GIS and Tourism Research - Home - SPINlab Vrije ... · PDF fileGIS and Tourist Consumer Research ... Geographical Information Systems ... Only the market research appears to be poorly

GIS and Tourist Consumer Research

The Spatial Profile of German Low-Cost Carrier Passengers

in Mallorca

Submitted in fulfilment of the requirements for the degree of Master of Science in GeographicalInformation Systems (UNIGIS)

Faculty of Earth and Life SciencesVrije Universiteit Amsterdam

The Netherland June 2009

Johannes Luberichs

GIS and Tourist Consumer Research -

The Spatial Profile of German Low-Cost Carrier Passengers in Mallorca

Submitted in fulfilment of the requirements for the degree of Master of Science in Geographical Information Systems (UNIGIS)

Faculty of Earth and Life Sciences Vrije Universiteit Amsterdam

The Netherlands Johannes Luberichs Rautenstraße 6 30171 Hannover Germany Phone: +49 163 1744992 e-mail: [email protected] Hannover, 21.07.2009

mailto:[email protected]

Disclaimer The results presented in this thesis are based on my own research at the Faculty of Earth and Life Sciences of the Vrije Universiteit Amsterdam. All assistance received from other individuals and organisations has been acknowledged and full reference is made to all published and unpublished sources. This thesis has not been submitted previously for a degree at any institution. Hannover (Germany), 21.07.2009 _________________________________ Johannes Luberichs

I

Abstract Space is an elementary part of tourism. This spatial relevance makes the field of GIS an ideal environment to analyse tourist phenomena. GIS is already applied in different parts of tourism business and research. The examples of applications range from site assessment over virtual tourism in 3-dimensional web space to the simulation of tourists’ spatial activity patterns in time and space. Only the market research appears to be poorly represented in the list of applications of GIS development and analysis in tourism science and research. Especially within a destination the knowledge about the distribution of different tourist consumer groups are valuable for destination management and tourist development planning. Therefore, this study develops a framework for spatially enhanced tourist consumer research. This framework will be deployed in order to analyse the spatial profile of German low-cost carrier passengers in Mallorca. The necessity is evident because Mallorca – being in a transformation from a low-budget and low-quality mass tourism destination to a destination of high class and quality tourism - currently faces a growing number of tourists in general and of visitors arriving by low-cost carriers in special. The study explores the distribution patterns of the German low-cost carrier passengers over the island. It discovers the existence of very distinct consumer groups in the different tourism spaces of the island and the differences to the German tourists in general. The strength of GIS in combining different spatial data has been exploited in order to reveal factors that influence the spatial distribution of the low-cost carrier passengers and the different consumer groups. As the spatial enhancement of tourist consumer research introduces some difficulties, this thesis names common pitfalls and discusses the problems arising from spatial data analysis.

II

Foreword Although I studied the geography of tourism, I did not encounter Geographic Information Systems (GIS) during my student life. In the 1990s GIS was not part of the curriculum in this discipline at our university. We studied the traditional cartographic techniques, but that was the basic knowledge in geography. When I entered the world of GIS by taking a detour after getting my university degree, I ended up in the environmental planning business. This is one of the areas where GIS had been applied first. The use of it is very widespread and the methods of GIS analysis seem to be well practiced. As a result of my studies, I am still interested in the development of tourism research. I had always searched to see if there were some opportunities for the applications of GIS in tourism. I saw the development of routing engines and web mapping on the internet. A well positioned destination or tour operator will show maps on their website, offer spatial hotel reservation systems, and in some cases Location Based Services (LBS) are available on-site. Increased access to the internet, high speed internet connections and more powerful mobile devices have boosted the development of GIS in this area over the last few years and this rapid development will continue. But even if these tools offer spatial query options (for example “show me all accommodations under 100 € per night within a radius of 3 km of a certain point of interest”) they are at the end of the day only pure information tools. I have spoken to people working in tourism about the application of GIS and geospatial analysis for research or management purposes. This has given me the impression that GIS is not wide spread in tourism science. I heard that the use of GIS as a tool in student education in tourism science grew over the last years. Mostly because making maps is easier with modern GIS software than it has been in earlier times. But on a first scan I did not see many essays or reports about GIS analysis in the tourism literature. Similarly I did not come across many tourism topics during my frequent scans through the GIS journals. So the combination of GIS and tourism spun around in my head when I was looking for a topic of my master thesis. I thought that there must be some use of GIS in tourism research since tourism is a spatial issue by definition. So I began to dig deeper in the matter…

III

IV

Table of Content 1. Introduction ......................................................................................................................1

1.1. Aims and Objectives .................................................................................................2 1.2. Data and Methodology..............................................................................................4 1.3. Structure of the Work ................................................................................................5

2. Geospatial Analysis in Tourism Science and Research...................................................6 3. Framework for Spatial Tourist Consumer Research ........................................................9

3.1. Tourist Consumer Research .....................................................................................9 3.2. Spatially Enhanced Research Process...................................................................10

3.2.1. Purpose of Research .......................................................................................13 3.2.2. Survey Design .................................................................................................13

3.2.2.1. Scale Consideration .................................................................................13 3.2.2.2. Data Sampling ..........................................................................................14

3.2.2.2.1. Sampling Technique ............................................................................14 3.2.2.2.2. Sample Size.........................................................................................15

3.2.3. Data Collection ................................................................................................15 3.2.4. Data Input and Management ...........................................................................16 3.2.5. Genuine Data Analysis Process ......................................................................17

3.2.5.1. Exploratory Spatial Data Analysis ............................................................17 3.2.5.2. Explanatory Analysis ................................................................................19

3.2.6. Data Synthesizing and Communication...........................................................20 3.3. Data Quality Issues.................................................................................................21

3.3.1. Non-Uniformity of Space and Edge Effects .....................................................21 3.3.2. Modifiable Areal Unit Problem and Ecological Fallacy ....................................22 3.3.3. Data Quality Documentation............................................................................23

4. The Research Location: Mallorca ..................................................................................25 4.1. Geography of Mallorca............................................................................................25 4.2. Tourism in Mallorca.................................................................................................27

4.2.1. The Spatial Distribution of Tourists in Mallorca ...............................................30 4.2.2. Low-Cost Carrier in the Air Traffic to Mallorca.................................................35

5. The Research Data ........................................................................................................37 5.1. Survey Design and Data Collection ........................................................................37

5.1.1. Sampling of the Mallorca Polls ........................................................................37 5.1.2. Geographical Data...........................................................................................38 5.1.3. Spatial Level of Analysis..................................................................................38

5.2. Data Input and Management ..................................................................................44 5.2.1. Geocoding of the Mallorca Polls ......................................................................44

5.2.2. Data Cleaning ..................................................................................................46 6. Spatial Tourist Consumer Analysis ................................................................................47

6.1. General Spatial Distribution ....................................................................................47 6.2. Spatial Distribution Compared to German Tourist in General.................................51 6.3. Distribution of Socio-Demographic Characteristics.................................................53 6.4. Trip-Related Behaviour ...........................................................................................58

6.4.1. Booking and Accommodation ..........................................................................58 6.4.2. Sojourn Time ...................................................................................................61 6.4.3. Travel Company ..............................................................................................63

6.5. Expectations and Preferences ................................................................................67 6.6. Spatial Consumer Profiles ......................................................................................73

6.6.1. Regional Consumer Profiles ............................................................................73 6.6.2. Municipal Consumer Profile.............................................................................75 6.6.3. Zonal Consumer Profile ...................................................................................76

6.7. Differences to German Tourists in General ............................................................78 6.7.1. S’Arenal / Playa de Palma ...............................................................................78 6.7.2. Costa de Ponent (South) .................................................................................79 6.7.3. Capdepera .......................................................................................................81 6.7.4. Summary .........................................................................................................82

7. Summary and Conclusions ............................................................................................86 7.1. Spatiality of German Low-Cost Carrier Passenger on Mallorca .............................86 7.2. Spatial Enhancement of Tourist Consumer Research............................................90

8. Outlook...........................................................................................................................97 References……………………………………………………...………………………………………i Attachments……………………………………………………..…………………………………..A-1

---------------------------------------------------------------

V

Table of Tables Table 1: Tourism density indexes at the coast of Mallorca for 2005 ......................................31 Table 2: Tourist arrivals in Mallorca 2005 differentiated by nationality...................................34 Table 3: Pearson’s R Correlation for selected variables on aggregated municipality level....50 Table 4: T-Tests for age means on regional level for the Costa de Ponent ...........................53 Table 5: T-Tests for age means for the Tramuntana zone.....................................................55 Table 6: Z-Test results for comparison of proportions of visitors in the training phase of life

(pupils, students, apprentices) for the Capdepera and Santanyi zone (selection) .........55 Table 7: Tests for spatial autocorrelation for the proportion of travel company on zonal level

.......................................................................................................................................64 Table 8: Activities the LCCP on a flight to Mallorca rated for their importance for this Mallorca

trip ..................................................................................................................................67 Table 9: What LCCP especially like of Mallorca.....................................................................68 Table 10: Z-Test results for the zone of Playa de Palma/ S’Arenal for the comparison of

proportion of LCCP naming culture as a special amenity of Mallorca they like..............71 Table 11: Significantly different characteristics of German tourists and German LCCP at the

Playa de Palma/S’Arenal................................................................................................79 Table 12: Significantly different characteristics of German tourists and German LCCP at the

Costa de Ponent (South) ................................................................................................81 Table 13: Significantly different characteristics of German tourists and German LCCP at

Capdepera......................................................................................................................82 Table of Figures Figure 1: The System of Tourism.............................................................................................1 Figure 2: Linear project design...............................................................................................11 Figure 3: Spatially enhanced data analysis framework ..........................................................12 Figure 4: Maps in the analytical process ................................................................................18 Figure 5: Geovisualisation in the cartography cube ...............................................................20 Figure 6: Development of tourist arrivals in Mallorca .............................................................28 Figure 7: Supply-side matrix of low-cost carrier offering service to Mallorca .........................36 Figure 8: Geocoding respondents ..........................................................................................44 Figure 9: Box plot of the age distribution at the central Costa de Llevant ..............................54 Figure 10: Activity rating of LCCP ..........................................................................................68 Figure 11: Age of tourists in Mallorca.....................................................................................83 Figure 12: Type of accommodation of Mallorca tourists.........................................................84

VI

Table of Maps Map 1: Geographic location of Mallorca and population in Mallorca......................................25 Map 2: Mallorca’s geographical segmentation and climate....................................................26 Map 3: Distribution of accommodations in Mallorca 2006......................................................30 Map 4: Municipality of destination for all tourists in Mallorca 2006 ........................................32 Map 5: Proportion of second homes per municipality 2002 ...................................................33 Map 6: Region of destination in Mallorca for tourists differentiated by nationality for 2007 ...35 Map 7: Spatial distribution of German LCCP in Mallorca on municipality level......................39 Map 8: Gi* application for customised zoning design.............................................................41 Map 9: Spatial levels of analysis ............................................................................................43 Map 10: Random location of respondents..............................................................................46 Map 11: Visited places in sample of the LCCP survey 2005..................................................47 Map 12: Density Equalising Cartograms for Mallorca’s LCCP tourism ..................................48 Map 13: Correlation of tourist beds and LCCP in Mallorca ....................................................49 Map 14: Distribution of German LCCP in relation to motorway connection ...........................51 Map 15: Differences in the regional distribution of German LCCP and German tourists .......52 Map 16: Socio-demographic characteristics of LCCP in Mallorca .........................................57 Map 17: Travel behaviour of LCCP in Mallorca: package tourism and accommodations ......60 Map 18: Travel behaviour of LCCP in Mallorca: sojourn time ................................................62 Map 19: Travel behaviour of LCCP in Mallorca: travel company ...........................................63 Map 20: Travel company in Mallorca’s tourist zones .............................................................66 Map 21: Regional activity rating of German LCCP in Mallorca ..............................................70 Map 22: What LCCP especially like on Mallorca....................................................................72 Map 23: Distribution of German LCCP and some selected spatial profiles of them in tourism

zones on Mallorca ..........................................................................................................87

VII

Abbreviations AEA Association of European Airlines AENA Aeroportes Espanoles y Navigacion Aera ANOVA One-way Analysis Of VAriance AR Augmented Reality AURISA Australian Urban & Regional Information Systems Association CAIB Comunitat Auònoma Illes Balears CASA Centre for Advanced Spatial Analysis CITTIB Centro de Promoción de la Investigación Y Las Tecnologías Turísticas

de les Illes Balears CSISS Centre for Spatially Integrated Social Science CSR Complete Spatial Randomness EDA Exploratory Data Analysis ELFAA European Low Fares Airline Association ESDA Exploratory Spatial Data Analysis ESRI Environmental Systems Research Institute GOP Grup Balear d’Ornitologia i Defensa de la Naturalesa GDP Gross Domestic Product HAZ Hannoversche Allgemeine Zeitung IBESTAT Instituto de Estatistica de las Illes Balears IDEE La Infraestructura de Datos Espaciales de España INE Instituto Nacional de Estadistica de España INESTUR Instituto de Estrategia Turística de las Illes Balears IQR Inter-Quartile Range IRPUD Institut für Raumplanung Universität Dortmund LBS Location Based Services LCC Low-Cost Carrier LCCP Low-Cost Carrier Passenger LIC Largest Inscribed Circle LISA Local Indicators of Spatial Autocorrelation MAT centre of Minimum Aggregate Travel MBC Minimum Bounding Circle MBR Minimum Bounding Rectangle NCGIA National Center for Geographic Information & Analysis PDA Personal Digital Assistant PIP Point-In-polygon analysis POI Point Of Interest

VIII

RS Remote Sensing SAL Spatial Analysis Laboratory at the University of Illinois SDI Spatial Data Infrastructure UN United Nations UNWTO United Nations’ World Tourism Organisation

IX

1. Introduction

1

1. Introduction Tourism and the related phenomena build a complex system influenced by all parts of the human environment and development (see figure 1; FREYER 2004; SCHLOEMER 1999; WALCH 1999). That is why tourism is the object of research in many scientific disciplines (FREYER 1998, p. 35; STORBECK 1992, p. 11). As manifold as the individual disciplines are the definitions of tourism1. The United Nations’ World Tourism Organisation (UNWTO) agreed on a very broad definition: “Tourism is defined as the activities of persons travelling to and staying in places outside their usual environment for not more than one consecutive year for leisure, business and other purposes not related to the exercise of an activity remunerated from within the place visited” (UNWTO 2000).

generatingarea destinationtransfer region

economy

ecology technological development

social and politicalenvironment

Composed according to Walch 1999, p. 25, Schloemer 1999, p. 3, and Freyer 2004, p. 5.

generatingarea destinationtransfer region

economy

ecology technological development

social and politicalenvironment

Composed according to Walch 1999, p. 25, Schloemer 1999, p. 3, and Freyer 2004, p. 5.

Figure 1: The System of Tourism

This definition, which includes business travel as well as day trips, indicates that moving in space (travelling) is the elementary part of tourism. That is why geography is heavily involved in tourism research. Tourism geography tries to explain the flows and patterns of tourism, investigates the impact of tourism and researches the spatial behaviour and activities of 1 There is an ongoing discussion on the definition of tourism (HOPFINGER 2004, p. 29) and its differentiation from leisure and recreation (HALL & PAGE 2006, p. 4/76). A widely accepted position is that the precise meaning depends on the research topic. Therefore, criteria like distance of travel, length of stay, or motivation lead to different aspects of tourism research (cp. FREYER 1998, p. 2; HALL & PAGE 2006, p. 4/76; HOPFINGER 2004, p. 29).

1. Introduction

tourists. The findings of tourism geography support decisions in regional development planning, destination management and tourism marketing. With the technical advances in computer sciences and techniques, geography got new powerful tools and methods to capture, handle, manipulate, and analyse huge amounts of data. These “organized collection of computer hardware, software, geographic data and personnel design to efficiently capture, store, update, manipulate, analyse, and display all forms of geographically referenced information” (ESRI ENVIRONMENTAL SYSTEMS RESEARCH INSTITUTE 1990, p. 1-2) is called Geographic Information Systems (GIS)2. With its capabilities for business mapping, geospatial analysis and its contribution to decision support GIS seems to be a very useful tool for the geography of tourism. To what extent the capabilities of geospatial analysis can contribute to tourist consumer research will be investigated exemplified for the consumer group of German low-cost carrier passengers (LCCP) in Mallorca3.

1.1. Aims and Objectives

In order to research the spatial pattern of tourism and the spatial behaviour of tourists, data have to be collected about the destinations and about the tourists. Data are nowadays collected in an enormous amount. Especially spatial sciences profit from that development. It is easier to get useful datasets through national or regional spatial data infrastructures (think of the European INSPIRE directive now in force) than it used to be some years ago. But for social sciences the data availability did not change in the same pace. The major part of the new data on individuals is held by companies. The data are neither freely available, nor have they been collected for scientific purposes and thus are often irretrievably flawed (O'SULLIVAN 2004, p. 98). Therefore, social sciences still rely on the same datasets as years ago. In most spatially enhanced social analyses the scientist works with census data that represent the population aggregated to certain spatial units like census blocks. In tourism there is no complete statistic available in a form similar to census data. Most countries have tourism statistics, but these statistics usually offer only a fraction of possible information. Often only the country of origin and the length of stay of a visitor are recorded. Package tour operators and other tourism businesses can work with data from their customer database. But destinations often do not have that kind of data because they do not interact with the visitors directly. Information on tourists can only be derived from field surveys. That is why one of the most used instruments in tourism destination management are tourist polls 2 Aware of the discussion around GIS as a science or tool (WRIGHT, GOODCHILD, & PROCTOR 1997) and the ambiguity of its acronym as GI System, GI Studies, or GI Science (LONGLEY et al. 2001) this paper deploys the cited technocratic view on GIS to do GI studies and thus contribute to the GI science. 3 Throughout this thesis the original Spanish spelling form of “Mallorca” will be used instead of the English version “Majorca” to avoid confusion because all other place names are in Spanish as well.

2

1. Introduction

at the destination (FREYER 1998, p. 239).Not much is written about the use of GIS as an analysis tool for the data derived from such polls (see chapter 2). The global motivation for this thesis was the general question how to explore those tourism statistics by geospatial analysis and which spatial information can be derived from these data. But why make all this fuss about space? Knowledge about the whereabouts of tourists assists tourism development and planning. With this knowledge, the location and allocation of tourism services (e.g. tourist information, transportation etc.) and facilities (e.g. accommodation, catering, additional attractions etc.) can be optimised. These services can then be adjusted to accommodate special interest groups that are clustered in a particular region. On the other hand, the agglomeration of certain tourist groups could be discouraged to minimise effects on the local citizens. To exemplify the utilisation of the spatial component in tourism research, a poll executed in Mallorca in 2005 serves as case study. Mallorca stands as a synonym for sun-and-sand mass tourism in the region around the Mediterranean Sea. The word “balearización” created in Spanish is derived from the tourist development on the Balearic Islands including Mallorca. It is a term describing negative side effects of mass tourism, e.g. coastlines of concrete, loud and drunken young people, or the loss of indigenous culture (KOCH & WACHOWIAK 2007, p. 1). Since the 1990s, the destination Mallorca has tried to diversify its product lineup to get rid of this old, stereotype image. In order to open new markets to attract different kinds of tourists, Mallorca has implemented different kinds of new products. These new products brought tourism to underdeveloped regions like the hinterland by means of agro- or cycling tourism. High quality and high yield promising market segments are promoted, e.g. golf, nautical, and first class tourism (SCHMITT 2000, p. 54). The pursuit of more high quality tourism seems to be contradicted by still growing numbers of visitors (SCHMITT 2007, p. 22) and a growing market position of low-cost carriers (LCC)4 in Mallorcan air traffic (STOSCH 2007, p. R/4; WACHOWIAK & NEUMANN 2008, p. 3). Nevertheless, the tourism in Mallorca is changing. Mallorca’s tourism statistics prove that fact (decreasing number of package tourists, (slowly) increasing high class accommodation bookings, less distinctive seasonality). The indicators for that change summarise the complete tourism to some figures. These figures neither differentiate very much between different consumer types nor do they take the spatial component much into account. Therefore, this thesis will account for both, the different characteristics of tourists and their spatial distribution. The focus will be on the new but huge group of low-cost carrier passengers (LCCP). The LCCs revolutionised the flight business and with that the mass 4 Low-cost carrier is a common, but not consistent term in the literature to describe of airlines operating on a cost focussed business model and are typically characterised by low fares (SCHREIBER 2006, p. 139ff). Further information is available in chapter 4.2.2.

3

1. Introduction

tourism market. The travel behaviour of LCCP will heavily influence the tourism development in Mallorca and other mass tourism destinations. This will in turn affect the tourism policy, the destination management, and strategic planning of the tourism industry. For those planning and management reasons, the factors influencing distribution of the tourists around the destination are of great interest. That is why this thesis is going to answer the questions:

• Which tourism spaces do the German LCCP consume? • Is the spatial distribution of the German LCCP different from other tourists? • What might be the factors influencing the spatial distribution of the German LCCP?

Working with georeferenced data in tourist social science is an unattended field of research. Due to this disregard a framework for spatially integrated tourism consumer research must be developed. In order to do so, some additional detailed questions regarding the feasibility of geospatial analysis in tourism consumer research will be answered:

• What is the appropriate spatial scale or level of analysis? • How to set the boundaries of tourism spaces? • Which methods are applicable for analysing spatial distribution of visitors? • Which information can be derived from geospatial analysis regarding the visitors’

characteristics? • What are the data requirements? • What risks are involved with geospatial analysis and how do they affect the

interpretation of the results?

1.2. Data and Methodology

The University of Applied Science of Bad Honnef / Bonn put an exploratory focus on tourism in the Balearic Islands, a group of Spanish islands in the Mediterranean Sea. In 2005 the researchers of this university started in cooperation with the University of Trier to investigate the current consumer profile of Germans travelling to Mallorca, the biggest of the Balearic Islands. Coordinated and supervised by Prof. Dr. Helmut Wachowiak from the University of Bad Honnef / Bonn different polls with different foci have been made. The 2 major ones not focussing on special interest tourism are chosen for this work. One captures the profile of the German LCCP travelling to Mallorca (WACHOWIAK 2005b). The data from this will mostly be exploited for this work in order to outline the spatial component in the LCCP consumers’ profiles. The other poll surveys the behaviour of all German tourists (WACHOWIAK 2005a). It will be used as standard of comparison where appropriate. Additional to the raw statistical data from the Mallorca surveys, data on the tourism in Mallorca and geographic information were supplied by different Balearic and Spanish agencies or have been derived from former publications on Mallorca.

4

1. Introduction

The results from literature reviews and data collection then will be fed into an analytical process in order to answer the questions raised in chapter 1.1. This analytical process will be developed to accommodate the special issues introduced into tourist consumer research by the spatial component.

1.3. Structure of the Work

After defining the aims and objectives for this work in the foregoing chapter the next will review current literature on the matter of GIS and geospatial analysis in tourism research in order to set this thesis in context to the existing research. In chapter 3 a framework for spatial consumer research will be developed accounting for the special issues and problems introduced by spatial information. From chapter 4 onwards the case study about the spatial component in the consumer profile of German low-cost carrier passengers will start. Chapter 4 will give an introduction of the research area in terms of geography, tourism development and distribution. In addition, the term low-cost carrier will be defined and its role in Mallorca’s air traffic will be researched. Chapter 5 shows the stages of data collection and preparation of the surveys this thesis is based on. In chapter 6 the genuine spatial tourist consumer analysis takes place. Different characteristics are highlighted and occurring phenomena are described along with possible explanations for them. The results are then summarised to spatial profiles for the different areas and for the different spatial levels of analysis. That way they can be compared to the profiles of German tourists in general, what happens in the chapter 6.7. In the conclusion (chapter 7) the findings of this thesis will be summarised and critically reviewed, in order to answer the questions about German LCCP raised in chapter 1.1. Finally chapter 8 gives an outlook where the development of LCC in Mallorca leads to and what GIS might contribute to consumer research in the future.

5

2. Geospatial Analysis in Tourism Science and Research

6

2. Geospatial Analysis in Tourism Science and Research As indicated in the introduction, GIS seems to be predestined for the use in tourism development and management. There are already some reviews on the use of GIS in that field. At the end of the 1990s MCADAM (1999) and BAHAIRE & ELLIOT-WHITE (1999) outlined where GIS could be applied in tourism and tried to find example applications. They claimed to have had some difficulties finding essays on the topic. They named scenery and visual impact analysis, land cover change modelling (BAHAIRE & ELLIOT-WHITE 1999, p. 166/168; MCADAM 1999, p. 79/80), detection of conflicts between ecological protective and recreational landuse, and site selection for ecotourism in Canada (BAHAIRE & ELLIOT-WHITE 1999, p. 163ff). They both state that GIS is a formidable tool for tourism development and sustainable tourism planning but its utilisation was rare and awareness amongst stakeholders in tourism has been small. That might have been due to the high costs for software, missing data and skills or even complete unawareness (MCADAM 1999, p. 87). GIS matured more and more in the new millennium and is now widespread. So the idea itself suggests that the situation has changed and GIS is common in tourism. The 2004 review from FARSARI & PRATACOS did not support that assumption. They based their essay on GIS in tourism planning and management on the articles of BAHAIRE & ELLIOT-WHITE and MCADAM. They only extended the list of applications a bit and added marketing as a new aspect. In that area GIS assists tourism management by promoting the destinations via web-mapping information tools and with geomarketing (FARSARI & PRATACOS 2004, p. 598/601). But since nearly all cited essays with the exception of 3 are from the 1990s it is not surprising that they conclude that “the number of GIS applications for tourism planning has not mushroomed as the technology has in other fields” (FARSARI & PRATACOS 2004, p. 596). This is supplemented by the notation that this lack of GIS applications is mostly apparent for existing mass tourism destinations. The latest reviews do not come to other conclusions. In the chapter on the role of GIS in the geography of tourism HALL & PAGE rely heavily on the article of BAHAIRE & ELLIOT-WHITE or even some older sources (HALL & PAGE 2006, p. 349ff). LEVY & DICKSON mostly name possible applications of GIS and Remote Sensing (RS) technologies for destination management without exemplifying them with existing applications (LEVY & DICKSON 2006). And when CHEN complained recently that “few researchers have applied GIS to tourism planning and management practices” (2007, p. 290) she referred to the article of FARSARI & PRATACOS (2004). So what about the current state of GIS in tourism? FARSARI & PRATACOS said that GIS is mostly used in assessing suitable sites for tourism development. This still seems to be a field where GIS is applied, as the examples from


Australia, Turkey and India show (ARROWSMITH & NTUWAH 2002; ERKIN & USUL 2005; KATIYAR & JAIN 2007; KUMAR & DATTA 2007; ÖCAL & USUL 2006). Another field already mentioned by FARSARI & PRATACOS is the use of maps on the internet to promote destinations. The research in that field is much devoted to the development of spatial web-applications like spatially enhanced hotel reservation engines (JOUN & RYU 2004), tourist route planers (cp. e.g. CUBEROS et al. 2000; or DICKMANN 2005, p. 44), or even 3-D web presentations of cities, already referred to as “virtual tourism” (HUDSON-SMITH & EVANS 2003, p. 51). The information on the destinations website aims to help the visitor make decisions where to go and what to do when he/she is there. To accommodate the tourist within the destination, new guidance by electronic devices is developed. These are location based services (LBS) and offer information at points of interest (POI) and help navigating. Web-mapping and LBS are of much interest in current literature (DICKMANN 2005; DYE & SHAW 2007; O'LOONEY 2004; SCHILLING, COORS, & LAAKSO 2005; ZIPF 2002) and reach up to the research on the feasibility of augmented reality (AR) for tourist guidance (COORS, HUCH, & KRETSCHMER 2000). All the examples listed so far have one thing in common. They do not focus on the research about the tourist. However the demand of ELLIOT-WHITE & FINN (1998) for a reconsidered geomarketing in tourism did not find an echo in the literature of the following years. Only the work of FICHTNER (2004, p. 510ff) about the catchment area of a German theme park calculated by the gravity theory integrated into GIS analysis was found from the field of geomarketing in tourism. For destination management and business LARBIG, KÄMPF & KELLER (2004) developed a GIS-based benchmarking tool for destinations. One prominent variable for distinguishing different tourism regions in Switzerland is the source market in terms of country of origin (LARBIG, KÄMPF, & KELLER 2004, p. 82/84). But the tourists have not been distinguished further in that application, although sometimes that could be very useful. For example, CHEN (2007, p. 291ff) introduced a case of site assessment for a tourism-oriented retail store for clothing. The target group is clearly circumscribed as women between 18 and 45 from nearly all income levels. In the final GIS-based site selection analysis she used only census data for the resident people. Especially for tourism-oriented retail in villages with small populations and many tourists, the information about the tourists in the nearby surroundings would be important because the purchasing power of the local people do not suffice to keep such a shop financially viable. These data are unfortunately often not available or have to be collected with huge efforts. So it might not be surprising that no literature has been found where the stationary distribution of tourists and different tourist groups within a destination has been investigated. Compared to the lack of literature on tourism distribution, much research is devoted to the spatial activity pattern of tourists within a particular place. The earliest work has been

7


conducted by DIETVORST (1995) who used GIS software to model visitor flows in a Dutch theme park. The results aimed at helping to resolve congestion problems at several attractions on busy days (DIETVORST 1995, p. 175). In that kind of research the time-factor is just as important as space. Detailed information on times and places, that are collected e.g. by diary (e.g. DIETVORST 1995; PIYATHAMRONGCHAI & TRIPATHI 2000), by GPS devices (see ISAACSON & SHOVAL 2005), or even by mobile positioning from cell phone roaming data (see AHAS et al. 2007), are fed into simulation software. The simulation modules are then coupled to or integrated in GIS software. Thus maps are produced that show when and where peaks of visitors have a negative impact on the natural environment (PIYATHAMRONGCHAI & TRIPATHI 2000) or even on tourism itself (GIMBLETT, RICHARDS, & ITAMI 2001). This review is not exhaustive but mirrors the development of GIS use in tourism. During the 1990s GIS was mostly implemented in site selection or impact assessment processes with focus on environmental and ecological issues. The reason for this might be the migration of GIS from the related field of environmental research where it already had been well established by that time (BAHAIRE & ELLIOT-WHITE 1999, p. 163; FARSARI & PRATACOS 2004, p. 596). Today, dynamic web-maps and simple web-GIS with spatial query functions, e.g. for hotel search or routing are well established in tourism business. It is even such a classic application of GIS that WORBOYS & DUCKHAM (2004) name a tourist information system as an example application for GIS in there introductory work to GIS. Because geographic tourist information systems have matured the current research focuses on the development of new products like 3-D-Web-GIS, LBS, or even AR for tourist information and guiding. In contrast to the research on spatial activities and time-space patterns of tourists, there is no evidence in the tourism business that GIS-supported market research is widely in use.

8

3. Framework for Spatial Tourist Consumer Research

9

3. Framework for Spatial Tourist Consumer Research What is known as market research in tourism comprises 2 main aspects: the competitor analysis and the consumer analysis. The first deals with market shares and marketing strategies of business rivals. The consumer analysis explores the demand (FREYER 1998, p. 235ff). Because this study is focussed on tourists’ distribution in Mallorca it will deploy the measures of consumer analysis. It is a well developed field of tourism research. But as outlined in chapter 2 not much is published about the spatial exploration of tourist consumer research data. That is why a framework for geospatial consumer research is missing. In order to reveal the spatial consumer profile of German LCCP in Mallorca and answer the questions raised in chapter 1.1 a framework for spatial tourist consumer research will be developed.

3.1. Tourist Consumer Research

Consumer analysis is the main focus point in tourism market research because in an increasing individualised market managerial decisions need to be based on detailed information about target groups and market segments. The necessity to divide the mass of tourists into groups is evident for the supplier of tourism products and services because they have to orient on the needs and wants of the tourists. The segmentation of different tourist groups enables the tourism companies, and destinations amongst them, to developed customer oriented products (STEINECKE 1988, p. 326; STEINGRUBE 2004, p. 138). Therefore, consumer research “yields information about the motives and needs of different classes of consumers” (THE FREE DICTIONARY 2008). During the last decades, different approaches to differentiate the tourist demand were developed. The main approaches to market segmentation in the international tourism research are the following (cp. FREYER 1998, p. 71; STEINECKE 1992, p. 180):

• Definition of target groups by socio-demographic parameters (e.g. age, size of household, income, graduation).

• Definition of target groups by tourist behaviour (e.g. booking behaviour, travel distance, preferred destination landscape).

• Definition of target groups by psycho-graphic parameters (tourist typologies by motives and expectancies; life-style groups).

The socio-economic approach is very simple and easy to conduct even for small enterprises. On the other side it does not get far enough in the world of more and more complex consumers. Today people are more used to travelling and thus more experienced in it, more people obtain a high school degree, and older people are more flexible than some decades ago. Due to these changes in social conditions the socio-demographic information alone


does not suffice to exactly distinguish groups of consumers. Therefore, consumers are often distinguished by their tourism related behaviour. But as this behaviour is only the result of some specific decisions, methods have been developed to find out more about the motives and the environment that influence the attitudes and behaviour of the consumer. This approach, working with psycho-graphic variables, is very complex. That is why a combination of the above approaches is most often implemented in practice (discussed in FREYER 1998, p. 71ff; STEINECKE 2006, p. 57ff). A special aspect of tourism is that the product is a place, a region. It is product and place of consumption at the same time (FREYER 1998). This place is commonly managed by a department that is responsible for the coordination of the different stakeholders (hospitality, gastronomy, car rental, retail etc.), image campaigns, and pull marketing to attract tour operators and tourists. Therefore, the destination in tourism is also a business unit (BIEGER 1997, cited by BECKER 2004, p. 464). A destination can comprise regions from as small as a lake to whole continents (ASHWORTH & GOODALL 1990, p. 6). Since it is the place where to meet the tourists, most tourism data for research are derived from polls at the destination (FREYER 1998, p. 239). That is one reason why the destination is the area of interest in this study. Producing customer oriented products is a target of consumer research. For destinations, the development of these products is highly spatially related. The different facets of this product have to be spatially arranged to suit the visitors’ wants. That is why the information of the spatial distribution of different groups of tourists is crucial for the allocation of tourism facilities and services.

3.2. Spatially Enhanced Research Process

Consumer analysis is applied empiric social research. As tourism with all its aspects is part of many research disciplines (sociology, psychology, geography, anthropology etc. (see chapter 1)), all these sciences contribute to the research approaches and methods in empiric research in tourism. Therefore, the process of consumer analysis in tourism is not very different from other disciplines. SEITZ & MEYER (2006, p. 130) roughly distinguish the 4 phases survey design, data collection, data analysis, and publishing. Of course this is framed by an overall project aim that raises a question and triggers the research. This seems to be a straightforward, linear process. In this kind of linear project design testable hypotheses are derived from the overall questions. These hypotheses are tested on a well-defined sample and conclusions are drawn on the questions raised in the beginning of the project. This is a still manifold applied structure of the process of (statistical) research (see HUIZINGH 2007, p. 2/3; LONGLEY). As outlined by MAASDAM (2000, p. 27) this linear process design had a

10


clearly stated objective and with that well defined hypotheses to be tested with little to no opportunities to recast the them (see figure 2).

Figure 2: Linear project design

Following this approach, the methods and tests executed on the data would be predestined. The weakness of the linear project outline is to miss interesting, unexpected results when following only one, predefined way of analysis. This way of analysis ignores the necessities of consumer research. Of course even in tourist consumer analysis the research starts with a general purpose. But the starting question is mostly very diffuse as it asks for a profile of customers in general or of a certain group of customers. Specific characteristics or behavioural patterns are difficult to express in hypotheses beforehand. In order to detect those specialities or unexpected phenomena a wider, more open-minded, explorative approach is necessary. Because explorative data analysis (EDA) is said to be “the process of using statistical tools […] to investigate data sets in order to understand their important

11


characteristics” (TRIOLA 2008, p. 119), it fits the demands of tourist consumer research. But it is more than just this technical definition. It is a philosophy of how to do data analysis (MAASDAM 2000, p. 29; TUKEY 1977, p. 21). The spatial information is important to the destination management (see above). Introducing the spatial component into consumer research increases the amount of data to be handled. But in the same way it enhances the possibilities of insight. Spatial data are special and address different requirements to data management and data analysis. But the exploratory philosophy behind the research process stays the same. Therefore, Exploratory Spatial Data Analysis (ESDA) is a spatially enhanced EDA. By working with geographic information in tourist consumer research the consumer data could be combined with any possible information available for the research areas. These additional data are geographic information (e.g. climate data, infrastructure networks, or administrative boundaries) or secondary statistics for different spatial levels (like economy or population statistics). These data can be used for exploration or the explanation of phenomena discovered already in the genuine analysis process. If the financial budget allows, even new data can be obtained when requested by results of the data exploration. As a rule, this does not include the conduction of a whole new survey because it is too expensive and time consuming.

Purpose of research

aim, scope, field

Survey design

Interview forms, scale consideration, statistical and

spatial sampling, etc.

Data input and management

Data pre-processing, geocoding, transformation,

reduction

Data synthesising and communication

Reports, graphics, maps and other geovisualisation techniques

Genuine analysis process ESDA

interpretationVerification

Data collection

Field survey (poll)Secondary

statistics and geographic data

New knowledge

Purpose of research

aim, scope, field

Survey design

Interview forms, scale consideration, statistical and

spatial sampling, etc.

Data input and management

Data pre-processing, geocoding, transformation,

reduction

Data synthesising and communication

Reports, graphics, maps and other geovisualisation techniques

Genuine analysis process ESDA

interpretationVerification

Data collection



Data collection



New knowledge

Figure 3: Spatially enhanced data analysis framework

12


The genuine analysis process is the part of the research framework where the data are explored by all available measures of EDA and ESDA. As these techniques make it a “detective work finding and revealing clues” (TUKEY 1977, p. 21), new knowledge can even be gained from the exploratory part of the analysis process. But often it will raise new questions and vague ideas that require confirmatory analysis. That is why the explorative paradigm to research is an extension but not a denial to the explanatory analysis with its hypothesis testing. But without ESDA a lot of hypothesis would not even be discovered. The single parts of the spatially enhanced research process will be outlined in the coming paragraphs along with the mentions of specific issues that arise from analysing spatial data.

3.2.1. Purpose of Research

The purpose of research is in tourist consumer research a vaguely defined objective the research aims to. In destination based surveys the information about the needs and wants of the tourists is mostly wanted. Sometimes only a special interest group is in the focus of a survey like the LCCP in this thesis. The market segmentation that is per definition part of consumer research cannot tell beforehand which might be the outstanding characteristics that lead to the classification of the tourist into different consumer markets. This information is to be derived from a process of surveying the tourists and then explore the data resulting from this survey.

3.2.2. Survey Design

The survey design includes the creation of an interview form and the pre-testing of it. The information asked for in this interview depends on the population defined by the purpose of research and the special interest of the research. In order to make the data fit for geospatial analysis spatial information must be collected along with the interviews. Other special spatial issue to be considered are the scale of analysis and the spatial data sampling.

3.2.2.1. Scale Consideration

Working with spatial data means to always consider the scale because the scale affects the observation, which will be made. The scale influences the representation of objects (point or polygon), the way data must be collected, and of course methods of analysis and the scale of the analysis results (O'SULLIVAN & UNWIN 2003, p. 33). The biggest possible scale of analysis is determined by the dataset with the smallest scale. That is why the scale has to be considered already in the planning phase. A survey in regional level will deliver appropriate results for regional analysis. Upscaling of this information to a subregional level is not possible afterwards because particular information is missing (cp. HAINING 2003, p. 94ff). Downscaling by aggregation of spatial information to bigger units is always possible

13


afterwards. Therefore, the spatial precision should be as fine as possible within the timely and monetary budget.

3.2.2.2. Data Sampling

The collection of information on the whole population usually is too time-consuming and cost intense or often even utterly impossible. That is why real world information has to be abstracted by a sample that is drawn from the population. In order to get reliable and valid results the sample has to be representative. That will be reached by a sufficient sample size and a sample design that ensures random sampling.

3.2.2.2.1. Sampling Technique

Random sampling means that every element of the population must have the same probability to be selected by the sample (BAHRENBERG, GIESE, & NIPPER 1990, p. 18). In field surveys on the street, interviewees can be randomly selected by throwing a dice, or an approach for systematic random selection can be applied by asking every e.g. third person (SEITZ & MEYER 2006, p. 84). But that system only works for single places. In destinations with many different places the sample has to account for the differences between these places. For example, a destination with 2 villages is the object of research. One village receives 30% of the visitors, the other 70%. Asking the same number of people in both villages would distort the results. The data could not be employed for conclusions on the whole destination without corrections. To account for that problem either a spatially systematic sampling is applied or the single cases of the questionnaire have to be weighted afterwards. Spatially systematic sampling requires sample locations to be spread systematically over the research area to mirror the real distribution of objects in the sample. To avoid distortions in the sample the number of interviews per location should be defined beforehand (cp. BAHRENBERG, GIESE, & NIPPER 1990, p. 19; HAGGETT 2001, p. 713/796; LONGLEY et al. 2001, p. 104). In addition, corrections to the sample are necessary afterwards because in destination-based polls people with a longer sojourn have a higher probability to be interviewed than those with a short sojourn (BUSCH & LEITNER 1992, p. 168). Thus spatially systematic sampling requires base data and in most cases includes some assumptions during design and correction processes. Therefore, it is time-consuming and cost-intensive as well as prone to the introduction of errors if not handled very carefully. A workaround for that problem is to choose a location for the interviews where all visitors of a destination have to pass by. With a simple random selection at this point, a reliable sample can easily be achieved. If location is one of the variables in the interview, the spatial distribution of tourists can be measured by data surveyed in this way. Unfortunately, this way is not feasible for most destinations as they are accessible from many directions by different

14


modes of transport. Only for special kinds of destinations is this way of sampling an option. National parks with restricted entrances or islands like Mallorca with only a few access points like airports or ports are amongst them.

3.2.2.2.2. Sample Size

The scale considerations also influence the sample size. There is no precise rule for sample size in statistics but the bigger the better. The more the values of a variable tend to scatter the bigger the sample should be (BAHRENBERG, GIESE, & NIPPER 1990, p. 18). This rule especially applies if a project aims at drawing a distinct conclusion on different groups within the sample. A survey with the objective to find out something about the spatial distribution of a certain kind of visitors within a destination has to gather a high number of cases to cover even less frequented places (SEITZ & MEYER 2006, p. 95). A critical number in classic and in spatial statistic is 30. Below 30 cases some statistical figures like the z-score calculation change their formula (MYATT 2007, p. 68) and some measure in spatial analysis like LISA or the Gi* are not reliable (cp. MITCHELL 2005, p. 174/180). To have less than 30 respondents per spatial unit would even make it difficult to further subdivide the groups and attaining statistical confidence at the same time. For spatial analysis, that means at least 900 interviews would have to be made when the interviews are evenly distributed over the minimum number of spatial units. Since this is only feasible under certain conditions, a survey can work with a smaller number of interviews or a smaller number of areal units when the research question can be answered without certain methods. Therefore, the sample size in spatial tourism consumer research depends on the size of the population, the desired spatial resolution, and the research question. In reality, time and money also play a decisive role.

3.2.3. Data Collection

The data collection for spatially enhanced tourism consumer analysis is twofold. Data about the tourists’ characteristics have to be collected. In addition, geographic information and secondary statistics are necessary to georeference the attribute information of the tourists to places on earth and to set the tourists in a spatial context of other information. The destination based consumer survey is a very cost-intensive part of a project. That is why this part of data collection usually can only be done once. Therefore, the questions asked to the interviewees have to be well considered and pre-tested (cp. SEITZ & MEYER 2006, p. 45). The polls can either be conducted as face-to-face interviews or as written self-applied questionnaires. For the face-to-face interviews tourists are asked on the streets, in hotels, or in other places directly by the interviewer. For the self-applied questionnaire the interview forms are usually handed out to the visitors in specific places or placed in the hotel rooms. All

15


forms of destination based questionnaires have to solve problems to ensure representativeness (see chapter 3.2.2.2 above). For spatial analysis the location of the tourist is of high interest. It can be recorded in different ways. The tourist can show or sketch locations on a map. The precision of the location will then be determined by the scale of the map used during the interviews. This scale will be the highest possible resolution for geospatial analysis. For self-applied questionnaires add-ons like maps or pictures are often impracticable. In that case, the tourist can be asked for the place of accommodation (or what location ever is of interest to the research question). The interviewee names or writes down the locations. The precision of the spatial information then depends on the interviewee’s knowledge of the region and thus can differ significantly. To locate the named places afterwards on a map can be tedious work, especially if an appropriate gazetteer is missing or place names are ambiguous. Hotel addresses would be more precise information. But for confidential reasons most people might refuse to answer such questions. The invention of new techniques related to GIS offer new possibilities in locating tourists and monitoring their spatial activities. They can be tracked by GPS very precisely. This is mainly applied to studies of spatial activity patterns of tourists (see chapter 2). Analysing the distribution of tourists by mobile position via mobile phone roaming data is a very recent approach. The weaknesses are that it only offers information on the nationality and excludes wide ranges of tourist groups (AHAS et al. 2007).

3.2.4. Data Input and Management

Data input is the crucial part of the analysis process. How the data are entered will later determine how appropriate they are for certain analysis methods. This regards to the attribute values and their coding as well as to the implementation of the geographic information. The geographic representation of the cases from the field survey depends on the way of collecting the variable space. If coordinates or addresses for each respondent are available, they could be represented each as a point in GIS environment. If only a village or region is known as location, data from cases within this area will be summarised and joined as attributes to the particular geographic feature. Depending on the scale of analysis it can either be represented by a point or a polygon. The data are fed into a system consisting of hardware and software that is appropriate for the research focus. GIS usually lacks the capabilities for sophisticated classical statistical analysis like cross tabulation, correlation or factor analysis, and significance tests. Statistic packages as a rule do not offer many possibilities for mapping and spatial analysis. But for spatially enhanced tourist consumer research data must be accessible to GIS software and statistical software. Nowadays, both categories of software offer a wide range of data access

16


and exchange capabilities or provide interfaces to common databases. To manage the data in this system means to “create, update, check, validate, control, recode, extract, aggregate, or question” data sets (JAMBU 1991, p. 14).

3.2.5. Genuine Data Analysis Process

The data analysis is the core of tourism consumer research. The engine of the whole project is running in the phase of data analysis in terms of applying statistical and geospatial analysis methods. In spatial tourism consumer analysis based on questionnaires process steps and methods of statistical analysis and geospatial analysis have to be combined. The order of method application and tool selection depends on the data and the purpose of research. But some basic principles are not deniable: Start with a simple overview and go on to the more complex methods. The start will be the exploration of single variables followed by the search for relations and patterns between variables or spatial objects. If the exploratory part uncovers indicators for special and spatial patterns, further steps will assist in defining and localising clusters (see for example the structure of MITCHELL'S guides to GIS analysis MITCHELL 1999; MITCHELL 2005). This analysis is not a straight forward process following a scheme step by step to the results. It is more a kind of analytical circle where the process is considered to be iterative. At each step it is possible and sometimes even necessary to look back to the previous one (DE SMITH, GOODCHILD, & LONGLEY 2007, p. 55ff). When from the results of ESDA questions are raised and hypotheses are generated the test of them is the confirmatory part.

3.2.5.1. Exploratory Spatial Data Analysis

The basis for ESDA is the data provided through the data collection process. The search for information is guided by the purpose of research. ESDA techniques look in the data for interesting characteristics and patterns. Indications of patterns then form the basis for generating ideas of relationships, framing precise questions and producing hypotheses. The statistical methods in tourism consumer analysis does not differ from those in other disciplines dealing with similar topics (SEITZ & MEYER 2006). Therefore, all techniques and methods from simple, one dimensional summary statistics to sophisticated multivariate statistical analysis are available. That also includes the visual techniques from EDA like histograms, box plots, or scatter plots. Very common in tourist consumer research are factor and cluster analysis to extract consumer profiles and target groups from a dataset (exemplified e.g. by HALLERBACH 2004; and KLEMM, LUND-DURLACHER, & WOLF 2004).

17


In statistics data are commonly organised in tables. Tables are an effective tool for organising and analysing data. But as long as they are not linked to geographic objects they ignore spatial components like connectivity or contiguity. For example, a table with information on communities in Mallorca can be ordered alphabetically. It places Muro next to Palma de Mallorca. But they are located at opposite sides of the island. Illustrating the information from the table on a map allows comparing communities to their neighbours. The map is the special way of illustrating spatial data. It makes the difference between spatial and non-spatial data. But it is more than just a means of data communication. It is also a tool for spatial analysis. The map is “a way of empowering the human eye and brain to perform intuitive spatial analysis” (GOODCHILD & JANELLE 2004b, p. 6). And the “primary tool for analysis is the human imaginative mind, and all other tools are supplementary” (ANDRIENKO & ANDRIENKO 2006, p. 163). Most scientists tend to agree that the human mind can best cope with images (ibid.). Therefore, maps are considered as powerful analysis tools that are part of all stages in a geospatial analysis process (as simplified illustrated in figure 4 (O'SULLIVAN & UNWIN 2003, p. 17)). Combined with spatial queries, point-in-polygon (PIP) search, overlay and buffer analyses, as well as dynamic linking between different techniques of visualisation, mapping belongs to what in spatial data analysis is known as geovisualisation. It integrates techniques from scientific visualisation, information visualisation, and cartography that are processed by a computer to explore, analyse, and present spatial data (after MACEACHREN, cited by GABLER-MIECK & DUTTMANN 2007, p. 2). It is especially the dynamic interaction GIS offers as advantage to geovisualisation. With that the maps on the computer are much more flexible than their analogue counterparts made of paper. The appearance of maps can be altered rapidly, combined and linked to graphics. In that way spatial patterns of different variables within a dataset can be visually detected by the researcher.

Figure 4: Maps in the analytical process

18


But the sheer sight of patterns does not mean that they are not created by chance. Because the human mind is easily tricked and mislead or patterns are too complex for visual detection (MONMONIER, cited by OPENSHAW.S. 1991, p. 393) (MONMONIER 1996), different methods have been developed to identify patterns in a spatial dataset by spatial statistics. The most frequently used methods for pattern detection in social sciences are measures of spatial autocorrelation (SWEENEY & FESER call them “the workhorses of social science research on spatial processes” (2004, p. 244)). The standard measures for global autocorrelation, like Moran’s I or Geary’s C, measure the existence of global patterns in terms of clustered, dispersed or randomly distributed with one single index. In addition to these methods, local measures for spatial autocorrelation localise the clusters within the research area. These local indicators of spatial autocorrelation (LISA) developed by ANSELIN are decompositions of global measures like Moran’s I. They are additions to the toolbox of ESDA and can serve 2 purposes: On the one hand, they may be interpreted as indicators of local pockets of nonstationarity, or hot spots. On the other hand, they may be used to assess the influence of individual locations on the magnitude of global statistics and to identify outliers (ANSELIN 1995, p. 93/112). ORD & GETIS already developed in 1992 a local measure for autocorrelation. Their measures Gi and Gi* are especially useful when global statistics do not identify significant pockets of clustering (ORD & GETIS 1995, p. 287). Those local indicators of spatial autocorrelation are already implemented in some GIS packages.

3.2.5.2. Explanatory Analysis

Once the ESDA showed interesting patterns, relationships, or anomalies, the questions for the reasons crop up. Hypotheses will be generated and can then be tested. That field is well developed for tabular data and thus can be used for the attribute data of the special objects. A way of spatial confirmatory analysis is the development of dynamic models (e.g. of tourism flows or spatial activities as for example done by ARROWSSMITH & ITAMI (2004)or GIMBLETT, RICHARDS & ITAMI (2001)). These models are tested for their accuracy of modelling the real world processes. These modelling and validation processes require enormous amounts of data. Next to the information that is incorporated in the model an independent sample is necessary to validate the model. Modelling works best for regions with frequent and detailed polls. Most scientists agree that spatial analysis is more suitable for hypothesis generation than testing and that description is the main focus of spatial analysis (OPENSHAW.S. 1991, p. 392). But ANDRIENKO & ANDRIENKO (2006, p. 164 and p. 579ff) name 2 very important tools for interpretation of statistics and explanations of relationships: the human mind and domain knowledge. Both come into operation on the interface between exploratory and explanatory

19


analysis and can help to draw conclusions from data even when mathematical formulas do not work.

3.2.6. Data Synthesizing and Communication

Statistical and geospatial analysis is nowadays a very complex, highly visual and interactive matter. As the cartographic cube shows (figure 5), there are graphics for the analyst as means of investigation and there are graphics for the end user. In order to present results that decisions can be made on, the data have to be synthesised. That means to gather the most telling and most significant results and present them in a user-friendly way (cp. JAMBU 1991, p. 4/17). The same applies to the research of this work. Since the geovisualisation techniques applied during this study can only be presented partly and in examples on paper, the results will be presented mostly in forms of static cartography but with some more information than they would contain for business presentations. The map is a very important means of communication for spatially enhanced tourist consumer research. Presenting maps has its difficulties and pitfalls. They can even be used to make the results meet the expectations and thus manipulate decisions. Therefore, the creation of maps is a delicate matter and has its own science. How to create decent maps and where awareness of misleading display is necessary is discussed for example by BREWER (2005) or MONMONIER (1996).

Low HighPublic

Private

Present

Discover

Interactivity

Audien

ce

Goals

Cartography

Geovisualisation

Worboys & Duckham 2004, p. 306

Low HighPublic

Private

Present

Discover

Interactivity

Audien

ce

Goals

Cartography

Geovisualisation

Low HighPublic

Private

Present

Discover

Interactivity

Audien

ce

Goals

Cartography

Geovisualisation

Worboys & Duckham 2004, p. 306 Figure 5: Geovisualisation in the cartography cube

20


3.3. Data Quality Issues

In spatial consumer analysis based on tourist interviews, data are available about individual tourists. This information can be spatially aggregated to areas that fit particular purposes. As long as the data are a sample, every step of aggregation introduces uncertainty that has to be accounted for. For knowing about the quality of an analysis result, the user needs something to know about errors, probability of confidence, and uncertainty in the data. Some of these issues have to be calculated for each process step, e.g. the significance level or measure of central tendency. Others are general effects that are necessary to consider when working with analysis results, e.g. the modifiable areal unit problem (MAUP) or the ecological fallacy. The quality issues can be derived from classical statistics and account for statistical as well as for spatial statistical analysis. The latter named examples only apply to spatial data and methods. Those requirements for spatial data will be outlined in the following chapters. Requirements for statistical analysis like confidence interval or significance tests shall not be explained in this thesis because it is focussing on the spatial matters in tourist consumer research. That does not mean they can be neglected when doing geospatial analysis in consumer research. They are important steps of the analysis process. Their application in a research context will be demonstrated in chapter 65.

3.3.1. Non-Uniformity of Space and Edge Effects

In spatial analysis of social science data the non-uniformity of space is a problem. Clusters of humans develop due to the environment and have to be considered in analysis. This is especially a problem with high-resolution data (O'SULLIVAN & UNWIN 2003, p. 33f). For example in tourism, visitors cluster in certain areas simply because the accommodation facilities are built in certain places. In other areas, there might be gaps caused by protected or inaccessible areas where no settlements exist. These appearances are not unexpected. Even if the methods for tackling the problem of non-uniformity of space are still poorly developed, they must be mentioned in analysis. A special case of non-uniformity of space arises in most studies due to the delineation of the study area. Events or features at the edge of the research area do not have neighbours in all directions in analysis processes although they might be influenced by neighbours from outside the study area. This so-called edge effect occurs mostly when study areas are arbitrarily defined to keep a study area manageable and spatial dependencies are to be analysed (DIGGLE 2003, p. 5). As a workaround, events or features from outside the study 5 Detailed discussion on standard statistical requirements can be found in statistical textbooks like Bahrenberg, Giese, & Nipper (1990), Heiberger & Holland (2006), Huizingh (2007), Ross (2005), or Triola (2008).

21


area can be considered in analysis, either as real objects or from simulation. The problem diminishes when working on islands, like Mallorca.

3.3.2. Modifiable Areal Unit Problem and Ecological Fallacy

Spatial data for social sciences are usually available in aggregated form associated with some areas. If they are available on individual level, spatial aggregation is a common way to achieve generalisation. The same technique is used to summarise smaller areas to bigger ones, either for generalisation or to compare with data only available on a courser spatial resolution. The areas used can rarely be seen as natural. Mostly the borders are arbitrarily created with no respect to the underlying phenomena of research. The best examples are census or economic statistics that are usually available within administrative boundaries (e.g. communities or census blocks). But the system of zoning heavily influences the outcome of analysis. OPENSHAW & TAYLOR are said to be the ones who termed the problem the Modifiable Areal Unit Problem (GOODCHILD 2005, p. 10). They showed that the relationship between the percentage of people above 65 and the percentage registered as Republicans could range from perfect positive correlation to perfect negative correlation, depending on the zoning design (ibid.). The MAUP has 2 different occurrences: the zoning effect and the scale effect. The zoning effect occurs if the zoning is changed but the scale of the data is stable. The scale effect describes the case of changing results caused by analysis on a different resolution (WONG & LEE 2005, p. 8). The scale effect of the MAUP is closely related to another common pitfall of geographic data, the ecological fallacy. The ecological fallacy denotes the mistake of drawing conclusions from an aggregated level to a finer scale. An individual does not necessarily share the characteristics of the group it is aggregated to by an area (WEEKS 2004, p. 388). While for the ecological fallacy acknowledging the problems is enough to keep people from drawing wrong conclusions, for the MAUP few solutions are developed. Some scientists like MONMONIER (1996, p. 196f), WONG & LEE (2005, p. 10), or WEEKS (2004, p. 388) suggest playing around with the data on different scales to see if results are stable. Another solution suggested by OPENSHAW & ALVANIDES (1996) is the avoidance of re-aggregation. That would mean to analyse the data on the finest available level of spatial resolution. That is often not feasible or due to the research topic not wanted. Another solution offered by OPENSHAW & ALVANIDES (ibid.) is to use the re-aggregation as a deliberate tool. Thus the zoning will be used to optimise a selected function or relationship by modifying the zoning system. The output is a partition of the research area into spatial units (OPENSHAW.S. & ALVANIDES

22


1996). Although some implementations of such kinds of algorithms have been made6, OPENSHAW’S idea did not become widely accepted. O’SULLIVAN & UNWIN (2003, p. 32) assume the computational complexity and the requirements for very fine scaled data to be the reasons. Finally it can be concluded that the spatial scale of analysis is to be considered very carefully. If zoning is possible, the zones have to be chosen carefully. If the analysis is not feasible or not wanted to be done without re-aggregation of the data, a multi-scale approach is recommendable to account for possible changing results due to MAUP.

3.3.3. Data Quality Documentation

To evaluate the quality and reliability of analysis output, sufficient documentation of errors and uncertainty in the data as well as documentation of confidence levels, error margins and statistical significance is necessary (cp. BURROUGH 1994, p. 187/192). For spatial consumer analysis in tourism this problem is twofold as data quality issues and error propagation have to be documented for spatial and statistical data. While for classical statistical data functions for measuring errors (confidence intervals) are available for spatial data no widely accepted theory of uncertainty and error propagation in GIS exists. Some different approaches have been developed7. Errors and uncertainty, however, affect processes in different ways, depending on the geospatial analysis in operation. That is why error documentation would have to be implemented individually for each process (BRIMICOMBE 2003, p. 206). This might be a reason why such error estimation and documentation is not widely implemented, neither in GIS software nor by the users. MAGUIRE (2005, p. 63) assumes that the complexity of the available methods keeps the majority of users from applying it. Others think that the users want to believe in the precision of computers and therefore avoid bothering about errors and uncertainty (see MAGUIRE, BATTY, & GOODCHILD 2005, p. 452). But that is not a problem only to geospatial analysis. Even for empiric research in tourism BUSCH & LEITNER (1992, p. 167) allege that some researchers deliberately do not mention the absolute sample size or the number of cases of a fraction because they want to distract from small numbers of cases. That hinders the people, who are reading the reports, to assess the reliability and precision of the results and could lead to wrong decisions made on basis of such bad documented analysis results.

6 See ZDES developed by the School of Geography of the University of Leeds (http://www.geog.leeds.ac.uk/software/zdes/) or the implementation of OPENSHAW’S algorithm by PALLADINI (PALLADINI 2004). 7 Some approaches are summarised and listed by KRIVORUCHKO & CRAWFORD (2005) or by BRIMICOMBE (2003, p. 206 - 234).

23

http://www.geog.leeds.ac.uk/software/zdes/


Therefore, it is good common practise in statistics to name or picture error margins directly or to provide figures that help the beholder to calculate or estimate the precision and significance of the outcome. Those figures are for example the number of cases in sample, the number of valid cases for a particular variable and fraction, standard deviation, or the confidence level for the result. For spatial data and results derived from spatial data the scale zoning level is to be stated, even if the data are not depicted as a map.

24

4. The Research Location: Mallorca

25


4.1. Geography of Mallorca

The Spanish island Mallorca is situated in the Mediterranean Sea and belongs to the archipelago of the Balearic Islands. Next to Mallorca, the Balearic Islands comprise Menorca, Cabrera, Ibiza (also known as Eivissa), Formentera and some small uninhabited islands. Mallorca is the biggest of the Balearic Islands. With its 3,640 km² Mallorca is roughly 3 times as big as the Dutch Ijsselmeer.

Llucmajor

Arta

PalmaManacorCalvia

Felanitx

Campos

Pollenca

Escorca

Santanyi

Inca

Petra

Algaida

Muro

Porreres

Andratx

Sineu

LlubíBunyola

Selva

Alaró

Alcudia

Soller

Santa Margalida

Sencellas

Montuíri

Capdepera

Marratxínet

Esporles

Sant Joan

Ariany Son ServeraPuigpunyent

Valldemossa

Campanet

Ses Salines

Sant Llorenc des Cardassar

Deia

Binissalem

Sa Pobla

Costitx

Fornalutx

ConsellBanyalbufar

Estellencs

S p a i nS p a i n

F r a n c eF r a n c e

I t a l yI t a l y!(

MallorcaMenorca

Eivissa (Ibiza)

Formentera

Cabrera

Palma de Mallorca

Mediterranean Sea

Population

293 - 3,5003,501 - 8,5008,501 - 17,00017,001 - 45,000380,000

Sources:Population:Institut Balear d' Estadística 2006Geodata:IDEE 2007 (Mallorca and Balearic Islands)ESRI 2003 (Europe)created by Johannes Luberichs, 2008

0 5 10 15Kilometers

Map 1: Geographic location of Mallorca and population in Mallorca

The Balearic Islands are the smallest Spanish autonomous region. In contrast to the other Spanish autonomous regions, the Comunitat Autònoma Illes Balears (CAIB) is not further subdivided into provinces. The Government of the Balearic Islands is situated in Palma de Mallorca, which is the capital of the CAIB. Mallorca is composed of 53 Municipios. These


Municipios are the smallest administrative units (SCHÜRGERS 2005, p. 3). They will be referred to in the following text as either municipalities or communities. 800,000 people are living in these 53 municipalities, nearly half of them in Palma (INSTITUT BALEAR D' ESTADÍSTICA 2006b). Mallorca can be subdivided into 3 main natural geographical regions. The mountains of the Serra de Tramuntana in the North West cover nearly a quarter of the island. It mostly consists of limestone from the Mesozoic, which formed bizarre karst formations in some locations. In most places, it falls steeply into the sea with only some small beaches. The highest peak is the Puig Major with 1,443 metres (BREUER 1992, p. 12ff).

a) natural landscapes

c) mean temperature January (°C)

b) yearly mean precipitation (mm)

d) mean temperature July (°C)

=< 400> 400 - 500> 500 - 600> 600 - 700> 700 - 800> 800 - 900> 900border of aridity

=< 8> 8 - 9> 9 - 10> 10 - 11> 11

< 24> 24 - 25> 25 - 26> 26

coastal typeCoast with integrated small bays (Cala)Sand beach coast in basinSteep coast

Sources:Breuer 1992; Schmitt 1999; Schürgers 2005

created by Johannes Luberichs, 2008

0 5 1015Kilometers

Map 2: Mallorca’s geographical segmentation and climate

On the eastern side of the island, there is a low mountain range called Serres de Llevant. It runs parallel to the Serra de Tramuntana but reaches the coast only in the north, where it covers the area between the eastern shores of the Bay of Alcudia and Cap de Capdepera. It reaches altitudes of around 600 metres. At the eastern side of the Serres de Llevant, marine abrasion terraces (named Marinas) from Mesozoic limestone characterise the coast. Caused

26


by ancient tectonic lifting, these Marinas rise above the sea and only some small bays8 grant easy access to the sea. The Marinas are part of the central lowlands of Mallorca. These lowlands (El Raiguer, El Pla, El Migjorn) spread in the middle between both mountain ranges (BREUER 1992, p. 16f). Their position on the lee side of the Serra de Tramuntana causes a dry-warm, wind protected climate (SCHÜRGERS 2005, p. 4). The mountains force the air to ascend and release its moisture as rain mostly at the north-western edge of the Serra de Tramuntana. That leads to low precipitation on the eastern side of the mountains that declines with growing distance and reaches its lowest level in the south of the island (BREUER 1992, p. 13/14). Another factor affecting climate is the surrounding sea. The island position mitigates the temperature amplitude over the year. The monthly mean temperature ranges from 7 degrees Celsius on the peaks of the Serra de Tramuntana in January to over 26 degrees Celsius in July in the centre of Mallorca (SCHÜRGERS 2005, p. 4). The island with its different landscapes, the warm and dry climate, and a pulsating capital offers good conditions to satisfy the tourist demand of very different target groups. These resources have been heavily exploited as will be outlined in the following chapter.

4.2. Tourism in Mallorca

Since the end of the 1950s, Mallorca has developed into a prototype of a mass tourism destination with all its concomitant phenomena. The number of visitors rose steeply from around 400,000 in 1960 to nowadays over 9.6 Million (see figure 6; SCHMITT 2007, p. 22). The tourism, which accounts for approximately 80% of the island’s gross domestic product (GDP) (SCHMITT & BLAZQUEZ I SALMON 2004, p. 2), makes Mallorca the richest province in Spain (SCHOLZ 2002, p. 239). Due to this importance of tourism, Mallorca highly depends on the development of the European tourism markets. Because the first phase of the tourism boom was characterised by uncontrolled development, the local government of the Balearic Islands started in the middle of the 1980s to implement some laws to control the building of new tourism facilities. Other laws and programs followed in the 1990s to improve the hotel quality through modernisation accompanied by public investment in the infrastructure. 1991 the first marketing plan was introduced (cp. BARDOLET 1992, p. 51/52). From then on the way was meant to lead away from mass tourism to more quality tourism (SCHMITT 2000, p. 54).

8 These bays are called Cala in Spanish and have been introduced in the names of a lot of tourism villages and resorts.

27


Figure 6: Development of tourist arrivals in Mallorca

Nevertheless, the tourism in Mallorca still shows the characteristics indicating that it will be a long way to achieve a substantial change. These characteristics were listed by different authors (BARDOLET 1992; BUSWELL 1996; KOCH & WACHOWIAK 2007; SCHMITT 1999; SCHOLZ 2002):9

• Mass tourism: 9,614,055 visitors in 2006. • (High) seasonality: 55% of visitors in high season, 28% in mid-season, 16% in low

season of 2006. • The low and middle quality segment dominates: In 2005 21% of the tourism

accommodations were low class, 54% belong to 3-star category. • Mostly package tours and hotel accommodations: 54.4% travel to the Balearic Islands

on package tours, 75.3% stay in hotels or similar tourist accommodations. • Tourism is foreigner dominated: 80.4% of visitors come from foreign countries. • Dependent on few source markets: In 2006, Germany and Great Britain accounted

for 62.6% of all visitors. • Tourism dominance on economy: 80% of GDP generated by tourism (see above). • Spatially concentrated at the coastline.

9 The list is conducted from the named sources. The values are, if not stated otherwise, derived from IBESTAT (INSTITUT BALEAR D' ESTADÍSTICA 2006a) for 2005 and from CONSELLER DE TURISME (2007) for 2006.

28


But why did Mallorca develop into such a mass tourism destination? At the end of the 1950s the Spanish dictator Franco changed visa regulations and opened some places to tourism and the associated foreign investments. In that time the economy of Western Europe grew fast. With more money and time at their disposal, especially people from Northern Europe discovered the Mediterranean countries as holiday destinations. The introduction of charter flights made even islands like Mallorca accessible. Despite the necessity of flying, the comparatively cheap costs of living in Mallorca made the holiday affordable for people from Northern Europe. Once the Mallorcan economy had changed from a poor agriculture-based economy to a rich tourism dependent one the Government of the Balearic Islands improved the relevant infrastructure to keep up with customer demand and to defend its market leading position (BARDOLET 1992, p. 36ff; KUNZE 1992, p. 105ff). All that, however, would not have worked if Mallorca would not possess a natural attractiveness. Mostly, its success is owed to its convenient climate with warm, sunny summers and mild winters, high quality beaches with clean water, and very different landscapes and sceneries (BARDOLET 1992, p. 39). The supply is complemented by cultural sights in Palma and artificial attractions like golf courses, casino, aquarium, and so on. For the last decade, the Government has tried to diversify more into new quality market segments developing offers in nautical, golf, agro- and residential tourism (SCHMITT 2000).

29


4.2.1. The Spatial Distribution of Tourists in Mallorca

Most accommodations of Mallorca concentrate at the coast around the Bay of Palma. The municipalities of Calvia and Palma comprise together more than a third of all tourist beds of Mallorca (INSTITUT BALEAR D' ESTADÍSTICA 2007).

Llucmajor

Palma

Calvia

Santanyi

Andratx

Ses Salines

Soller

Capdepera

Son ServeraSt Llorenc des Cardassar

Santa Margalida

Muro

Alcudia

Pollenca

FelanitxManacor

Sources:based on Institut Balear d' Estadística 2007 created by Johannes Luberichs, 2008

Number of Beds(count of municipalities)

=< 1000 (32)1000 - 8000 (5)8001 - 15000 (4)15001 - 19000 (3)19001 - 30000 (2)30001 - 59750 (2)

Quality Classification of Accommodations

5 Star hotels4 Star hotels and apartments3 Star hotels and apartmentslow class and unrated accommodation

(municipalities with over 1,000 beds)

0 5 10 15Kilometers

Map 3: Distribution of accommodations in Mallorca 2006

Of course these are also the spots where most tourists stay. Together these municipalities receive around 40% of all tourists. Other communities with a high number of beds are Alcudia in the north and Sant Llorenc de Cardassar at the Costa de Llevant. The hinterland and the Tramuntana mountain range offer only few places for accommodations and thus are the destination for only a small part of the visitors (see map 3 and map 4). The tourism and population density in these areas are very low.

30


Municipality Beds per meter coast

Max. people density per meter

coast*

Coast length per bed in meter

Min. coast length per person in meter*

Sant Llorenc des Cardassar 3.41 4.14 0.2 0.2 Muro 2.89 3.83 0.4 0.3 Son Servera 1.18 2.15 0.7 0.5 Santa Margalida 1.08 1.81 0.5 0.6 Calvia 1.03 1.69 0.8 0.6 Palma 0.99 9.32 0.8 0.1 Alcudia 0.66 1.01 1.2 1.0 Capdepera 0.58 0.86 1.5 1.7 Manacor 0.47 1.58 1.7 0.6 Ses Salines 0.34 0.69 2.3 1.4 Santanyi 0.33 0.52 2.8 1.9 Felanitx 0.31 1.11 3.1 0.9 Llucmajor 0.25 0.89 4.0 1.1 Soller 0.19 1.05 5.3 1.0 Fornalutx 0.13 1.01 7.8 1.0 Pollenca 0.1 0.31 9.2 3.3 Andratx 0.08 0.31 10.9 3.3 Deia 0.05 0.13 19.3 7.9 Arta 0.03 0.26 31.2 3.9 Banyalbufar 0.02 0.08 41.3 12.2 Campos 0.02 0.78 18.2 1.3 Estellencs 0.02 0.07 56.2 14.2 Valldemossa 0.02 0.25 66.2 4.0 Escorca 0.00 0.01 - 105.1 The coast length is calculated with from data for a maximum scale of 1:25,000 (IDEE 2007a) * calculated from the population (INSTITUT BALEAR D' ESTADÍSTICA 2006a) and tourism statistics. Because for the tourists no absolute figures were available on municipality level, the maximum number of tourists is derived from the number of beds and the maximum occupancy rate in Mallorca, what was 92.5% in August 2005 (CONSELLER DE TURISME 2006, p. 35). Holiday homes have not been taken into account.

Table 1: Tourism density indexes at the coast of Mallorca for 2005

In the areas of high concentration tourism reaches a density of more than 1 bed per meter coast. To put it the other way round, in high season less than one meter coastline is spared for every tourist in the northern municipalities of Muro and Santa Margalida, in the southern municipalities of Calvia and Palma in the south as well as in some places at the Costa de Llevant. Taking the population into account that might use the beaches and coast as well for recreational activities, the space per person shrinks to as few as 0.1 meters in the area of Palma de Mallorca (see table 1).

31


Llucmajor

Palma ManacorCalvia

Felanitx

Pollenca

Santanyi

Petra

Muro

Andratx

Alcudia

Soller

Santa Margalida

Capdepera

Son Servera

Ses Salines


Sources:based on Conseller de Turisme 2007, p. 61 created by Johannes Luberichs, 2008

% of tourists(count of municipalities)

29.8 % (1)22.1 % (1)4.7 - 7.8 % (2)0.8 - 4.6 % (13)< 0.8 % (34)none (2)

0 5 10 15Kilometers

Map 4: Municipality of destination for all tourists in Mallorca 2006

These figures are an indicator of the high concentration of tourism and the impact on the coast. They do not show the whole picture of tourism concentration because a big share of the Mallorca tourism market is devoted to residential tourism. Nearly a fifth of the visitors stay either in their own holiday home or apartment or in the property of friends or relatives (CITTIB 2003bp. 1). According to a study of the Balearic University on Mallorca a fifth of the island’s area is already owned by foreigners. The majority of these foreigners are Germans (SCHMITT 2007, p. 23). About the occupancy rates nothing is known. That is why they are not included into the density indexes in table 1. In the municipalities Andratx and Calvia at the Costa de Ponent, in Santanyi and Alcudia, where more than 60% of the houses are holiday homes (see map 5), the density indexes would have shown an even higher concentration of tourism.

32


Map 5: Proportion of second homes per municipality 2002

The Germans are the biggest group of tourists on Mallorca. The second largest group of visitors comes from Great Britain. Together these 2 source markets account for nearly 2 thirds of the visitors to Mallorca. Spanish tourists come only third in line with a share of less than a fifth of the visitors (see table 2). The different nationalities prefer different areas for their holidays. Spanish visitors prefer the city of Palma as place of stay in Mallorca. The area around S’Arenal is also highly favoured (see map 6). 65% of all Spanish visitors come to these 2 regions (CONSELLER DE TURISME 2008, p. 22). The British visitors dominate the Costa de Ponent with 937,000 visitors in 2007. That is around 40% of all British tourists in Mallorca. Most of them stay at the Bay of Palma, the part of the Costa de Ponent belonging to the municipality of Calvia (CONSELLER DE TURISME 2007, p. 61) where the tourism industry is adjusted to the British taste (BUSWELL 1996, p. 313; SCHOLZ 2002, p. 241).

33


Nationality Total arrivals % of total

German 3,378,986 38.4% British 2,230,045 25.3% Spanish* 1,517,507 17.2% French 238,071 2.7% Swiss 159,894 1.8% Irish 136,262 1.5% Italian 134,616 1.5% Swedish 129,004 1.5% Austrian 125,859 1.4% Dutch 119,349 1.4% Danish 95,908 1.1% Belgian 94,343 1.1% Norwegian 92,767 1.1% Portuguese 82,938 0.9% Luxemburger 27,330 0.3% Finish 18,544 0.2% Other European 57,937 0.7% Russian 17,945 0.2% Other 80,858 0.9% Not differentiated foreigner arrivals by sea* 63,995 0.7% Total 8,802,155 100 * The number of Spanish tourists is the only that includes the arrivals by sea. All other values only picture the arrivals by air. Details on the nationality of foreigners arriving by sea are not available. Therefore, all but Spanish arrivals by sea are summarised in the last row. (CONSELLER DE TURISME 2006, p. 19)

Table 2: Tourist arrivals in Mallorca 2005 differentiated by nationality

The Playa de Palma with its in Germany famous place S’Arenal is a very small but crowded stretch along the coast. One fifth (around 700,000) of the German visitors choose it as their destination. With just above 800,000 visitors in 2007, only the much bigger region Costa de Llevant receives more German visitors (CONSELLER DE TURISME 2008, p. 22). From all the municipalities the region Costa de Llevant is composed of, Capdepera at the northern end is the most visited by Germans (CONSELLER DE TURISME 2007, p. 62). Here the famous tourism village Cala Rajada is to be found.

34


N o r t ho u t s i d e

t o u r i s t r e g i o n s

C o s t a d

e Ll e

v a n t

P a l m a

C o s t a d e T r a m u n t a n a

C o s t a d e P o n e n t

P l a y a d e P a l m a

Sources:based on Conseller de Turisme 2008, p. 22 created by Johannes Luberichs, 2008

# of tourists(count of regions)

200,000 - 320,000 (2)1,500,000 - 1,750,000 (3)1,995,441 (1)2,356,767 (1)

0 5 10 15Kilometers

% of tourists

GermanBritishSpanishOthers

Map 6: Region of destination in Mallorca for tourists differentiated by nationality for 2007

4.2.2. Low-Cost Carrier in the Air Traffic to Mallorca

Despite the proliferation of low-cost airlines in Europe since the beginning of the new millennium, no generally accepted term for these new airlines exists. They all have in common that they differentiate from the established airlines in their business model and that the price is the central marketing instrument to gain new customers and bigger market shares (SCHREIBER 2006, p. 139ff). Next to low-cost carrier, common terms are no-frills carriers (AEA ASSOCIATION OF EUROPEAN AIRLINES 2005, p. III-2) describing the focus on service abandonment or low-fare airline (ELFAA 2008) describing the focus on cheap tickets. According to SCHREIBER the low-cost carrier is the most common term in the literature on that matter (2006, p. 193). In its origin meaning it describes the business model of airlines “to operate against lowest possible costs” (ZWAN 2007). The customer only perceives the price but not the business model. That is why from the customer’s point of view an airline appears to be a LCC if it offers “a very good deal”, as ZWAN put it (2007).

35


For this thesis the term low-cost carrier will be used. Not only because it is mostly used in literature, but also because it indicates both: the business model operating at low costs and the low costs the customers have to pay for their tickets. Schreiber took the possible techniques of cost reducing to identify LCC and categorise them according to their origin and business model. Germanwings, the airline whose passengers are interviewed for this study, is what Schreiber calls a hybrid airline. They have no frills but do not tap the full potential of cost reducing. With their standing in the LCC market and as a growing airline in Mallorca air traffic10 they are considered representative for huge parts of the Mallorca LCC market.

bmibaby (140,237)FlyMe (2,020)

EUJet (3,451)germanwings (48,718)HLX (164,004)Skyeurope (5,501)Thomsonfly (75,159)Vueling (77,458)

easyjet (451,297)WIZZ (2,031)Ryanair (entered Mallorca

Market 2006)

Helvetic.com (15,371) Sterling (24,872)Transavia (67,601)Virgin express (19,356)Windjet (11,194)

Airberlin (2,285,194)Niki (122,722)

„Spec

ialist

“

„Aban

donment“

„Serv

ice“

„Hyb

rid“

No-fri

llsAp

proac

h

Genuine Low-Cost Approachlow middle high

lowmi

ddle

high

Matrix adopted from Schreiber 2006, p. 195Passenger arrivals from AENA 2008

Figures in brackets are the passenger arrivals at the airport of Palma de Mallorca in 2005.

bmibaby (140,237)FlyMe (2,020)

EUJet (3,451)germanwings (48,718)HLX (164,004)Skyeurope (5,501)Thomsonfly (75,159)Vueling (77,458)

easyjet (451,297)WIZZ (2,031)Ryanair (entered Mallorca

Market 2006)

Helvetic.com (15,371) Sterling (24,872)Transavia (67,601)Virgin express (19,356)Windjet (11,194)

Airberlin (2,285,194)Niki (122,722)

„Spec

ialist

“

„Aban

donment“

„Serv

ice“

„Hyb

rid“

No-fri

llsAp

proac

h

Genuine Low-Cost Approachlow middle high

lowmi

ddle

high

Matrix adopted from Schreiber 2006, p. 195Passenger arrivals from AENA 2008

Figures in brackets are the passenger arrivals at the airport of Palma de Mallorca in 2005.

Figure 7: Supply-side matrix of low-cost carrier offering service to Mallorca

STOSCH (2007, p. R/4) recently mentioned that the LCCs dominate now the transportation business to Mallorca. With Air Berlin one single LCC accounted 2005 for 21.6% of all arrivals in Mallorca and 2007 already for 24.8% bringing 2.8 million passengers to Mallorca (AENA 2008a; AENA 2008b). In the year of the surveys 2005, all airlines classified by Schreiber as LCC (see figure 7 above) accounted for 33.2% of the passengers to Palma de Mallorca (derived from AENA 2008a). Assuming that the 3.4 million German tourists in 2005 showed a similar behaviour would mean approximately one 1.1 million Germans arriving by LCC in Mallorca.

10 In 2007 already 143,146 passengers arrived in Palma de Mallorca by Germanwings (AENA 2008b).

36

5. The Research Data

37

5. The Research Data The Mallorcan tourism department publishes yearly statistics on the tourism in Mallorca. But they provide only limited information on the spatial distribution of tourists and usually only basic information on the characteristics of the visitors. For the year 2002, data on the profiles of the main visitor groups were available, but they cannot be spatially differentiated (CITTIB 2003a; CITTIB 2003b). That is why the main attribute data for this research are taken from tourism surveys. These surveys are part of a bigger research project focused on the Balearic Islands by the University of Applied Sciences Bad Honnef / Bonn. Within this research project the profiles of the German visitors to Mallorca are investigated by different polls. These polls have been supervised by Prof. Dr. WACHOWIAK of the University of Applied Sciences Bad Honnef / Bonn. The data were provided as raw statistical data (WACHOWIAK 2005a; WACHOWIAK 2005b).

5.1. Survey Design and Data Collection

This study is based on a dataset derived from a poll on German tourists flying by LCC to Mallorca. It was conducted in 2005 on flights of the airline Germanwings between Cologne Bonn Airport and Aeroport de Son Sant Joan, the only international Airport in Mallorca. Some of the questions between outbound and return flights differ, but many are equal or similar and can be compared11 (WACHOWIAK 2005b). A second poll was conducted during May 2005. German visitors answered a personal face-to-face questionnaire on different places in Mallorca (WACHOWIAK 2005a). Because the questions have been similar to the ones of the LCCP survey both groups can partly be compared what will be done in several selected areas.

5.1.1. Sampling of the Mallorca Polls

The passenger questionnaire has been conducted as self-administered interviews on randomly selected flights. The sample can be considered as sound statistical sample (WACHOWIAK & NEUMANN 2008). Assuming Germanwings as a representative for the German LCC, the sample represents the population of German LCCP visiting Mallorca. The place of accommodation is one of the variables in the survey. For spatial distribution analysis the named places of accommodation have to be regarded as sample locations. As the sample is random in the classical statistical view, the spatial sample follows the rules of what is known in spatial analysis as simple random sampling. It is a type of probabilistic sampling where the sample locations are chosen by some kind of random generator (cp.

11 The interview forms are to be found in the attachments.


BAHRENBERG, GIESE, & NIPPER 1990, p. 19; HAGGETT 2001, p. 713/796; LONGLEY et al. 2001, p. 104), in this case a random sampled questionnaire. There are some drawbacks to pure random sampling. It can produce clusters of samples by accident (cp. HAINING, WISE, & MA 2000, p. 94/95; LONGLEY et al. 2001, p. 105). Compared to spatial systematic sampling (see chapter 3.2.2.2.1) the simple random spatial sample introduces only a small amount of problems. It is considered the best possible practice to sample the spatial distribution of LCCP. During the second poll German visitors where interviewed at certain selected places around Mallorca. But the number of interviews per area does not align with the distribution of the German tourists as known from the Mallorca statistics. Without further processing (e.g. weighting the cases) of the data, conclusions on the German tourist in Mallorca cannot be readily drawn. Moreover, since none of the interview locations were in the hinterland of Mallorca, comparisons between the spatial distribution of LCCP and the common German tourists are not possible. The data from the face-to-face interviews can however be treated as single polls for each particular area around the interview location. Conclusions on particular regions can therefore be drawn and compared to the ones derived from the poll on the LCCP.

5.1.2. Geographical Data

Geographical data are the objects in space to which the attribute information about the tourists, the infrastructure or the environment are bound. Depending on their geographical extent and the scale of analysis they can be represented as point, line, or polygon entities. The geographical data necessary for geospatial analysis in this thesis have been collected by a range of different techniques. While some of the data have been available in ready-to-use vector format, other data had to be created by scanning and digitising paper maps or by transforming raster data. The newly created geographical data always have been tested for plausibility by additional maps to draw comparisons, for example the Mallorca Accommodation website that provides aerial photos on a fine scale together with information on tourist accommodation places (CONSELLER DE TURISME et al. 2007).

5.1.3. Spatial Level of Analysis

During the survey design phase, no exact target scale for analysis was considered. The scale of analysis is only limited by the geographic information derived from the tourist interviews. This information is the tourists’ place of stay, i.e. a village (see map 9d). In a questionnaire like the LCCP survey it is impossible to catch respondents from every possible location. To account for the missing places and places with few respondents, locations will

38


be aggregated into bigger areas. The group of tourists in an area will then be represented by descriptive statistics (count, sum, mean, standard deviation etc.). But which level of aggregation should be chosen? There are some predefined areas. The Mallorcan tourism department defined tourist regions and sectors in order to report tourism statistics (see e.g. CONSELLER DE TURISME 2006, p. 22; and CONSELLER DE TURISME 2008, p. 37/38). These areas have large geographical extents and divide Mallorca in only a few regions. They are good for generalisation and overview. The available statistics for these levels will be used for comparisons. Due to some inconsistencies in the statistics for the regions, the 2 northern regions are aggregated to the sector north to avoid confusion (see map 9a). The tourist regions generalise the information and may hide differences within the large areas. For a spatially more detailed analysis, smaller areas are necessary.

48

59

46

2

22

28

7

2 27

504

85

257

72

1

29

56

3

93

63

6

4

55

2

7

3

36

79

179

21

4

3

13

2

4

Arta

Llucmajor

PalmaManacor

Calvia

Felanitx

Campos

Pollenca

Santanyi

Andratx

Alcudia

Soller

Santa Margalida

Capdepera

Son Servera

Ses Salines

St Llorenc des Cardassar

LCCP % of total(count of municipalities)

< 0.75* (20) 0.76 - 3.0 (8)3.1 - 5.5 (6)10.6 (1)15.2 (1)29.7 (1)no respondent (16)

Calvia48

name of municipalitynumber of respondents

Sources:based on Wachowiak 2005b

created by Johannes Luberichs, 20080 5 10 15

Kilometers* below fault tolerance (confidence level 90%; n = 1696)

Map 7: Spatial distribution of German LCCP in Mallorca on municipality level

The administrative regions, named municipalities, are much smaller than the tourist regions. Most statistical data used for comparison are available only on municipality level. That is why the municipalities will be one of the levels utilised for spatial analysis. Caused by historic

39


settlement patterns, the main villages are not situated on the coastline. The tourist villages at the coast developed as recently as the mass tourism (KUNZE 1992, p. 106). For that reason the municipalities reach deep into the inland areas (see map 7). That shape makes them unsuitable level of aggregation because the areas comprise very different landscapes and the tourism areas are mostly located on the coast. Moreover, there are some municipalities that did not get into the sample and others that received a number of respondents below the fault tolerance. Only 17 municipalities are above that limit (label with a thicker outline in map 7). Therefore, some municipalities will be dissolved in order to analyse and compare data (see map 9b). Since the raw survey data are available, a customised zoning is possible. That is a very delicate issue since the zone design heavily influences the outcome (MAUP, see chapter 3.3.2). Automatic zoning algorithms are only available for specific purposes and software. They have some limitations making them unsuitable for the daisymetric map of the visited places (see PALLADINI 2004). That is why for this study, a zoning system is created by hand. This unfortunately handicaps the traceability. Therefore, the guidelines for the zoning are outlined here. The zoning follows the idea of OPENSHAW & ALVANIDIS (1996) to create units, which accommodate homogenous groups. In addition, a sufficient number of over 30 respondents have to represent every zone. Because of the concentration of the tourism at the coast a zone of 2.5 km along the coast was created. One exception is the Tramuntana mountain range. Here the number of respondents was too low to make a small area only along the coast. The area was created following the municipality borders along the geographic segmentation of the mountain area. In a similar way, the borders in all other zones have been delineated by municipality borders where appropriate. When a municipality border was close to the sea, deviations from the 2.5 km distance buffer have been made. Which of the locations are aggregated to a zone was decided mostly by the characteristics of the tourists, and partly by the characteristics of the landscape. While the latter could be derived from topographic base maps and satellite images, the cumulated characteristics of the tourists have been identified by hot-spot analysis with the Gi* measure. This local statistic of spatial autocorrelation enables the analyst to see where features with high values, or those with low values cluster (ORD & GETIS 1995). The calculation of the Gi* index works through a comparison of the local sum of neighbouring features with the sum of all features. A high difference from the expected value in the local sum leads to a high z-score showing that a feature is very likely to be surrounded by features with similar high or low values. The more the z-score is different from zero, the more intense the clustering. The Gi* has been calculated with randomised locations of the tourists (see illustration in chapter 5.2.1, map 10) for all variables with metric values. For variables with binary-coded nominal values the positive answers (“yes”) have been converted to a 1 and the negative (“no” or no answer) to

40


41

a 0. Places where the characteristic of the respondents are higher than the average appear red in map 8 below, and those places where low values cluster appear blue. In that way, differences between neighbouring places could be identified and considered in the zoning process (see map 8 below).

!!!!!!!

!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!

!!!! !!! !!

!!!!!!!!! !! !!!!!! !! !!!!! ! !!!! ! !!! !!!! !!!!!!!!!! !! !!! ! !!!!! ! !!!!!!! ! !! ! !!!!!!!!! !

!!!! !!! !! !!

!! !!! !!!!!! !!!! !! !! ! !!!! ! !!! !!! !! !! !! !!!!! !!!!!!! !!!!!!! !!!!!! !!!

!!!!!!!!

!!

!!!!!! !!! !!!!!!!

! !!!! !!!! !!!!!!!!! !!!!!!!!!!!!!!

!

!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!! !!!!!!! !!!!!!!!!!!! !!!!!!!!!!!!! !!!!!!!!!!!!!!!!!

!!!!!!!!

!!!!!! !!!

!!!

!

!

!

!!!!

!!

!!!!!!!!!

!!

!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!

!

!

!!!!!!

!!!!!! !!!!!!! !!!!!!!!!!!!

!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!

!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!

!!

!!!! !!

!

!!!

!!!!!!!!!!

!!

!

!!!

!

!!!

!!!

!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!

!!!!!!

!!!!!! !!!

!!

!

!!!!

!!!

!!

!!!

!!

!!!!

!!

!!!!!!!

!

!!!!! !

!

!!!! !!!!!!!!

!!!!!!

!

!!!!!!!!!!!!! !!!!

!!!

!!!!!!!!!!!!!!!!!!! !!!

!!!!! !!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!! !!

!!!!!!!!

!!

!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!

!

!!!

!!!!!!!!!!!!!!!

!!!!!!!

!!!!!!

!!!!!

!!!!

!!!!!!!!

! !!! !!!!!! !

!!!

!

!!

!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!

!!!!!!!

!!!

!!!!!!!!!!

!!!

!!!!!!!!!!!

!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!! !!!!!!!!!!!!!!!!

!!

!!! !!!!

!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!

!!

!! !!

!!

!!!!!!!!!!!! !!

!!

!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!

!!!

!

!!!!

!

!! !!! !!!!!!!!! !!!!!!!!!!!!!!! !!!!!!!!!!!

!

!!!!!!!!!!!!!! !!! !!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!

!

!!

!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!

!

!!!!!!!!!!!!!!!!!!!! !!! !!!!! !!!!!!!!!!!!!!!!!!!! !!!!!!!!!!! !!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!! !!!

!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!! !!!!!!!!!!! !!!!!

!!

!!! !! !!!! !!

!!!!!!!!!!!!!

!!

!

!!!

!

!!!

!!!

!! !!!!!

!!!!!!!!!!!!!!! !!!!!!!

!!!!! !

!!

!!

!! !!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!

!!!!

!!!!!!

!!!!

!

!!!

!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!

!!! !!!

!!

!

!!!!

!!!!

!!!!!!!

!!!

!

!!!!

!!!!!

!!!!!!!

!

!!!! !!!!!!!

!

!!!! !!!!!!!!!!

!!!!!

!

!!!!!!!!!!!!! !!!

!!!

!!!!!!!!!!!!!!!!!!!!

!!!! !!!!!!!!!

!!!!!!!!

!!!!!!!!

!!

!!!!!!!!!!!!!!!!!!!!!!!

!!!

!

!!!

!!!!!!! !!!!!

!!!!!!!

!!!

!!!!!!!!!!!

!!!!!!!!!!!!!!!

! !!! !!!!!!

!!!

!

!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!

!!!!!!!!!!!!!!!!!!!!

!!!

!!!!

!!!

!!!!!!!!!

!!!

!!!!!

!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!

!!!!!!!!!! !!!!!!!!!!!!!!!

!!

!!! !!!!

!!!!!! !!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!

!!

!! !!

!!!

!!!!!!!

! !!!!! !

!!

!!!!

!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!

!!!

!

!!!!!

!

!!!!!!!!

!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!! !!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!

!!! !!! !!!! !!!

!!!!!!! !!! !! !! !!!!!! !! !!!!! ! !!!! ! !!! !!!! !!!!!!!!!! !! !!! ! !!!!! !! !!!!!!! ! !! ! !

!!

!!!!! !!!!

! !!!!! !!! !! !

!!! !!!! !!!!!! !!!! !! !! ! !!!!! ! !!! !!! !! !! ! !! !!!!!! !!!!!!! !!! !! !!!! !!!!!! !!!

!!!!!!!!

!!

!!!!!!! !!! !!!!!!!!

! !!!! !!!! !!! !!!!!!!!!! !!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!

!!!!

!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!

!!!!!! !!!!!!! !!!!!!!!!!!!!!! !!!!!!!!! !!!!!!!!!!!!!!

!!!!!!!!

!!!!!!!!

!!!!! !!!!!! !!!!

!

!

!!!!!

! !!!!!!!!!

!

!! !!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!

!!!!!

!!!

!!!!!!!

!!

!!!

!!

!!!!!! !!!!!!!!!!!!!!!!!!! !!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!

!

!!! !! !!!! !!

!!!

!!!!!!!!!!!!

!!

!

!!!

!!

!!!

!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!

!!!!!! !!!

!!

!

!!!!

!!!

!!

!!!

!!

!!!!!

!!!!!!

!!!!!!!

!

!!!! !!!!!!!

!

!!!! !!!!!!!!!!

!!!!!!

!

!!!!!!!!!!!!!! !!!!!

!!!

!!!!!!!!!!!!!!!!!!!!! !!!

!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!! !

!!!!!!!!

!

!!!!!!!!!!!!!!!!!!!!!!

!!

!

!!!

!!!!!!!!!!!!!!!!!!!

!!!!!!!!!

!!!!!!

!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!

! !! ! !!!!!! !

!!!

!

!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!

!!!

!! !

!!!!!!!!!!

!!!

!!!!!!!!!!!!

!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!! !!!!!!!!!!!!!!!!

!!

!!! !!!!

!!!!!!! !!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!

!!

! !!

!

!!!!!!!

! !!!!! !!

!!

!!!!

!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!

!!!

!

!!!!!

!

!!!!!!!!

!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!

!!!!!!!!!!!!!! !!!!! !! !!!!!! !!!!!!!!!!!!!

!

!

!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!

!

!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!

!!

!!

!!

!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

! !!!!!! !!!!!!!!!!!!!!!!!!!!!!!

!!

!!! !!! !!!! !!!

!!!!!!! !!! !! !! !!!!!! !! !!!!! ! !!!! ! !!! !!!! !!!!!!!!!! !! !!! ! !!!!! !! !!!!!!! ! !! ! !

!!

!!!!! !!!!

! !!!!! !!! !! !

!!! !!!! !!!!!! !!!! !! !! ! !!!!! ! !!! !!! !! !! ! !! !!!!!! !!!!!!! !!! !! !!!! !!!!!! !!!

!!!!!!!!

!!

!!!!!!! !!! !!!!!!!!

! !!!! !!!! !!! !!!!!!!!!! !!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!

!!!!!!!!!!!

!!

!!!

!!!!!!

!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!

!!!!!!!!!!!!!!!!!!!!! !!! !!!!! !!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!! !!!

!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!! !!!!!!!!!!!!! !!!!!

!!

!!! !! !!!! !!

!!!

!!!!!!!!!!!!

!!

!

!!!

!!

!!!

!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!

!!!!!! !!!

!!

!

!!!!

!!!!

!!!!!!!

!!!

!!

!!!!!

!!!!!!

!!!!!!!

!!!! !!!!!!!

!

!!!! !!!!!!!!!!

!!!!!!

!

!!!!!!!!!!!!!! !!!!!

!!!

!!!!!!!!!!!!!!!!!!!!! !!!

!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!! !!

!!!!!!!!

!!!

!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!

!

!!!

!!!!!!!!!!!!!!!!!!!

!!!!!!!!!

!!!!!!

!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!

! !! ! !!!!!! !

!!!

!

!!!

!!! !!!!!!!!!!!!!!!!!!!!!!!

!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!

!!!

!! !

!!!!!!!!!!

!!!

!!!!!!!!!!!!

!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!

!! !!!

!!!!!!! !!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!

!! !!

!!!

!!!!!!!

! !!!!! !!

!!

!!!!

!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!

!!!

!

!!!!!

!

!!!!!!!!

!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!

!!!!!!!!!!!!!!!!! !!!!!!!

!! !!!!!!! !!!!!!!!!!!

!!

!

!

!!!

!!

!!!!!!

!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!

!!!!!

!

!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!! !!!!!!! !!!!!!!

!!!!!!!!!

!!

!!! !!! !!!! !!!

!!!!!!! !!! !! !! !!!!!! !! !!!!! ! !!!! ! !!! !!!! !!!!!!!!!! !! !!! ! !!!!! !! !!!!!!! ! !! ! !

!!

!!!!! !!!!

! !!!!! !!! !! !

!!! !!!! !!!!!! !!!! !! !! ! !!!!! ! !!! !!! !! !! ! !! !!!!!!!!!!!!! !!! !! !!!! !!!!!! !!!

!!!!!!!!

!!

!!!!!!! !!! !!!!!!!!

! !!!! !!!! !!! !!!!!!!!!! !!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!

!!!!!!!!!!!

!!

!!!

!!!!!!

!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!

!

!!!!!!!!!!!!!!!!!!!!! !!! !!!!! !!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!! !!!

!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!! !!!!!!!!!!!!! !!!!!

!!

!!! !! !!!! !!

!!!

!!!!!!!!!!!!

!!

!

!!!

!!

!!!

!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!

!!!!!! !!!

!!

!

!!!!

!!!!

!!!!!!!

!!!

!!

!!!!!

!!!!!!

!!!!!!!

!

!!!! !!!!!!!

!

!!!! !!!!!!!!!!

!

!

!

!!!!!!!!!!!!!! !!!!!

!!!

!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!! !!

!!!!!!!!

!!!

!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!

!

!!!

!!!!!!!!!!!!!!!!!!!

!!!!!!!!!

!!!!!!

!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!

! !!! !!!!!! !

!!!

!

!!!!

!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!

!!!!!!!!

!!!

!!!!!!!!!!

!

!!!!!!!!!!!!

!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!

!!!!!!!!!! !!!!!!!!!!!!!!!!

!!

!!! !!!!

!!!!!!! !!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!

!!

!! !!

!!!

!!!!!!!

! !!!!! !!

!!

!!!!

!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!

!!!

!!!!!

!

!!!!!!!!

!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!! !!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!

!!! !!! !!!! !!!

!!!!!!! !!! !! !! !!!!!! !! !!!!! ! !!!! ! !!! !!!! !!!!!!!!!! !! !!! ! !!!!! !! !!!!!!! ! !! ! !

!!

!!!!! !!!!

! !!!!! !!! !! !

!!! !!!! !!!!!! !!!! !! !! ! !!!!! ! !!! !!! !! !! ! !! !!!!!! !!!!!!! !!! !! !!!! !!!!!! !!!

!!!!!!!!

!!

!!!!!!! !!! !!!!!!!!

! !!!! !!!! !!! !!!!!!!!!! !!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!

!!!!!!!!!!!

!!

!!!

!!!!!!

!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!

!!!!!!!!!!!!!!!!!!!! !!!!!!!!!! !!!!!!!!!!!!!

!!!!!!!!

!!!!!!!!!!!!!!!!!!!!

!!!!! !!!!!! !!!!

!!

!!

!!!

!! !!!!

!!!!!!!

!! !

!

!!

!!!!!

!!!

!!!!!!!!!!!!!!!!

!!

!

!!

!!

!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!

!!!!!

!!

!!! !! !!!! !!

!!!!!!!!!!!

!!

!

!!!

!!

!!!

!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!

!!!!!! !!!

!!

!

!!!!

!!!!

!!!!!!!

!!!

!!

!!!!!

!!!!!!

!!!!!!!

!

!!!! !!!!!!!

!

!!!! !!!!!!!!!!

!!!!!!

!

!!!!!!!!!!!!!! !!!!!

!!!

!!!!!!!!!!!!!!!!!!!!!

!!!! !!!!!!!!!!

!!!!!!!!!!!!!!! !

!!!!!!!!

!!!

!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!

!

!!!!!!!!!!!!!!

!!!!!!

!!!!!!

!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!

! !! ! !!!!!! !

!!!

!

!!!!

!!!! !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!

!!!!!!!!

!! !

!!!!!!!!!!

!!!

!!!!!!!!!!!!

!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!

!!!!!!!!!!

!!! !!!!

!!!!!!! !!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!!!!!

!!

!! !!

!!!!!

! !!!!! !!

!!

!!!!

!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!!!!

!!!

!

!!!!!

!

a) age

c) sojourn time (less than 60 days)

b) package tourists

d) family company

e) visitors liking the beaches

f) visitors liking the landscape

Sources:based on data from Wachowiak 2005b


Spatial relationship conceptualised as zone of indefference with a threshold of 5,000 m Euclidean distance.

Gi* Z-Score per respondent< -2-2 to - 1-1 to 11 to 2> 2visited tourism areas in samplenew designed tourism zones

!

!

!

!

!

Map 8: Gi* application for customised zoning design


The zoning process resulted in 20 different zones with numbers of respondents ranging from 36 to 241. The areas with low numbers might not be suitable for certain analyses where the visitors are divided into subgroups. Nevertheless, the differences for some variables looked so strong that the creation of the zones seems appropriate. Of course this method of zoning has some limitations. The Gi* index can be influenced by outliers in the variable it is calculated for. That is why e.g. the Gi* z-score for the sojourn time (see map 8c) is calculated only for visitors staying less than 60 days. To provide different views on the data and thus acknowledge the MAUP and address the subjectivity problems introduced by customised zoning, the analysis will be done on multiple scales (shown below in map 9): regional level (map 9a), municipal level (map 9b), and zonal level (map 9c). For some analyses even the pure sample in form of the visited places could be used (map 9 d). This are for example spatial queries or buffer analysis, where ad-hoc areas are build to summarise the LCCP to ad hoc areas for this special geospatial analysis.

42

North

outside tourist regions

Costa de Llevant

Palma

Costa de Tramuntana

Costa de PonentPlaya de Palma

92

553

140

46

350254

261

Tramuntana

- East - West

Serres de Llevant

Palma

SantanyiCosta El Migjorn

Badia d'Alcudia

CapdeperaAndratx / Calvia

Alcudia / Badia de Pollenca

Cala d'Or

central Costa de Llevant

Bahia de Arta

Badia de Palma - West

- South

Playa de Palma/ s'Arenal

Can Pastilla Playa Sa ComaCosta de Ponent

- North

I n l a n d40

78

7645

49

62

43

36

48

58241

179

6468

39

181

71

221

56

InlandArta

Llucmajor

PalmaManacorCalvia

FelanitxCampos

Pollenca

Tramuntana

Santanyi

Andratx

Alcudia

Muro / Santa Margalida

Capdepera

Son Servera

Ses Salines


53

48

48

59

46

22

28

36

504

27

85

56

55

79

25793

179

21

a) Mallorca Tourist Regions

c) Mallorca Tourist Zones

b) Mallorca Municipalities

outside tourist regionstourist regions Source:

based on Conseller deTurisme 2008

coastal zonehinterlandTramuntana Source:

created byJohannes Luberichs

Source:based on IDEE 2007

created by Johannes Luberichs, 2008Map 9: Spatial levels of analysis

!

!!

! !

!

!

!

!

!

!

!

!!

!

!!!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!!

!!

!

!

!

!!

!

!

!

!

!

!

!

!

!!!

!!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!!

!

!

!

!

!

!

!

!!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

!

Cala Millor

Cala D'Or

Inca

Campos

Soller

Calvia

Sa Coma

Manacor

Alcudia

Sa ComaAndratx

Peguera Felanitx

Santanyi

S'Arenal

Pollenca

Llucmajor

Cala Rajada

Santa Ponca

Can Pastilla

Palma

d) Visited Places

0 5 10 15Kilometers

Source:based onDuMont Reisekartographie 2007ESRI 2003, ESRI 2007IDEE 2007Wachowiak 2005a/b

Artaoriginal municipality bordersaggregated municipality areas

visited places in sample!

original municipality borders179 count of respondents

179 count of respondents 179 count of respondents


5.2. Data Input and Management

5.2.1. Geocoding of the Mallorca Polls

In order make the data working in a GIS environment and to do geospatial analysis, the tabular data from the Mallorca tourist surveys have to be geocoded. Each case in the surveys is assigned to a place that is spatially referenced on earth. In this study it is done by a place name. These places are the ones the tourists named on the question where they stayed during their holidays. They are geometrically represented either as area drawn around the buildings of a location (village) or as point in the centre of it. The simple procedure of geocoding usually is to link the entries in the database with the geographic location by an identifier given as code for a unique place. This procedure is illustrated in figure 8.

Figure 8: Geocoding respondents

But the identification of the right place to synchronise the unique identifier in the database and the geographic data bears some problems. One is the spelling. Place names are spelled differently due to the bilingual nature of the Balearic Islands (Catalan, Castellan) or just simply misspelled by the respondent. For example, Port d’Andratx has 13 differently spelled versions in the database. That is why a list with standardised names and unique identifiers had been created to make the links between the data work. Nevertheless, the identifiers had to be assigned to the right entries in the place name column of the LCCP data table. Santa

44


Ponca and Santa Ponsa for example are 2 different places located far from each other. But each exists, depending on the language, in both notations. In this case one place had been excluded because it consists of only some farms more than 5 kilometres away from the sea and has no infrastructure to accommodate tourists. Therefore, the tourists have been assigned to the other Santa Ponca, which is a complete developed tourist village at the coast of Calvia. Another source of uncertainty might be introduced by misinterpretation of the question or by an abbreviation of the village name. An example for both cases is Sant Jordi. A village named Sant Jordi is situated 5 kilometres away from the Playa de Palma and does not seem to offer official tourist accommodations12. At the Playa de Palma a hotel named Sant Jordi exists. People could have had this in mind when answering the question. Giving a hotel name instead of a place name has not been an uncommon mistake. Or they possibly meant Colonia de Sant Jordi, a village in the south west with well developed tourism infrastructure, but abbreviated that name. Assigning the cases to the villages was done with the assumption that people knew the place where they stayed and did not misinterpret the question. Another problem cropped up by places with more than one occurrence in Mallorca. If 2 or more places have the same name, it cannot be doubtlessly guessed which one the tourist meant. Unfortunately that applies to the popular place Sa Coma. Excluding the visitors of Sa Coma from the analysis would cause quite heavy shifts in spatial analysis as Sa Coma counts for more than 3% of the interviewed LCCP. So the tourists have been assigned to the places with the same name after a key depending on their kind of accommodation, the hotel beds available, and the size of the village13. That of course means that some assumptions about that relationship influence the results of geospatial analysis. After the first exploratory data analysis it however became that clear this introduces smaller errors than the exclusion of these uncertain cases. One place of course could receive more than one tourist. Thus the connection to the survey data corresponds to a one-to-many relationship, depictured in figure 8 above. By establishing this link for each place, the statistics could be calculated and the number of respondents could be counted. For visual analysis the representation of many cases by one single feature introduces some difficulties. When doing queries, a place would be highlighted as selected regardless of whether one or many cases at that place are met by these queries. To make the data fit several spatial analysis approaches, random points within every built area according to the numbers of visitors has been created with a GIS operation. Every point now represents a single respondent which had been to that area. 12 Neither the use of the Accommodation Mallorca system nor the use of internet search engines brought results regarding accommodation in that Sant Jordi. 13 See details in the attachments.

45


Map 10: Random location of respondents

5.2.2. Data Cleaning

Not all people answered the question for their place of accommodation in a satisfactory manner. Some did not do it at all. Since this study aims to study the spatial component in the LCC consumer profile, these cases have been removed. Due to the definition of tourism, permanent residents fall out of analysis, too. Permanent residents are those respondents who stayed in Mallorca for 360 or more days. Since the survey had been conducted on flights to and from Mallorca, it had to be ensured that people did not take part in the questionnaire twice. Those people were filtered by a question on the interview form of the return flight. Once identified, these respondents were removed from the sample. Finally 1696 valid cases from the passenger survey (921 on the flight to and 775 on the flight from Mallorca) and 1103 from face-to-face interviews were left for spatial analysis.

46

6. Spatial Tourist Consumer Analysis

47


6.1. General Spatial Distribution

As already stated, the tourism in Mallorca concentrates at the coast line (see chapter 4.2). The LCCP are no exceptions. All in all around 80% of the LCC passengers spend their holiday in an area of not more than 2.5 kilometres away from the coast. This area makes up less than a quarter of the island’s surface. The directional distribution (see map 11 below) points to the main centres of German tourism in Mallorca. More than one half of the LCCP concentrate at the Costa de Ponent with the main places Peguera and Sa Ponca and around the Bay of Palma with the main spots being Palma de Mallorca and S’Arenal. Another 16% travel to the coasts at the north eastern part from Sa Coma to Cala Rajada.

[_

Cos ta de Cos ta de Ponen tPonen t Pa l maPa l ma

ou t s i de ou t s i de tou r is t reg ionstou r is t reg ions


C o s t a d e T r a m u n t a n aC o s t a

de L

l ev a n t

C o s t a d

e Ll e

v a n tPl aya Pl aya de Pa l made Pa l ma

Nor thNor th

Sa Coma

PegueraS'Arenal

Cala Rajada

Santa Ponca

Respondents per Location(count of location)

1 - 11 (84)12 - 31 (21)39 - 82 (7)100 - 147 (3)235 (1)

Central Tendency

[_ LCC passengers mean centredirectional distribution



Backgroundtourist regionN o r t hN o r t h

municipality border 0 5 10 15Kilometers

Map 11: Visited places in sample of the LCCP survey 2005

The density equalising cartograms in map 12 show how extreme the concentration in some areas is. This kind of map pictures the areas in a size depending on a selected variable (UNIVERSITÄT ROSTOCK: PROFESSUR FÜR GEODÄSIE UND GEOINFORMATIK 2008). In this case areas on the different spatial levels are resized by the percentage of the LCCP they received. The coastal areas squeeze the central inland areas into small shapes that are


hardly recognisable. This distribution is dominated by the south and east coast areas. This phenomenon is visible for all levels of analysis, but on municipal and zonal level the north of Costa de Llevant is clear to see. Despite its size, the Tramuntana mountain region receives only a few visitors and is hard to see on the 3 maps below.

Palma

Costa de LlevantPlaya dePalma

Costa de Ponent

Badia d'Alcudia

outside tourist regions

Costa de Tramuntana

Badia de Pollenca

PalmaCalvia

Capdepera

Andratx

Santanyi

SollerArta

Alcudia

Manacor

Son Servera

FelanitxLlucmajor Campos

Pollenca

St L. des Cardassar

Ses Salines

Santa MargalidaMuro

DeiaBanyalbufar

Petra

Estellencs

Palma

Capdepera

S'Arenal

Costa de Ponent - South

Tramuntana

Santanyi

Serres de Llevant Bahia de ArtaCan

Pastilla

Cala d'OrCosta El Migjorn

Playa Sa Coma

West of Badia de Palma

- North

Badia d'Alcudia

Andratx / Calvia


Alcudia / Badia de Pollenca

a) Tourists Regions

c) Tourist Zones

b) Municipalities

Sources:Data based on Wachowiak 2005b.These Density Equalising Cartograms use the methodologydeveloped by Mark Newman and Michael Gastner at the Universityof Michigan.


LegendThe colours in the map are only for differentiation of areasbut do not represent numeric values.

original areas

Map 12: Density Equalising Cartograms for Mallorca’s LCCP tourism

This unequal distribution along the coast leads to the assumption that there must be reasons for the distribution other than just the closeness to the sea. Mallorca has a reputation of being a sun-and-sand destination. This indicates 2 factors that might not only attract the visitors to the island but also influence the distribution on the island. 71% of all LCCP stay in places with beach access. 45% of the LCCP stay in the two communities of Calvia and Palma, in the south of Mallorca, where the where the climate is dry and not too hot (for climate information see map 2 in chapter 4.1) and long beaches invite the visitor to take advantage of this climate. 40% of Mallorca tourist beds are located here. Another tourism hot spot is the Bay of Alcudia in the north of Mallorca. The climate here is hotter and the precipitation higher. At the long beaches in the communities of Alcudia, Muro, and Santa Margalida a fifth of Mallorca’s beds are accumulated. Only the LCCP are not attracted by it. Just 5.4% of them spend their holidays here.

48


% of LCCP30252015105

% of

beds

20181614121086420

Sources:based onInstitut Balear d' Estadística 2007Schürgers 2005Wachowiak 2005b


0 5 10 15Kilometers

coastal typeCoast with integrated small bays (Cala)Sand beach coast in basinSteep coast

Beds / meter coast(count of municipality)

0.0 (1)=< 0.05 (3)> 0.05 - 0.25 (3)> 0.25 - 0.40 (3)> 0.40 - 0.70 (3)> 0.70 - 1.30 (3)> 1.30 - 3.41 (2)

% of LCCP

% of LCCP(count of municipality)

1.2 - 1.6 (3)> 1.6 - 2.8 (3)> 2.8 - 3.2 (3)> 3.2 - 4.0 (3)

> 4.0 - 6.0 (3)

> 6.0 - 29.8 (3)

Calvia

Palma

Capdepera

Muro/Santa Margalida

Andratx

Alcudia

Sant Llorencde Cardassar

Map 13: Correlation of tourist beds and LCCP in Mallorca

Another fifth of the tourism beds are offered by the climatically similar municipalities of Sant Llorenc de Cardassar, Son Servera, and Capdepera at the north of the Costa de Llevant; even though the beaches are not as long as in the Bay of Palma or in the Bay of Alcudia. Around 19% of the LCCP stay in these places. In the hot inland of Mallorca and in the nearly beachless, wet and cold mountain region of Tramuntana tourism supply is very low. So is the demand of the LCCP. Although the correlation tests show that the LCCP are attracted by infrastructure in the form of overnight accommodations and gastronomy services (see table 3 below), they show a somewhat different pattern. In the Bay of Palma the focus is on the city of Palma with Playa de Palma on the east side. The number of LCCP in Calvia, west of the Bay of Palma, is low compared to the share of tourist beds in this municipality (see map 13). As both areas have the same climate and long and nice beaches, there must be more reasons for the decision

49


for a destination. The small amount of LCCP compared to the number of hotel beds in the north of Mallorca supports this point. However, it does not seem to be the quality of the accommodation that influences the distribution as the correlation tests indicate (see table 3 below). In both cases, the explanation could be the existence of another main visitor group. The community of Calvia is a main destination for the British tourist in Mallorca (cp. Map 6, chapter 4.2.1). Around a third of all British tourists in Mallorca come here (CONSELLER DE TURISME 2008). Magalluf, one of the tourist villages at the coast of Calvia, is to British travellers what S’Arenal is to the Germans: a beach resort for the low quality tourism segment and alcohol abusing young people (BUSWELL 1996, p. 313; SCHOLZ 2002, p. 241). Therefore, a lot of beds are blocked by British tour operators or are only distributed through British channels. The developed supply, mainly for British tourists, might not be appealing to a lot of German visitors. The municipalities of Alcudia, Muro and Santa Margalida along the bay of Alcudia with long and flat beaches are said to be ideal for families with children (FRÜNDT et al. 2006;TUI 2008). Since there are not many groups with children among the LCCP, a low share of visitors in the north of the island might not be surprising. % LCCP # beds # holiday homes # gastronomy

% LCCP 1 # beds 0.705(**) 1 # holiday homes 0.844(**) 0.614(**) 1 # gastronomy 0.901(**) 0.556(*) 0.931(**) 1 % high class accommodations 0.142 % middle class accommodations 0.299 % low class accommodations -0.249 (*) The correlation is on a level of 0.05 (2-tailed) significant. (**)The correlation is on a level of 0.01 (2-tailed) significant.

Table 3: Pearson’s R Correlation for selected variables on aggregated municipality level

Another advantage that made Mallorca a destination for mass tourism is its easy accessibility from central Europe by air. For the travellers, however, the journey does not end at the airport. The road connection to the final destination is part of the journey. The motorway network only connects the area around the Bay of Palma and Alcudia to the airport. The Tramuntana Mountains and the Costa de Llevant have no direct motorway access. Therefore, the journey from the airport to a destination in the northern part of the Costa de Llevant can take as long as the flight to Mallorca. For example, the road distance from the airport to Cala Rajada is around 80 kilometres. Most tourists go to areas with a good connection. A buffer analysis revealed that nearly half of all LCCP chose places to stay in an area within 2.5 km of the motorways. These are the areas close to the airport around the bay

50


of Palma as well as the well-connected places in the north around Alcudia (see map 14). While the cumulative number of visitors ascends only slowly with growing distance from the motorway, some places heavily attract visitors despite their remoteness. The tourist villages Cala Rajada, Cala Millor, and Sa Coma at the north end of the Costa de Llevant are far away from motorways yet are mainly responsible for the high numbers of visitors in the distance bands of 35 and 37.5 kilometres (see chart in map 14). Therefore, the factor accessibility is here outweighed by some other factors. That might be the good accommodation infrastructure and the cultural and natural attractiveness of these villages. The places combine a natural beauty with cultural sites (like the castle of Capdepera) and a wide variety of gastronomy and nightlife locations in settings that are less spoiled by architectural sins like those on Playa de Palma.

!.

!.

!.!.

!.!.

!.!.

!.

!.!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!. !.!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.!.

!.

!.

!.!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.!.

!.

!.

!. !.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.!.

!.

!.

l(

5

2.5510 km 252015 30

5

10 7.57.5

1512.5 22.5

35

32.5

17.520

40 km

37.5

Palma

SollerInca

Peguera

Alcudia

s'Arenal

Cala Rajada

Santanyi

Sources:Roads: IDEE 2007; ESRI 2003Statistics based on Wachowiak 2005b


MotorwayRespondents per Location (% off all)

!. < 1%!. 1% - 1.9%!. 2% - 4.9%!. 5% - 8.9%!. 9% - 14%

MiscellaneousVisited places (LCCP survey 2005)

l( Airport

0 5 10 15Kilometers

* LCCP accummulation based on spatial intersection of visited places and distance bands. Distance bands are 2.5 km each.

0102030405060708090

100

0 5 10 15 20 25 30 35 40 45Euclidean Distance from motorway in km

cummulative % of LCCP

!.

!.

!.!.

!.!.

!.!.

!.

!.!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!. !.!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.!.

!.

!.

!.!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.!.

!.

!.

!. !.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.

!.!.

!.

!.

l(

5

2.5510 km 252015 30

5

10 7.57.5

1512.5 22.5

35

32.5

17.520

40 km

37.5

Palma

SollerInca

Peguera

Alcudia

s'Arenal

Cala Rajada

Santanyi

Sources:Roads: IDEE 2007; ESRI 2003Statistics based on Wachowiak 2005b


MotorwayRespondents per Location (% off all)

!. < 1%!. 1% - 1.9%!. 2% - 4.9%!. 5% - 8.9%!. 9% - 14%

MiscellaneousVisited places (LCCP survey 2005)

l( Airport

0 5 10 15Kilometers

* LCCP accummulation based on spatial intersection of visited places and distance bands. Distance bands are 2.5 km each.

0102030405060708090

100

0 5 10 15 20 25 30 35 40 45Euclidean Distance from motorway in km

cummulative % of LCCP

Map 14: Distribution of German LCCP in relation to motorway connection

6.2. Spatial Distribution Compared to German Tourist in

General

The distribution of LCCP correlates with the distribution of tourist infrastructure on Mallorca but showed in some places huge deviations from the trend. In a similar way the deviations from the distribution of all German tourists occur. A remarkably lower percentage of all LCCP visits the north of Mallorca compared to German tourists in general. The Costa de Llevant again seems to attract a much bigger share of the LCCP than of German tourists in general (see map 15). Although the data for comparison are from the years 2002 and 2007, the stability of the results supports the significance of the differences (CONSELLER DE

51


TURISME 2003; CONSELLER DE TURISME 2008). Even the smaller difference in the area around the Bay of Palma is stable and thus trustworthy. On a municipality level, only categorised percentage values are available (CONSELLER DE TURISME 2007, p. 62). Therefore, conclusions from those comparisons have to be handled with care. Although, the huge differences of some figures suggest that mainly the two ends of the Costa de Llevant are responsible for that more of LCCP. Namely, these are the municipalities Santanyi with the vacation spots Cala d’Or and Cala Figuera, Son Servera with Cala Millor, and Capdepera with Cala Rajada.



C o s t a d

e Ll e

v a n t

P a l m a


t a d e e n t


$+5.4%

!(1.6%

!(1.6%

!(22.8%

!(16.1%!(4.3%

!(17.1%!(18.4%

!(19.7%

$+5.4%

$+32.6%

$+8.3%$+2.7%

$+15%

$+20.6%

$+15.4%

Sources:based on Wachowiak 2005b andConseller de Turisme 2008, p. 22


0 5 10 15Kilometers

Differences in % of total for German tourists (2007) and LCCP (2005)(count of region)

-7.9 (1)-4.4 to -2.0 (2)

2.1 to 4.0 (2)9.9 (1)LCCP % of totalGerman tourists % of total

-1.9 to 2.0 (1)


Map 15: Differences in the regional distribution of German LCCP and German tourists

Less LCCP in the areas close to the airport and motorway and many more LCCP in areas of Costa de Llevant without a proper motorway connection indicate that the LCCP are not as much influenced by accessibility in terms of travel time and travel distance as the German tourist in general. As the supply of tourist infrastructure, natural environment and other attractions is the same to both groups, the differences in distribution must be caused by different preferences between these groups or maybe even by a higher flexibility of the

52


LCCP. A closer view of the spatial characteristics of the LCCP consumer groups might help explain that phenomenon.

6.3. Distribution of Socio-Demographic Characteristics

As shown in chapter 3.1, the simplest approach to consumer analysis is to explore their socio-demographics. This information is also important to the more sophisticated approaches of tourist type classification and thus the starting point of any market segmentation. The average LCCP is 39 years old. On a regional level the variation between the different parts of Mallorca is small. Most areas show a mean age that does not differ much from the average. Only the Costa de Ponent and the area outside the tourist regions are a bit above the average. The visitors of Costa de Ponent are the oldest with a mean age of 41.8 years. Although their average is nearly one year above the area that comes second, the difference is not statistically significant due to the big error margins. Moreover, T-Tests revealed that the other regions’ mean ages are statistically seen significantly lower. The only exception is the Costa de Tramuntana (see table 4). The small sample size for this area causes a high error margin. Despite a confidence level of only 90% the average age of visitors ranges from 35.5 to 41.3 years. The age of the visitors to the Costa de Tramuntana cannot be compared to any of the other regions because the results are statistically not reliable on regional level. Region Age Valid

cases St. Dev. Margin of

error T Degrees of

Freedom Significant difference

Costa de Ponent 41.8 312 14.13 1.32 Outside tourist regions 40.9 87 13.77 2.43 0.54 397 No

Palma 39.7 230 12.70 1.38 1.80 540 Yes North 38.6 128 12.44 1.81 2.25 438 Yes Costa de Llevant 38.4 509 13.75 1.00 3.40 819 Yes Costa de Tramuntana 38.4 45 11.85 2.91 1.55 355 No

Playa de Palma 36.0 245 13.01 1.37 5.02 555 Yes Margin of error and significance have been calculated with a 90% confidence level.

Table 4: T-Tests for age means on regional level for the Costa de Ponent

The socio-demographic variables of the visitors to the Costa de Llevant seem to be on average (see map 16a). A closer look on municipal or zone level reveals the very heterogeneous structure of this area (see map 16b and 16c).The area with the oldest visitors is the central area of the Costa de Llevant. The visitors here are 44.8 years on average and thus, more than 5 years above the general mean. The initial assumption that this might have been a result of extreme outliers could be rejected by a simple look at the box plot of the age distribution in this area (see figure 9). 53


Figure 9: Box plot of the age distribution at the central Costa de Llevant

The zone with the very youngest visitors also is located at the Costa de Llevant. Capdepera, lying at the northern end of the Costa de Llevant, is visited by LCCP with a mean age of 34.6 years. The Tramuntana zone on the north east side of Mallorca has a slightly different shape than the region Costa de Tramuntana. With 38.5 years, the mean age is close to the overall average age. Because this is now based on 72 valid cases the Tramuntana zone is now better comparable to some other zones (see table 5). Nevertheless, significance tests show that only values with a big difference are statistically significant. Next to the old visitors to the central Costa de Llevant, the visitors to the eastern inland zone and the southern part of the Costa de Ponent are considerably older than the average LCCP. All in all, there are more areas with a mean age above the average than those with a significant lower age. The deviation is higher than for the areas with a mean age below the average.

54


Region Age Valid cases

St. Dev.

Margin of error

T Degrees of

Freedom

Significant difference

Central Costa de Llevant 44.8 51 13.18 3.04 2.80 121 Yes Costa de Ponent – South 43.0 161 14.24 1.85 2.37 231 Yes Bahia de Arta 41.1 66 14.27 2.89 1.17 136 No Costa de Ponent – North 39.8 61 12.54 2.64 0.63 131 No Tramuntana 38.5 72 11.68 2.26 Alcudia/ Badia de Pollenca 37.7 54 12.18 2.73 0.36 124 No

Playa de Palma/ S’Arenal 35.7 206 13.52 1.55 1.53 276 No Capdepera 34.6 169 13.97 1.77 2.06 239 Yes Margin of error and significance have been calculated with a 90% confidence level.

Table 5: T-Tests for age means for the Tramuntana zone

Because the phase of life is much related to age, Capdepera is the area with the highest share of people still in the apprentice or education phase of life. Over a quarter of the visitors are pupils, students, or job trainees. That is high above the average of 15.7%. Compared to the proportions of learning people in other zones the figures are seldom significant. Again, the small number of cases are the reason, e.g. for Santanyi (see table 6 below). The other zone with very young visitors, the Playa de Palma around S’Arenal, has a significant lower proportion of learning people than Capdepera.

Region % of people in

the training phase of life

Valid cases

2-Tail actual confidence level

Significant difference to Capdepera

Significant difference to

Santanyi

Capdepera 28.8 % 170 -5% - No Santanyi 28.6 % 42 -15% No - Andratx/Calvia 22.6 % 31 36% / -6% No No Serres de Llevant 20.3 % 69 80% / 54% No No Playa de Palma/ S’Arenal 19.6 % 199 94% / 65% Yes No Palma 14.2 % 218 100% / 96% Yes Yes Costa de Ponent – North 13.9 % 65 97% / 87% Yes No Costa de Ponent – South 8.8 % 160 100% / 100% Yes Yes Central Costa de Llevant 6.5 % 46 100% / 98% Yes Yes

Table 6: Z-Test results for comparison of proportions of visitors in the training phase of life (pupils,

students, apprentices) for the Capdepera and Santanyi zone (selection)

Another palpable difference between these 2 places is the relation of the genders. While in general more women came with LCC to Mallorca (52.7%) at the Playa de Palma male passengers dominate with a proportion of 53.2%. Capdepera on the other hand has one of the highest ratios of female visitors (60%). It shares the top position with Playa de Sa Coma

55


and Santanyi at the southern end of the Costa de Llevant and the Badia d’Alcudia in the north of Mallorca. With the available data no correlation could be found that supports a reasonable explanation for the spatial differences in the socio-demographics of the visitors. From a visual analysis of the distribution, it seems that the young people are attracted by the areas where the nightlife is notorious and the beaches are large. Older LCCP seem to prefer areas with more cultural supply and intact historical villages. The differences in socio-demographics, and here especially of age, are probably caused by different preferences people have in different phases of their life. For example, VADELL found out that golf tourists on Mallorca are mainly middle aged between 46 and 60 (2005, p. 30). The survey at hand did not ask for golf activities. But the golf tourists are individual tourists, mostly avoiding package tours (cp. STENAU 2009, p. 162/180) and thus, are similar in parts of their profile to the LCCP. Assuming that there are several golf tourists among the LCCP, they might have raised the average age in the region of the Costa de Ponent where a high density of golf courses is to be found. In 2005, 6 of Mallorca’s 15 places together with 8 courses have been in or close this region (CONSELLER DE TURISME 2006p. 130). This is only one factor and it does not help explain why, along the Costa de Llevant, the visitor groups differ as much in age as they do. That might be caused by the different supply. The beaches along the central Costa de Llevant are small and difficult to access. But for older LCCP the visit to the beach might not be the centre of their holiday because they cannot stand the whole day in the hot sun. So they look for additional attractions like beautiful landscape, natural attractions and nice, tranquil villages. All this is offered between Portocristo and the Cales des Mallorca. That might be one reason for the high average age in this area. There might be other clues for the differences in the visitor groups in the characteristics of the LCCP that will be investigated in the following chapters.

56



C o s t a d

e Ll e

v a n t

P a l m a





I n l a n d

A r t a

L l u c m a j o r

T r a m u n t a n a

M a n a c o r

C a m p o s

P o l l e n c a

P a l m a

C a l v i a

F e l a n i t x

S a n t a n y i

A n d r a t x

A l c u d i a

M u r o / S a n t a M a r g a l i d a

C a p d e p e r aS o n S e r v e r a

S e s S a l i n e s

S t L . d . C a r d a s s a r

Tr a m u n t a n a

I n l a n d - E a s tI n l a n d - W e s t

S e r r e s d e L l e v a n t

P a l m a

S a n t a n y i

C o s t a E l M i g j o r n

B a d i a d ' A l c u d i a

C a p d e p e r a

A n d r a t x / C a l v i a

A l c u d i a / B a d i a d e P o l l e n c a

C a l a d ' O r

c e n t r a l C o s t a d e L l e v a n t

B a h i a d e A r t a

B a d i a d e P a l m a - W e s t

- S o u t h- N o r t h

P l a y a d e P a l m a / S ' A r e n a l

C a n P a s t i l l a

P l a y a S a C o m aC o s t a d e P o n e n t

C o s t a d e P o n e n t Sources:based on Wachowiak 2005b

n = 1544

n = 1597

Working Life Status for all LCCP

Distribution of Sexes for all LCCP

Only those areas with at least one category next to work within fault tolerance are displayed.

Age (Mean)Mean 39.0, n = 1556

40.1 - 42.042.1 - 44.044.1 - 46.0

Working Life Status (%)

learning (pupils, students, apprentices)retiredworking

Gender (%)

malefemale

36%

a) Tourist Regions b) Municipalities (partly aggregated)

c) Tourist Zones

0 5 10 15Kilometers

Map 16: Socio-demographic characteristics of LCCP in Mallorca

36.1 - 38.034.6 - 36.0

38.1 - 40.0

miscellaneous


6.4. Trip-Related Behaviour

6.4.1. Booking and Accommodation

In the past decades the proportion of package tourists who book transportation, meals and accommodations arranged by one tour operator has been remarkable high. Although their share dropped just below 50% recently (CONSELLER DE TURISME 2008, p. 62), Mallorca is still a destination, which is very important for tour operators and travel agents. The decreasing share in Mallorca’s tourism might be caused by the rising number of LCC in the air traffic to Mallorca. Only 16% of the LCCP choose this form of travel organisation. Package tourists go to the regions of Playa de Palma, Costa de Llevant and the north of Mallorca. Nevertheless, there are big spatial variations visible on zonal level. A real package tour destination is Playa de Sa Coma belonging to the municipality of Sant Llorenc de Cardassar (see map 17c). 59% of LCCP that came here were booked on a package tour. Although this place only receives 3.3% of all LCCP, it has a share of 11.8% of all package tourists arriving by LCC. The other hot spots of package tourism for the LCCP are the Costa de Llevant north of Playa de Sa Coma around Capdepera and along the whole northern coast. All these places have a fraction of between 20% and 30% of package tourists. In the southern and western part of Mallorca, only Playa de Palma has a share of package tourists that is significantly above the average. All together, the named areas accommodate 62.1% of all package tourists coming by LCC while receiving only 37.3% of the German LCCP. These huge differences are probably caused by the supply of package tours. The forenamed areas are known for a big share of tour operator owned or blocked hotels. Another possible reason for the high concentration in Sa Coma might be a special offer that perhaps has been advertised with Germanwings tickets. Since a package tour includes amongst other services the lodging, the places with a high portion of package tourists are those with high rates of LCCP staying in rented accommodation. Playa de Sa Coma and Playa de Palma together with Can Pastilla are the extreme with over 80% paying for their own lodging. These can either be hotels, hostels or pensions as they are the dominant accommodation types for the LCCP in Playa de Palma or holiday homes or fincas like in the hinterland or in the Tramuntana Mountains (see map 17a). Residential tourism is one of the market segments supported by the marketing strategy of the government of the Balearic Islands. Residential tourism means people do not have to rent an accommodation because they either own a holiday home or apartment or they can stay in one owned by friends or relatives. One third of the LCCP stay in such a non-rented accommodations. On municipal level, with a result of 0.849, Pearson’s correlation coefficient indicates a strong, statistically significant correlation between the number of secondary homes (INE 2004) and the number of LCCP staying in non-rented accommodations. The 58


municipality Arta in the north east of Mallorca receives the highest proportion of residential LCC tourists. The high share of people visiting a home of friends or relatives in combination with a very small number of cases, however, leads to the conclusion that this might be a distortion caused by the only 25 valid cases. Other communities with more valid cases and a proportion of more than 50% residential tourists among the LCCP are Llucmajor, Andratx, and the aggregated community area of the hinterland. On zonal level, there are also 4 areas where over 50% of the LCCP stay in non-rented properties. 3 of them do not have access to the coast. One of the areas comprises the villages of Calvia and Andratx (nearly 4 of 5 people here stay in self- or friend- or relative-owned houses or apartments) and happens to be one of the areas with no package tourists. The only coastal zone with non-rented accommodations as the dominant form of lodging is the north of the Costa de Ponent, an area with rugged coastline and only a few small coves with sandy beaches. This area at the Costa de Ponent is an area with one of the highest shares of holiday homes and the top destination for residential tourists among the LCCP. After the figures of the 2002 census in Spain, 27% of all secondary homes of Mallorca are located in the municipalities of Andratx and Calvia (INE 2004). These municipalities are the top destinations for home owners amongst the LCCP. 36.7% of them come here. In addition, more than a quarter of the people staying in properties of friends or relatives come here as well.

59



C o s t a d

e Ll e

v a n t

P a l m a





I n l a n d

A r t a

L l u c m a j o r

T r a m u n t a n a

M a n a c o r

C a m p o s

P o l l e n c a

P a l m a

C a l v i a

F e l a n i t x

S a n t a n y i

A n d r a t x

A l c u d i a



S e s S a l i n e s


Tr a m u n t a n a



P a l m a

S a n t a n y i



C a p d e p e r a



C a l a d ' O r







P l a y a S a C o m aC o s t a d e P o n e n t


% of LCCP booking Package Tours16.4% for Mallorca; n = 1696

0,00>= 1.8 - 10.0> 10.0 - 20.0> 20.0 - 30.0> 30.0 - 40.0> 40.0 - 50.0> 50.0 - 58.9

a) Tourist Regions b) Municipalities (partly aggregated)

c) Tourist ZonesType of Accommodation of LCCP in Mallorcan = 1620

rented accommodation 66.4%

non-rented accommodation 33.6%

Source:based on Wachowiak 2005b

0 5 10 15Kilometers

Map 17: Travel behaviour of LCCP in Mallorca: package tourism and accommodation

Type of Accommodation (%)

Type of Accommodation (%)rented accommodationproprty of friends/relativesown property

Type of Accommodation (%)

rented accommodationnon-rented accommodation

hotel/hostel/pensionholiday home/ fincaholiday clubproperty of friends/relativesown poperty


6.4.2. Sojourn Time

The sojourn time is an important indicator for the tourism structure and tourist revenue in a destination. The range in the sojourn time amongst the LCCP is very wide with some outliers at the upper end (see box plot in map 18). These visitors stay for more than 2 months, mostly as residential tourists. To account for this wide variation the sojourn time is analysed based on the median and a classification of the length of stay. Can Pastilla is the top short time visitor destination with a median duration of stay of 4 days and over 50% of the short time visitors staying less than 5 days. The other zones around the Bay of Palma, the northern part of Costa de Ponent and the hinterland have disproportionately high rates of short time visitors amongst the LCCP. The eastern part of Mallorca’s hinterland is ranked third in this category, with 39.5%. At a first glance, that might look like an error perhaps caused by the small absolute numbers of respondents from this zone. However, it can be explained by the good road connection to this area via a motorway that is well developed from Palma through Inca up to Alcudia. A short distance or a good connection to the airport reduces the travel time to the different places on the island and makes them more suitable for short visits. Map 18c shows that the zones with the lowest median for sojourn time are intersected by the motorway. On the regional and municipality level this phenomenon is not as concise but still visible. As seen in chapter 6.1, the connection to the motorway network does not influence the distribution of the LCCP beyond a certain point. Nevertheless, it certainly influences the travel behaviour in terms of the length of stay und thus, the distribution of short trip and long trip visitors.

61

r(



C o s t a d

e Ll e

v a n t

P a l m a





r(

I n l a n d

A r t a

L l u c m a j o r

T r a m u n t a n a

M a n a c o r

C a m p o s

P o l l e n c a

P a l m a

C a l v i a

F e l a n i t x

S a n t a n y i

A n d r a t x

A l c u d i a



S e s S a l i n e s


r(

Tr a m u n t a n a



P a l m a

S a n t a n y i



C a p d e p e r a



C a l a d ' O r







P l a y a S a C o m aCos ta de Ponen t

Boxplot sojourn time

Hinge 3.00

days

1901801701601501401301201101009080706050403020100

a) Tourist Regions

c) Tourist Zones

b) Municipalities (partly aggregated)

Median Sojourn TimeMedian = 7, n = 1601

456

88,5912

Sojourn Time (%)

5 - 7 days8 - 14 days> 14 days

Mobilitymotor wayhighway dual carriage

r( airport

27%< 5 days


Sojourn Time for all LCCP

n = 1601

7

0 5 10 15Kilometers

Map 18: Travel behaviour of LCCP in Mallorca: sojourn time

Only those areas with at least one category within fault tolerance are displayed.


6.4.3. Travel Company

Making a decision for a holiday destination depends highly on the group of travellers. Families with children have other needs than groups of friends or elderly couples without children. The supply offered in a certain place has to be adjusted to the main target groups of visitors. 39% of the LCCP come with their partner to Mallorca and 34% with friends. Despite the cheap tickets, Mallorca does not seem to be a family destination for the LCCP. Only around a quarter of them travel with their families to Mallorca. Families are an important target group. But families are said to not get along well with other tourist groups like couples or travelling friends (WELT ONLINE 2007). Disproportionately high are the portions of travelling families at the Costa de Ponent and in the northern region of Mallorca. Here, the families make up around a third of the visitors.



C o s t a d

e Ll e

v a n t

P a l m a





Palma

Tramuntana

LlucmajorManacorCalvia

Pollenca

Santanyi

Andratx

Alcudia

a) Tourist Regions

b) Municipalities (partly aggregated)

Travel Company for all LCCP

n = 1688


Travel company (%)

28friendspartnerfamilymiscellaneous

0 5 10 15Kilometers

Map 19: Travel behaviour of LCCP in Mallorca: travel company

On zonal level the areas along the Costa de Llevant mostly have the shares of families above average, except Capdepera. Cala d’Or is top with 36% while in the other zones

63


between 30% and 34% of the visitors come with family. The accumulation of high values here and the zones with families above the average at the Costa de Ponent produce signs of clustering (cp. Map 20). A test for spatial autocorrelation proved that these patterns are not randomly generated. The strength of the clustering, however, is only weak (see table 7) although even there is a cluster of low percentage of families at Playa de Palma (14%) and Palma city (15%). These are the zones where people prefer to come with friends. Around the Bay of Palma that applies to all zones except Can Pastilla. Other spots dominated by travelling groups of friends are the south and north end of the Costa de Llevant. The central zone of this coast including Playa de Sa Coma is the exact opposite. While the other zones along this coast have a share of friend-accompanied visitors between 40% and 50% the central part has only 20% of those visitors. In contrast to that 40% travel with their partner to the central Costa de Llevant and Playa de Sa Coma. The 6 zones with the highest percentage of LCCP travelling with friends are at the same time those with the lowest proportions of travelling partners. Next to the central Costa de Llevant, couples are the dominant form of travel groups in the Tramuntana Mountains, in the north of Mallorca, and in Can Pastilla as a kind of couple enclave at the Bay of Palma. Although all these listed areas have high shares of travelling couples, they only account for 23% of all couples amongst the LCCP. A quarter of the visitors travelling with a partner go to Palma or Playa de Palma although they are outnumbered there by people coming with their friends. Travel company Moran’s I Z-Score Distribution

Partner -0.12 -0.82 Random Friends -0.11 -0.68 Random Family 0.24 3.24 Likely to be clustered (99%) Miscellaneous 0.05 1.18 Random For the calculation of the indexes the spatial relationship has been conceptualised as zone of indifference with a threshold distance of 3,000 meters (Euclidean Distance)

Table 7: Tests for spatial autocorrelation for the proportion of travel company on zonal level

Different travel groups have distinct demands, as were already mentioned above for families. Although hard facts are missing to prove it, the different distribution can be assumed, based on the knowledge about the supply in different areas around Mallorca. Groups of friends dominate in the zones where relaxing, beach activities and going out at night are the focus of the supply (Playa de Palma and Capdepera). Travelling partners often seem to prefer more peaceful areas like the Tramuntana Mountains (younger couples) and the central Costa de Llevant (older couples). The differences in age between these partner areas might be caused by the more challenging landscape and the colder and unsteady weather in the mountains. Much family accommodations are in the northern part of Mallorca, at the long beaches of the Bay of Alcudia. The water on these beaches is shallow and the tourist infrastructure is newer

64


than in the south. The area is promoted as a family destination by travel books and tour operators alike (cp. e.g. FERRER 2007, p. 21/40; FRÜNDT et al. 2006, p. 62/108; TUI 2008, p. 59). That applies to the zones Cala d’Or and Santanyi too. Nevertheless, the families amongst the LCCP do not occupy these places, which are mostly recommended for and adjusted to family holidays, in a significant amount. Although the zones in the north have a share of visiting families above the average, most families among the LCCP travel to the Costa de Ponent. That many families come here, although other areas are more recommended for family holidays, might be caused partly by the more expensive hotels in the north. The newer apartment hotels in the north are to a larger extent from the upper end of the star scale (see map 3, chapter 4.2.1). For the price-conscious LCCP family the bigger choice of low and middle class accommodations in combination with a lot of available leisure time activities (Marineland, Aqualand etc.) might seem to be more attractive. In addition, many owners of a holiday home come with family (over 50%; based on WACHOWIAK 2005b). Costa de Ponent has a high density of German holiday homes and is a cluster for LCCP in this tourism sector. This accommodation structure might also contribute to the high number of families in this area that is not explicitly recommended for families.

65

Tramuntana

Inland - East

Inland - West

Serres de Llevant

Palma

Santanyi

Costa El Migjorn

Badia d'Alcudia


Alcudia/ Badia de Pollenca

Cala d'Or


Bahia de Arta

Costa de Ponent Playa de Palma/

s'Arenal

Playa Sa Coma


Tramuntana

Inland - East

Inland - West

Serres de Llevant

Palma

Santanyi

Costa El Migjorn

Badia d'Alcudia



Cala d'Or


Bahia de Arta


s'Arenal

Playa Sa Coma

Tramuntana

Inland - East

Inland - West

Serres de Llevant

Palma

Santanyi

Costa El Migjorn

Badia d'Alcudia



Cala d'Or


Bahia de Arta


s'Arenal

Playa Sa Coma

a) Partner (Mallorca average 39.7%)

c) Family (Mallorca average 25.2%)

b) Friends (Mallorca average 34.9%))


Tramuntana

Inland - East

Inland - West

Serres de Llevant

Palma

Santanyi

Costa El Migjorn

Badia d'Alcudia



Cala d'Or


Bahia de Arta


s'Arenal

Playa Sa Coma

c) Miscellaneous (Mallorca average 7.8%)

Accompanied by friends (%)17.74 - 22.5022.51 - 27.5027.51 - 32.50

32.51 - 37.5037.51 - 42.5042.51 - 47.5047.51 - 51.28

Misc. travel company (%)3.57 - 5.005.01 - 10.0010.01 - 15.00

Accompanied by partner (%)

Map 20: Travel company in Mallorca's tourist zones

Accompanied by family (%)13.57 - 17.5017.51 - 22.5022.51 - 27.50

27.51 - 32.5032.51 - 37.5037.51 - 38.89

23.44 - 27.5027.51 - 32.5032.51 - 37.50

37.51 - 42.5042.51 - 47.5047.51 - 51.11

0 5 10 15Kilometers


6.5. Expectations and Preferences

The knowledge about the customers’ wants and needs is important to the tourism industry and for tourism marketing. They cannot be readily concluded from socio-demographic characteristics or from travel behaviour. That is why in the poll at hand people on their way to Mallorca have been questioned about their expectations. They were asked to rate the importance of a list of activities. The marks range from 1 for “very important” to 5 for “very unimportant” for the activities shown in table 8. Activity n % 1 % 2 % 3 % 4 % 5

Average rating

Relaxing 687 51.2 29.0 10.9 4.2 4.7 1.82 Enjoying landscape and nature 664 42.3 36.0 10.8 5.0 5.9 1.96 Visiting typical Mallorcan gastronomy 656 31.1 38.6 17.2 7.6 5.5 2.18 Going out in the evenings, meeting people (nightlife) 646 32.5 31.3 20.6 9.8 5.9 2.25

Making trips or excursions, exploring the neighbourhood 635 28.0 35.1 20.3 7.6 9.0 2.34

Experiencing local culture and way of living 640 14.2 39.5 25.0 11.9 9.4 2.63 Lying often on the beach; bathing 661 18.9 22.8 23.8 14.5 20.0 2.94 Contact to local people 636 14.6 21.1 32.4 17.8 14.2 2.96 Cultural activities (e.g. museums, concerts, exhibitions, visiting historical monuments) 625 12.0 23.5 27.2 19.8 17.4 3.07

Shopping for typical Mallorcan products 612 5.6 23.0 31.4 24.0 16.0 3.22 Shopping (in general) 628 6.2 22.1 32.0 20.4 19.3 3.24 Doing sports 622 9.0 15.8 20.9 19.3 35.0 3.56

Table 8: Activities the LCCP on a flight to Mallorca rated for their importance for this Mallorca trip

In addition, all passengers on their way to and from Mallorca were requested to check on a list what they especially like in Mallorca. The lists were slightly different on the both ways but have the most important amenities of Mallorca in common (see table 9). Together, both questions result in a huge amount of data. Therefore, the research focussed on the most telling information. In order to do so, some variables have been disregarded for further research. As criterion for disregarding variables, the coefficient of variation for the areas at the different geographic levels has been chosen. Those variables with a very low coefficient have not been further investigated.

67


Amenity of Mallorca Percentage

Landscape / nature 67.5% Accessibility 62.4% Beaches 49.9% Food 36.7% People 29.6% Price (value)(MCADAM 1999) / low priced destination 29.6% Culture 22.4% Language 18.8% Leisure time facilities 14.9% n = 1696

Table 9: What LCCP especially like of Mallorca

Because of the lower number of absolute cases in the questionings on expectations, an analysis on a municipality or zonal level is not possible for the entire island. The numbers of cases for most of these areas are below what is necessary for statistically-reliable results. That is why most of the analysis on the wants and needs of the LCCP refers to the regional level. Moreover, even on this level the count of valid cases is too low in the Tramuntana region for most of the activities. In map 21 the numbers of valid cases for beach activities are displayed. Even less people answered the questions for most of the other activities. For some examples a zoom to smaller areas is made for high frequented zones. The question on the distribution of what people consider to like especially in Mallorca can be discussed even on a zonal level.

1.00

1.50

2.00

2.50

3.00

3.50

4.00

4.50

5.00

relaxing

enjoying

landsc

ape an

d nature

visit typ

ical ga

stronom

y

nightlife

making e

xcursio

ns

experie

nce loc

al cultu

re

beach a

ctivities

meeting

locals

cultura

l activit

ies

shoppin

g Mallo

rca pro

ducts

shoppin

g gene

ral

sports

Figure 10: Activity rating of LCCP

The most important activity to all LCCP is relaxing. It is, however, not everywhere the central focus of the holiday. The LCCP visiting the northern part of the island (and in Tramuntana, 68


but without statistical reliability) rate it as most important and the highest compared to the other regions. But the people outside the tourism regions rate it with 2.13 far below the average. They rate the visiting of typical gastronomy, enjoying the nature, and making day trips higher. In combination with the highest rating for sports amongst all regions the latter mirrors the unrest of the LCCP in the interior of Mallorca. One reason for their active behaviour might be the fact that over half of the visitors outside the tourism regions names the beaches as an amenity of Mallorca they especially like. To get there, they have to travel. Like in every other region besides the Playa de Palma, most people here name the landscape and nature as a special amenity of Mallorca. The highest percentage of LCCP liking the landscape and nature can be found in the Tramuntana mountain range and in the northern region of Mallorca. More than 3 quarters of the visitors in these regions name it as what they especially like in Mallorca. In the same time the highest importance is given to the enjoyment of the nature by these visitor groups. In general, the percentage of people liking the landscape and nature of Mallorca is high and so they rate the enjoying of it very high. The LCCP visiting the urban area of Palma and the mass tourism spot at the Playa de la Palma from S’Arenal to Can Pastilla are much less fond of nature. Only 60% of the visitors to Palma and even less than a half of the visitors to the S’Arenal region like very much the nature and landscape of Mallorca. Especially the visitors to S’Arenal express this, giving by far the lowest rating for enjoying the nature. All areas, including Palma with the second lowest mark for joy of nature, are in the range between 1.7 and 2. S’Arenal visitors rate it only 2.5 on average. In this region it is all much more about the beach and the nightlife. The ratings for it deviate nearly as much from those of the other regions like for the nature, only in the opposite direction. The visitors to Playa de Palma value the beach as much more important, with a mean mark of 2.3, than the average LCCP with 2.9. They also have the highest percentage of people liking especially the beach of Mallorca. The distance to the second place, the visitors to the Costa de Llevant, is marginal but it is significant compared to the other regions (see map 22). The people at the Costa Llevant like the beaches in Mallorca, but they do not attach great importance to it for their holiday in Mallorca. Regarding the importance of nightlife, the visitors to the Playa de Palma are only matched by visitors to Palma. The other regions are behind (compare map 21). Going out at night and meeting people is in the area of the Playa de Palma even more important than relaxing. For both nightlife and beach activities, Playa de Palma is even on zonal level (where Can Pastilla is not part of this area) unrivalled by other regions.

69


11.25

1.51.75

22.25

2.52.75

33.25

3.53.75

relaxing enjoy thelandscape and

nature

visit typicalgastronomy

going out in theevenings(nightlife)

makingexcursion

experiencelocal culture

beach activities

avera

ge ra

ting

very important

unimportant

Northoutside

tourist regions

Costa de Llevant

Palma

Costa de Tramuntana

Costa de Ponent

Playa de Palma

32

202

5915

127

12532

101

Numbers show the countof valid cases for the rating of beach activities.

11.25

1.51.75

22.25

2.52.75

33.25

3.53.75

relaxing enjoy thelandscape and

nature

visit typicalgastronomy

going out in theevenings(nightlife)

makingexcursion

experiencelocal culture

beach activities

avera

ge ra

ting

very important

unimportant

Northoutside

tourist regions

Costa de Llevant

Palma

Costa de Tramuntana

Costa de Ponent

Playa de Palma

32

202

5915

127

12532

101

Numbers show the countof valid cases for the rating of beach activities.

Map 21: Regional activity rating of German LCCP in Mallorca

Compared to the coastal regions in the south east, the north western region of the Tramuntana Mountains attracts rarely people who favour the beaches of Mallorca. Only a quarter of the LCCP name them as special amenities of Mallorca. The interest in cultural activities is everywhere much lower than the interest in most other activities. And less than a quarter consider culture a special amenity of Mallorca. On regional level, the north receives a disproportionately high ratio of people that especially like the culture of Mallorca. Despite a difference of over 7%, that is not even statistically significant different from the lowest value at the Playa de Palma, where only 17.6% like the culture. On zonal level the range in the proportion of people with a special liking for Mallorca’s culture is much larger. The highest and lowest values differ by 39.3%. Next to Cala d’Or, where only 2 of 39 people checked culture on the list of Mallorca’s amenities, the Playa de Palma around S’Arenal has a very low ratio of culture liking people. Only 14.9% ticked this. This is significantly less than in the zones of the southern Costa de Ponent and Andratx/ Calvia (see

70


Table 10), that are close by. The latter is the area with the most culturally interested LCCP in Mallorca.

Zone % liking especially the

culture of Mallorca Valid cases

2-Tail actual confidence

level

Significant difference

Playa de Palma/ S’ Arenal 14.9% 221 Costa de Ponent – North 20.6% 68 67.5% No Costa de Ponent – South 22.1% 181 90.7% Yes Tramuntana 24.4% 78 89.7% No … … … … … Santanyi 33.3% 48 99.4% Yes Andratx/ Calvia 44.4% 36 100% Yes

Table 10: Z-Test results for the zone of Playa de Palma/ S’Arenal for the comparison of proportion of

LCCP naming culture as a special amenity of Mallorca they like

In the list of Mallorca’s amenities just above the culture come the low prices in Mallorca. Although the visitors travel with a LCC to Mallorca and thus, are likely to have paid only a low air fare only 29.6% consider Mallorca’s cost-performance ratio as a special amenity. The only region really above this average is Playa de Palma. On zonal level the both zones constituting this region (Can Pastilla and S’Arenal) are considerably high above the average with around 40% of people appreciating the prices in Mallorca.

71



C o s t a d

e Ll e

v a n t

P a l m a





Inland

Arta

Palma

Tramuntana

LlucmajorManacor

Calvia

Felanitx

Campos

Pollenca

Santanyi

Andratx

Alcudia

Capdepera

Son Servera

Tramuntana

Inland - East

Inland - West

Serres de Llevant

Palma

Santanyi

Costa El Migjorn

Badia d'Alcudia



Cala d'Or


Bahia de Arta


s'Arenal

Playa Sa Coma


What LCCP especially like of Mallorca (% of all LCCP)

n = 1696

0 5 10 15Kilometers

What LCCP especially like of Mallorca (%)

landscape and naturebeachescultureprice/value

43%

Map 22: What LCCP especially like on Mallorca


6.6. Spatial Consumer Profiles

The spatial tourist consumer analysis of LCCP in Mallorca showed that the tourism spaces occupied by them are located mainly at the coast. In absolute numbers, their favourite region is without a doubt Costa de Llevant. But even relatively, in comparison to the distribution of other tourists, they favour this region. In the region of Costa de Llevant the zone of Capdepera is the main destination where the German LCCP travel to. Here, they are overrepresented in relation to other tourists and in relation to tourist beds. Nevertheless, together with the southern part of the Costa de Ponent, Capdepera comes only third in the list of favourite zones after Palma de Mallorca and the Playa de Palma around S’Arenal. The analysis of the different consumer characteristics discovered, within these and other zones, quite different consumer groups, which are precisely described and named in the following.

6.6.1. Regional Consumer Profiles

On a regional basis there is the easy to distinguish group of visitors to the Playa de Palma, from Can Pastilla to S’Arenal. It is the group of “the young and quick nocturnal sun-and-sand

tourist”. These mostly young and middle-aged LCCP, with disproportionately many men amongst them, tend to have short stays and come either with their partner or in a group of friends. While families do not especially favour this region most of the few package tourists

among the LCCP come here. They prefer to stay in hotels and hostels, which might owe to the structure of accommodation supply in this early developed tourism region. The LCCP visiting Mallorca give an impression of where the reputation of Mallorca as a sun-and-sand destination might come from. The S’Arenal visitors consider the nightlife as most important

and rate it together with beach activities much more important than visitors to any other region. On the other hand, they are not very interested in Mallorca’s nature or culture and are not particularly very active or mobile. So these people, uninterested in the cultural and natural environment of their destination, are simply satisfied by the supply of nice beaches and nightlife facilities, which exist in abundance along the Playa de Palma. People with similar attitudes but with much less preferences for beach activities and a bit more cultural and shopping interests visit the Palma region, often for very short trips of only up to 4 days. It is a kind of “short-

term urban sun tourist”. Contrasting to these visitors at the Bay of Palma, there are the visitors with the highest preferences for natural and cultural experiences. These

73


nature lovers have also a strong desire for relaxing and consider the nightlife as unimportant. These people travel to the Tramuntana Mountains or the north of Mallorca. The mountains with the steep and stony coast are the destination for those who do not see the beaches as a special amenity of Mallorca. They are “the relaxed nature lovers”. Those who also like the beach go to the northern region of the island with its mixture of rugged coastline and long, sandy beaches in the bays of Pollenca and Alcudia. They are a kind of “relaxing, nature liking sun-and-sand tourists”. The group of LCCP visiting the area outside the tourism regions is again very different. This huge, mostly flat area is targeted by the agro-tourism program of the local government of the Balearic Islands, as a part of its diversification efforts. The visitors here, nearly 2 years older than the average LCCP, prefer lodging in holiday homes or staying in their own property or

that of friends/relatives. They are interested in the local culture and people and are very active: They value enjoying the environment and making excursions as the second most important activities close after the eating in typical Mallorcan restaurants. They even rate doing sports higher than the people in every other region and consider relaxing by far

least important. It is “the older, locally interested active finca14 tourist”.

The visitors to the other regions are somewhere in between these other consumer groups. The visitors to the Costa de Ponent, staying preferably in their own property or that of a friend/relative, are the oldest and rate beach activities less important than the visitors to any other region. They could be called “the old residential tourist”.

The Costa de Llevant receives masses of tourists that influence and mirror the average of the LCCP at the same time. On this level of spatial aggregation this group of visitors represents “the average LCCP tourist”.

14 The Spanish word “Finca” formerly meant an agricultural estate with a farm house. Nowadays, in the tourism business, it is used as synonym for a rural holiday home in Spain (WIKIPEDIA).

74


6.6.2. Municipal Consumer Profile

A detailed analysis of the consumer profile for each of the 53 of Mallorca’s municipalities was not possible because of low respondent numbers for several communities. This fact caused the aggregation of some municipalities to bigger units. The profiles of these areas, which are similar in shape to the area outside tourist regions and the Tramuntana Mountain region, show no significant shifts. Caused by the shape of the municipality of Palma it comprises most of the visitors that made up the regions of Palma and Playa de Palma. Therefore, this profile is a combination of both of them. The visitors to the municipality of Palma are comparatively young, tend to make short visits, stay predominantly in hotels, and appreciate as much as no other region the nightlife of Mallorca. Their likings of beach and culture are indifferent while nature does not play an important role in their travel decision. Therefore, we have here on municipality level “the short-termed, nocturnal tourist”.

The region known as Costa de Llevant receives tourists that represent nearly the average LCCP visiting the island. The region comprises 8 different municipalities. A differentiated examination on municipal level reveals that this visitor group is composed of some very heterogeneous visitor groups in the single municipalities. Noteworthy is the extreme young visitor group at the north end of the Costa de Llevant, in Capdepera. Many of them are still in

the learning phase of life and have comparatively little Mallorca experience. Only 3 of 4 have already been to Mallorca, which is one of the lowest values of all zones. But despite their age they do not differ much in their likings and expectations from the average LCCP in Mallorca. They mostly stay around a week and spend their holidays rather

with friends than with partner or family. They are “the young friends on sunny vacations”. Another revelation is produced by the division of the Costa de Ponent in the 2 municipalities of Andratx and Calvia. Calvia, where the tourist villages Peguera and Santa Ponsa are located, receives very old visitors. They are on average 3 years older than the visitors to the neighbouring

community Andratx (statistically significant within a 90% confidence interval). In most other characteristics these visitor groups coincide. Only the sojourn time is different. While the people in Andratx stay either very short or between 8 and 14 days the behaviour of the visitors to Calvia is indifferent. The tendency to longer stays in

75


Andratx might be caused by the higher ratio of people staying in their own property or make holiday in a property of a friend or relative.

6.6.3. Zonal Consumer Profile

The customised zones have a different shape and size than the tourism regions and municipalities. Besides the Tramuntana zone they are all smaller. In some cases the differences are big, and thus are the differences to the profiles. In some cases, however, the differences are small. From the region along the Playa de Palma around S’Arenal, for example, only Can Pastilla was separated. The customer profile stayed nearly the same, but is, in some aspects, more distinctive. Amongst the visitors to Can Pastilla is a higher proportion that appreciates the landscape and nature of Mallorca as well as the culture as amenities of Mallorca. Therefore, these mostly couples of LCCP choose a place a bit away from the party zone around S’Arenal. Still, they stick to the Playa de Palma because very many of them like the beaches and the accessibility to Mallorca that is very high in this vicinity to the airport. This visitor group has the highest proportion of short-time travellers. They are a kind of “stop-over-at-the-beach visitors”.

As the visitors to Can Pastilla are not part of the LCCP group at the zone of Playa de Palma/ S’Arenal, the profile of beach life and especially young people favouring nightlife and disregarding nature and culture around S’Arenal is even further strengthened. It is the extreme form of “the young

and quick nocturnal sun-and-sand tourist”. In a similar way, the profile of the LCCP travelling to the mountain area of the Tramuntana is confirmed. The borders for this zone are drawn a bit wider including municipalities at the intersection to the hinterland. The results are now based on a higher number of respondents (78) what makes them more stable. They show a profile of people highly favouring relaxation and loving to explore the nature and are very uninterested in nightlife. They consider beaches neither important nor a special amenity of Mallorca but favour the food in typical Mallorcan restaurants. They are “the relaxing, nature-loving gourmets” under the LCCP. Very detailed results regarding the visitor profiles reveal the fine scaled zoning of the Costa de Llevant. While the visitors to the region named Costa de Llevant did not show any

76


extreme values in their tourist profile, that picture of an average tourist breaks up into very different visitor groups along the Costa de Llevant:

The very young visitor group at Capdepera does not change as the shape is that of the municipality of Capdepera. It is still “the average young

friends on vacations”. Compared to the other zones along the Costa de Llevant they rate the nature lower and nightlife higher. But in contrast to the other young group of visitors in Playa de Palma they still consider

enjoying the nature more important than nightlife. In their personal preferences nightlife comes third after relaxing and enjoying the nature. At the central zone of Costa de Llevant the oldest visitor group is located. Compared to the young bunch at Capdepera these people prefer longer sojourn times, travel with their partners and less than a half of them see the beach as a special amenity of Mallorca. But like the youngest visitor group, they do not deviate much from the average in their attitudes. It is “the old couple from next door on holiday”. The visitors to Playa de Sa Coma compose of another distinct visitor group. Amongst these

predominantly pairs of package tourists more people especially like the beaches of Mallorca than in most of the other zones. They consider the beach activities important for their Mallorca holiday. But enjoying the nature and relaxing are even more important to them. At the same time they rate cultural activities and shopping higher than the LCCP in any

other zone. Even all the other activities are more important to them as to the average LCCP. And most of them try to squeeze all these activities into one week or less of holidays. This I-want-it-all mentality can be best summarised naming this LCCP consumer groups the “the

all-interested, twosome package tourist”. In contrast to these all-interested package tourists, on the other side of the central Costa de Llevant meet culturally ignorant, highly immobile tourists that do expect nothing from their Mallorca trip but relaxation. Therefore, this “relaxed beach-lying tourists” come to Cala d’Or in groups of friends or with their families.

77


6.7. Differences to German Tourists in General

The German LCCP and the German visitors in Mallorca have the same favourite spots on the island but show deviations in their distribution. While the LCCP especially frequent less the northern regions of Mallorca and a bit less the regions Palma and Playa de Palma, they favour the Costa de Llevant remarkably more than the German tourists in general. Because of flaws in the spatial sampling the data of the face-to-face questionnaire cannot be used to compare the distribution of the German visitor and the LCCP. But they can be used, taken as independent sample for every place, to compare the characteristics of both consumer groups for particular zones. The zones that had more than 100 respondents in both polls are compared.

6.7.1. S’Arenal / Playa de Palma

The area around S’Arenal is the number one destination for Germans in Mallorca. It has developed in the early years of mass tourism and is characterised by sky scraping hotels and a high-density of buildings. Here the profile of the common German tourists is characterised by much less learning and more retired people than are to be found within the LCCP visitor group. The German tourist in general is on average significantly older than the LCCP (nearly 9 years). The ratio male to female visitor is also different between the LCCP and the German tourists in general. While 56% of the LCCP around S’Arenal are male 56.5% of the common German tourists are female in this area. It is not only the socio-demographic characteristics that are different (see table 11 below). The LCCP show significantly different travel behaviour. They stay for shorter periods and less often in hotels. Less than a quarter of them book package tours in contrast to the 3 quarters of German tourists in general. In contrast to the LCCP, the common tourist considers the experience of local lifestyle and cultural activities significantly more important than the LCCP. More of them especially like the people and culture of Mallorca. The discrepancy between the LCCP and the common German tourists is even larger, regarding their attitudes towards the nature and landscape of Mallorca.

78


Characteristic German tourists in general German LCCP

Socio-demography

Mean age in years 35.7 44.4 % male 53.2 % 43.5 % % Learning visitors 13.4 % 17.7 % % Retired visitors 21.6 % 7.2 % Travel behaviour

% Package tourists 75.3 % 23.5 % % renting accommodation 99.3 % 86.1 % % not renting accommodation 0.4 % 12.6 % Median sojourn time 7 5 Diagram sojourn classes

0

5

10

15

20

2530

35

40

45

50

< 5 days 5 - 7 days 8 - 14 days > 14 dayssojourn time

%

German touristsLCCP

Dominant travel company in % Friends (59.8 %) Friends (44.3 %) Expectations and preferences

Top 3 amenities of Mallorca (% of answers) 1. beach (69 %) 2. landscape (60.9 %) 3. The people (42.1%)

1. beach (56.1 %) 2. landscape (42.1 %) 3. food (30.3 %)

Ratings: - Experience local culture 2.93 3.43 - Enjoying the landscape and nature 2.02 2.62 - Relaxing 1.59 1.97 n 271 221 Sources: Based on (WACHOWIAK 2005a; WACHOWIAK 2005b)

Table 11: Significantly different characteristics of German tourists and German LCCP at the Playa de

Palma/S’Arenal

6.7.2. Costa de Ponent (South)

The southern part of the Costa de Ponent comprises the tourist areas of Peguera and Santa Ponca. This area is a highly urbanised, densely built-up area, visited mostly by German and British tourists (FRÜNDT et al. 2006, p. 280). This area hosts next to hotels and pensions several “urbanisacions”, which is the Spanish name for tourist apartment villages. According to SCHMITT (2007, p. 23), the surrounding areas, the municipalities of Calvia and Andratx, have each a proportion of more than 60 % of the homes for tourism purposes. That indicates the importance of residential tourism as a business factor at the Costa de Ponent. But that fact is only mirrored by the 37.5% of LCCP who do stay in their own property or in one of a friend/relative at the southern part of the Costa de Ponent. The German tourists in general

79


stay mostly in hotels. Only 7.7% have their own property or stay in that of friends or relatives. Further differences in travel behaviour are the shorter sojourn time of the LCCP and a very small proportion of package tourists amongst them (see table 12). The LCCP are on average 6 years younger than the common German tourist in this area. This goes along with a remarkably small percentage of retirees among the LCCP (8.8%). Even if the 30% of retirees among the German tourists from the face-to-face survey might be above the year-round-average, it is still a notably high proportion. While both groups name the landscape and the beaches as what they like most of Mallorca, these amenities are much more important to the common German tourist. Relaxing is a very important holiday activity for them. The LCCP consider relaxing and enjoying the nature important but less than the German tourist in general. In contrast, they rate activities like going out for eating, and going out at night more important.

80


Characteristic German tourists in general German LCCP

Socio-demography

Mean age in years 49.0 years 43.0 years % Retired visitors 30.2 % 8.8 % Travel behaviour

% Package tourists 79.7 % 12.7 % % renting accommodation 90.6 % 58.6 % % not renting accommodation 7.7 % 37.5 % Median sojourn time 10 7 Diagram sojourn classes

0

10

20

30

40

50

60

70


German touristsLCCP

%

Expectations and preferences

Top 3 amenities of Mallorca (% of answers) 1. Landscape (85.7 %) 2. Beach (66.5%) 3. The people (33.0 %)

1. Landscape (72.4 %) 2. Beach (42.5 %) 3. Food (37.0 %)

Ratings: - Beach activities 2.45 3.22 - Nightlife 2.80 2.40 - Visit typical gastronomy 2.40 2.16 - Enjoying the landscape and nature 1.53 1.93 - Relaxing 1.58 1.94 n 182 181 Sources: Based on (WACHOWIAK 2005a; WACHOWIAK 2005b)

Table 12: Significantly different characteristics of German tourists and German LCCP at the Costa de

Ponent (South)

6.7.3. Capdepera

Capdepera is an old town with a historic castle set on a hill not far from the sea in the north east of Mallorca. The municipality named after the town comprises the tourist village Cala Rajada that once had been a fishing village and is now a favourite spot for German tourists, especially the LCCP (FRÜNDT et al. 2006, 164ff). This area is visited by very young LCCP, on average 6 years younger than the common German tourists. Only a few of them come with package tours and therefore stay not as much in rented accommodations as the German visitors in general. The differences in sojourn time are less pronounced where the LCCP have a higher percentage in the short trips and a smaller percentage in the sojourn time class of 8 to 14 days (see table 13). Both groups especially like the landscape, nature and beaches of Mallorca and spend their 81


holidays mostly with friends. Beach activities and enjoying the landscape are important for the LCCP, but are rated lower than by the German visitors in general. Instead, they give higher marks to experiencing the local culture. Characteristic German tourists in general German LCCP

Socio-demography

Mean age in years 40.3 years 34.6 years Travel behaviour

% Package tourists 79.6 % 22.9 % % renting accommodation 91.3 % 71.1 % Median sojourn time 7 7 Diagram sojourn classes

0

10

20

30

40

50

60


German touristsLCCP

%

Expectations and preferences

Top 3 amenities of Mallorca (% of answers) 1. Landscape (75.2 %) 2. Beach (71.5 %) 3. Food (32.1 %)

1. Landscape (68.7 %) 2. Beach (60.3 %) 3. The people (35.2 %)

Ratings: - Beach activities 2.29 2.63 - Experience local culture 2.97 2.56 - Enjoying the landscape and nature 1.75 2.03 n 137 179 Sources: Based on (WACHOWIAK 2005a; WACHOWIAK 2005b)

Table 13: Significantly different characteristics of German tourists and German LCCP at Capdepera

6.7.4. Summary

The choice of air carrier is not the only distinctive characteristic between the German LCCP and the German visitors in general. As has been seen in the 3 zones compared above, some characteristics are different. In addition, the figures from other sources indicate that those characteristics are different in general (cp. e.g. CITTIB 2003a; CONSELLER DE TURISME 2006; CONSELLER DE TURISME 2008; WACHOWIAK 2005a). These most telling differences are the age, the percentage of package tourists, the choice of accommodation, and the sojourn time. Compared to the German tourists in general, among the LCCP are more young people and less older people (see figure 11).

82


0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

under 30 30 to 45 46 to 60 over 60 age

LCCP 2005

German tourists 2002All tourists 2002

Source: based on Wachowiak 2005b (n = 1763); CITTIB 2003a, p. 1; CITTIB 2003b, p. 1 Figure 11: Age of tourists in Mallorca

Generally, the LCCP seldom book package tours compared to all visitors to Mallorca. In 2002 60.7% of all and even 64% of German tourists booked a package tour for their visit to Mallorca (CITTIB 2003a, p. 1; CITTIB 2003b, p. 1). Despite decreasing numbers of package holiday makers in Mallorca they still make up nearly a half of all visitors to Mallorca (CONSELLER DE TURISME 2008, p. 62). The LCCP prefer to arrange their holiday themselves and thus book flight, accommodation, and other parts of the trip separately on their own. Only 16% book packaged tours. The low percentage of package tourists comes along with a disproportionately high share of residential tourists and a comparatively low number of hotel guests. More than one third of the LCCP (34.1%) do not need to book any accommodation because they stay either in their own home or in the house or apartment of friends or relatives. Only 48% stay in a hotel or hostel. The usual proportion of hotels in the accommodation market for German tourists is around 70%. A fifth of them stay in their own properties or of friends or relatives. That is approximately the same as for all other tourists with only slight shifts between own properties and those of friends and relatives (see figure 12). The cheap air-fares and the regular connection to the European mainland make it much easier for owners of holiday homes and their friends to visit the property more often and in flexible time spans. These convenient aspects are likely to be responsible for the high share of people not staying in rented accommodations among the LCCP.

83


0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

hotel/hostel own property property of friends/relatives miscellaneous

LCCP 2005German tourists 2002All tourists 2002All tourists 2007

Source: based on Wachowiak 2005b (n = 1620); CITTIB 2003a, p. 1; CITTIB 2003b, p. 1; Conseller de Turisme 2008, p. 62 Figure 12: Type of accommodation of Mallorca tourists

The LCCP tend to stay for short periods. With an average of 8.7 days (biased by some extreme outliers) and a median of 7 days the sojourn time of LCCP is very low compared to the average of 11 days of all Mallorca visitors (CONSELLER DE TURISME 2006).The low average and median is caused by high numbers of short trips. 28.5% of the LCCP leave Mallorca at the latest after 4 days. The reason for the significant differences between LCCP and other German tourists is probably the product of LCC travel itself. The direct distribution over the internet demands technical affinity what more often is to be found among young people. The often missing cooperation of LCC with tour operators asks for flexible people booking all parts of the travel themselves. Therefore, they are more flexible in their planning and can even stay only a few days. Package tourists are bound to the length of offered packages, mostly either 1 or 2 weeks long. Especially for residential tourists this flexibility is welcome as they do not need to rent an accommodation. The differences between the LCCP and the German tourists in general in their attitudes are not that pronounced and differ between the regions. The comparison of the different areas shows that the general German tourist apparently likes the nature and beaches of Mallorca and rates them nearly in every researched area more important than the LCCP. The LCCP consider the nightlife frequently more important than the common tourists. Nevertheless, although there can be a preference for party and nightlife assumed amongst the LCCP, they do not visit the area around the Bay of Palma, which is famous for its nightlife, as much as the tourists in general. They are disproportionately highly represented in some areas at the

84


Costa de Llevant, especially Cala Rajada in Capdepera. This tourist village has a lot of activities to offer during the night but in addition provides natural beauty and cultural sites.

85

7. Summary and Conclusions

86

7. Summary and Conclusions This study implemented a spatially enhanced consumer research. That way the distribution patterns of German LCCP and the distinction between the spatially-defined visitor groups of LCCP in Mallorca could be discovered. In addition, the reasons of these patterns have been searched for and for the distinction to the common German tourists visiting Mallorca. The results will be summarised and critically reviewed in this concluding chapter and thus answering the questions asked in chapter 1.115.

7.1. Spatiality of German Low-Cost Carrier Passenger on

Mallorca

The spatial consumer analysis of LCCP in Mallorca showed that the tourism spaces occupied by the LCCP are located mainly at the coast. Their favourite region is the Costa de Llevant. Even relatively in comparison to other tourists they favour this region. Within this region, Capdepera is the area most favoured by German LCCP. In relation to other tourists and in relation to tourist beds they are overrepresented. On zonal level Capdepera comes together with the southern part of the Costa de Ponent only third in the list of favourite zones after Palma de Mallorca and the Playa de Palma around S’Arenal. The analysis of the different consumer characteristics discovered within these and other zones very different consumer groups (see map 23).

consumed tourism spaces

Comparisons to other tourists on Mallorca, especially German tourists, showed that the LCCP are different in their characteristics and their spatial behaviour. They prefer other places than the common German visitor, especially along the Costa de Llevant. Even if the profiles at certain places are not very distinct it could be shown that the LCCP are younger and more flexible than the German tourist in general.

differentiation of LCCP from other tourists

The discovered distinct consumer profiles of the LCCP of course triggering the questions for the reason of these differences. The decision for a certain holiday destination is a complex process. It is determined by so called push and pull factors (FEILMAYR 2007, p. 48; FREYER 1998, p. 38). Push factors are intrapersonal (motivation/reasons) and interpersonal factors (travel companion) and cannot be influenced by destination management (SILVA & CORREIA 2008). Nevertheless, the knowledge and understanding of them can help for marketing purposes. That applies also to the structural factors such as money and time limiting the travel choice. A destination offers a supply that is recognised by the consumer as pull factors attracting him to this destination. It is a combination of the marketing of a destination (price,

factors influencing the distribution of LCCP

15 Regarding the research questions answered in this chapter, see the side texts for additional guidance.


distribution, tourist/trip-planning information) and its local attractions and services (mainly in terms of accommodation, gastronomy, cultural sites, and natural environment) (BELLAND & BOSS 2008).



C o s t a d

e Ll e

v a n t

C o s t a d

e Ll e

v a n t

Costa de Ponent:

- North

- South

The beach-lyer

Tramuntana

Serres de Llevant

Palma

Santanyi

Costa El Migjorn

Badia d'Alcudia

Capdepera

Andratx / Calvia

Badia de Pollenca

Cala d'Or

Bahia de Arta

Playa de Palma / s'Arenal

Playa Sa Coma

LCCP % of total(count of zones)

=< 2.5 (4)2.5 to < 3.0 (4)3.0 to < 3.5 (2)3.5 to < 4.0 (2)4.0 to < 5.0 (4)10.0 to 15 (4)

Sources:based on Wachowiak 2005b created by

Johannes Luberichs, 20080 5 10 15

Kilometers

The young and quick, nocturnal sun-and-sand tourist

The relaxing, nature-loving gourmet

The all-interestedtwosome

package tourist

The old couple from next door on holiday

The average young friends on vacation

Map 23: Distribution of German LCCP and some selected spatial profiles of them in tourism

zones on Mallorca

Some of the factors influencing the overall distribution of the German LCCP in Mallorca could be discovered by spatial queries, buffer analysis and dynamic mapping in combination with summarising statistics:

factors influencing the distribution of LCCP

• The number of tourist accommodations available (hotels, beds, holiday homes), • the number of gastronomy available, • the road accessibility, • the closeness to the coast and access to beach.

The first named factors showed even linear relationships to the number of LCCP per municipality and correlated significantly positive with it. Regression analyses suggested that the gastronomy is the most influencing factor of them16. That might partly be true.

16 See regression analyses results in the attachment.

87


The masses of tourists are in the spots where nightlife is highly favoured. Nightlife needs a lot of gastronomy establishments like restaurant, bars, and clubs to be considered good and vibrant by the most people. Since the nightlife is important to many LCCP, they will follow the reputation of the bar- and club-scene. But on the other side, the connection might likely be the other way round. A huge amount of tourists will attract businessmen and entrepreneurs. When there are enough customers a new establishment might open. It is, however, highly unlikely that the opening of a new establishment will attract a certain amount of new visitors like the linear regression function suggests. A main reason why the multiple regression function did not deliver reliable results is that all the variables are correlated with each other. In tourist centres around the island the accommodations agglomerate together with the holiday homes. The correlation is weak but exists because the demand for the natural factors like beautiful landscape or beaches is similar to both groups. And where there are accommodations, there is, of course, gastronomy in an amount according to the number of accommodations. The role of the distance to the transport network or the coast and beaches could not be expressed in terms of such relationships. They speak by their mapped results and the sheer amount of LCCP close to the motorways or to the sea. Even these methods, however, showed that there are pull factors beyond the named ones. For example, the observed relation of distance to road (as an indicator for travel distance and accessibility) and the amount of LCCP is inverse. The number is high close to the motorway, low in a middle distance, and increases again in a further distance. The explanation will not be that a lot of people like the remoteness so much. For that they are too unevenly distributed over the badly-connected areas. Other pull factors are here stronger than the travel time. That is why the named are some, but by far not all the factors influencing the distribution of LCCP. They are of course part of an explanation for the distribution of tourism in general. That means the LCCP follow roughly the pattern of tourism in Mallorca. But they show significant deviations from the distribution of German tourists in general. Although not backed up by figures, one reason might be the higher flexibility of the LCCP. The German tourists in general come mainly through package tours. The tour operators supply their beds not evenly around the island but concentrate in certain areas. The German visitors mirror this spatial pattern by high numbers in the Playa de Palma region and in the north of Mallorca and low numbers in Calvia, where mostly British tourists reside. The LCCP can choose freely their place of stay and thus are able to adjust more easily their destination according to their taste. Another factor influencing the different distribution is probably the higher proportion of LCCP staying in

88


their own property or in one of friends or relatives. That raises the numbers in areas with many holiday homes like at the Costa de Ponent and to some extent at the Costa de Llevant (see chapter 6.4.1). The huge difference in the north of Mallorca can probably be explained by the fact that especially the bay of Alcudia is known as a family destination. LCC travelling is not a domain of families as the low percentage of accompanying children shows. In addition, it is a package tourist area with a comparably high share of 4- and 5-star accommodations and is quite far away from the airport, what makes it difficult to access, especially for shorter trips and for people with a lower budget. Even the families among the LCCP do not travel to the north very much, perhaps for the same reasons (see chapter 6.4.3). The families cluster in the southeast and in the southwest. That they are especially numerous in the southwest around the Costa de Ponent might be caused a combination of cheap hotels, a lot of available leisure time activities in theme parks and facilities alike, and the high numbers of holiday homes that are used by families. The reasons for the different spatial profiles of German LCCP could be discovered for some characteristics and had to be assumed for others. Short time visitors are influenced by the accessibility. The residential tourists follow the distribution of holiday homes. The reasons for the disparity in the socio-demographic composition of the LCCP consumer groups at the different places are not that easy to see. They might be caused by different supply and costs in the different areas. Amongst other factors, the preferences for certain holiday attractions are tightly connected to the socio-demographic characteristics of a person. Young people are likely to have less money and less travel experiences. The interest for culture and cultural activities might be lower than those for beach, sports and nightlife. The young consumer groups in places like Playa de Palma and in Capdepera support this assumption. In more peaceful places, where the beach is not the main focus of the tourism supply, the people are older. This view is supported by the findings of RUTOW & WACHOWIAK (2009, p. 16) who discovered that the cultural interested tourist on Mallorca is around 42 on average and travel predominantly with their partners. But it is only true for the hinterland and the central Costa de Llevant. Although Palma with all its old buildings, theatres and museums can be considered as the cultural centre of the island, the LCCP staying there are younger and not interested that much in cultural activities. Since it is in day trip distance from all other destinations, it can be visited from any place on the island. Studies prove that it is by far the most popular day trip destination for German tourists in Mallorca (cp. NEUMANN & WACHOWIAK 2009, p. 130). Therefore, it is not necessary to stay in Palma during the holidays.

89


That the consumer groups in the different areas have distinct expectations and likings is probably related to the distinct supply at the different places. The visitor will try to adjust his travel destination as good as possible to his likings. Especially the image of a place influences his travel decision. Studies show that especially the knowledge of a destination from former visits shapes this image to a great extent (cp. HAAK 2009, p. 45f). 80% of the visitors had already visited the island (derived from WACHOWIAK 2005b). Since the LCCP are seldom bound to the restrictions of package tours, they are free to choose their destination as they like. This is reflected, but not provable by numbers, in the profiles of visitor groups and the matching supply and characteristics of a place: People liking the nature and landscape are highly represented in the rough, natural Tramuntana area, people who like the beaches are mostly at the coast, and at the places of extended nightlife people happen to rate this as very important. In a similar way the supply probably determines the average age of a consumer group, its gender balance or the travel company. But those factors seem to influence each other, too. In the areas with more young people a lot of groups of friends are around and the areas with older visitor groups are dominated by travelling couples, for example. All in all it can be stated that a lot of factors influence the composition of a consumer group at certain places. Especially the quality of the supply (natural as well as man-made) is sometimes difficult to measure. Things like the public image of a place – what is very important in tourism (cp. e.g. HAAK 2009) – belong to this category as well. Therefore, some of the results are still speculative ideas. These ideas however are the result of the spatially enhanced tourist consumer analysis and would not have been generated without it.

7.2. Spatial Enhancement of Tourist Consumer Research

One of the main aims of this thesis is to identify the tourist spaces that are consumed by German LCCP. That triggers right away the question how to define the boundaries of these spaces and what might be the right spatial scale of analysis (cp. Chapter 1.1). In this work the boundaries were set on 4 different spatial levels (cp. Chapter 5.1.3). This approach makes working with the data elaborate because there is 4 times the preparatory and manipulative work to do on the datasets. In addition, it complicates the reporting and mapping of the results. But to make use of as much available data as possible to set the LCCP distribution into context of their environment, detect phenomena and try to explain them, a multi-scale approach was indispensable. In addition, working on different scales with different shaped spatial units helps to show

appropriate spatial scale of analysis

90


the effects of the MAUP and minimise their influences on the interpretation of the results (cp. Chapter 3.3.2). In 2 cases the boundaries of the tourism spaces were adopted from other authorities that report statistics about Mallorca: the tourist regions and the municipalities. Those arbitrary areas are not always optimal for analysis. The big regions can only show an overview and obscure some patterns because of the high level of aggregation. Some indexes for comparison, however, have been only available on regional level. The municipalities are ill shaped for tourism analysis because they reach far into the hinterland where the landscape and the structure of tourism are much different than at the coast. Therefore, they may distort the results to a certain extent. Unfortunately most of the statistics for economics and infrastructure are only available on municipality level. Several relations between the spatial pattern of the German LCCP and the tourist supply would not have been discovered without working on municipality level.

boundaries of tourism

For this study the raw data of the surveys have been available (like it will be in most destination based tourist consumer studies) with a spatial resolution that theoretically made analysis on village level possible. The daisymetric character of this spatial data and the fact that it has been a sample, made summarising of the single cases to groups in research-area-covering, aggregated spatial units for most analyses necessary to account for the missing places. This aggregation is a common way in social sciences to interpolate the sample data. The availability of the sample data made a customised zone design possible. Following the idea of OPENSHAW & ALVANIDIS (1996; cp. Chapter 3.3.2) this opportunity was used to apply the aggregation as a tool for analysis. The boundaries of the tourism spaces were mainly set by the characteristics of the visiting LCCP (see chapter 5.1.3). Using local measures of spatial autocorrelation as well as landscape information and tourist supply information the zone design was made manually. That unfortunately exacerbates the traceability and reproducibility of this method. Algorithms, however, have not been finally developed for sample data that are not yet interpolated. To do so would have exceeded the scope of this work but might be an interesting proposal for future research. The advantage of customised spatially aggregated units is that the self-designed zones accommodate heterogeneous groups of tourists and thus meet the requirements of tourist consumer research very well (cp. chapter 3.1). Additionally, they enable a fine-scaled approach that serves the longing for high resolution conclusions. In combination with the other levels of analysis they account for the MAUP and minimise the risk of ecological fallacy in result interpretation.

91


So a multi-level approach is appropriate for tourist consumer research and the boundaries of tourism spaces are to be defined depending on the available data. A customised zone design is appropriate. But zoning has to account for units of statistical reporting, too. Spatial analysis has a tradition of some decades. But still scientists cannot agree on a commonly accepted definition (O'SULLIVAN & UNWIN 2003, p. 2f). Since there is no final definition, there is not a final set of methods that are part of spatial analysis. Therefore, this study cannot say exactly which methods of spatial analysis were left out. But it could show which methods it used to analyse the spatiality of the German LCCP in Mallorca and why.

methods of analysis

First of all, the way of spatial data conceptualisation is to be mentioned. The LCCP are single objects with several attributes and different behaviour. So they cannot be interpolated like continuous data and thus cannot be represented by a surface. The representation as an object in point or vector format is the appropriate perspective for tourist consumer analysis as it is the most appropriate perspective for social science in general (FISCHER 2006). In the majority of cases in social sciences, these objects are geographically represented by spatial units that aggregate many individuals to a group. The work with such zone-based data bears some problems; either in interpretation (MAUP and ecological fallacy) or in calculation (edge effects). Kwan (KWAN 2000) adds that the aggregate spatial data framework limits the number of appropriate applicable methods. For example, point-based methods for fine-scaled analysis cannot be applied. Just to convert areas to be represented by points, as mentioned by Fischer (FISCHER 2006) does not change the problem that the values of the individuals are aggregated and thus an analysis of their individual behaviour and interaction is not possible. Analysis on the level of individuals or households is of course only interesting to micro-geographic analysis. Although for this study the information on the individuals have been available, such a fine-scaled approach has been infeasible. The location of the cases has been on village level. One village can, of course, accommodate many respondents. Therefore, an aggregation to a spatial unit is already given by the survey design. But in this form of aggregation the data do not cover the whole research area. The locations have only a small extent and since the study works with a sample, not all locations on Mallorca are covered by the survey. With this daisymetric map only a limited set of analysis were possible. As the sample was representative, the directional distribution could be calculated and show the main spots of German LCCP tourism. For other

92


analysis like spatial queries, buffer or overlay analysis ad-hoc-areas were built where the data have been summarised to. For comparative analysis of areas and to account for the missing places in the consumer analysis process the information of the visitors have to be interpolated somehow to cover the whole research area. This was done by aggregating the individuals to spatial units. The group of LCCP tourist within a unit are represented by descriptive statistics. This concept has its forenamed weaknesses. But since this work could access the data of the individuals the choice of the zoning was free (see chapter 5.1.3 and paragraph above). Therefore, a multi-level approach has been chosen (see above). That way the conceptualisation of the data offers a wide range of methods to analyse the distribution of the German LCCP and their characteristics in Mallorca. First of all, homogenous groups of tourists were accommodated in a customised zone design. These groups and their characteristics were than compared to each other and to that of groups aggregated to spatial units that are commonly used for statistical reporting (regions and municipalities). For that the data have been manipulated in GIS, mapped in different ways and dynamically linked to other graphical representations of the data like box plots or scatter plots. All these tools can be summarised under the method of geovisualisation as part of the geospatial analysis. They contributed to the insight into the distribution of the German LCCP. That would not have worked without the assistance of classical statistical methods. There were the descriptive statistics that described the characteristics of the tourist groups in the spatial units. Estimation statistics had to show the error margins of means or proportions. For comparison of different units test statistics had to be calculated to verify whether a numerical distinction of 2 areas are really significantly different in the statistical sense. Correlation coefficients have been calculated to verify relations discovered in maps or graphs. Even spatial statistics have been calculated, like the directional distribution or the spatial autocorrelation of the visitors in the zoning process (cp. Chapter 5.1.3). Measures of spatial autocorrelation are called the workhorses of social sciences (SWEENEY & FESER 2004) because they are heavily deployed in social sciences (as manifold demonstrated in the book of GOODCHILD & JANELLE 2004a). In this study they have not been applied for pattern identification or cluster detection for the aggregated spatial units because the clusters of similar values were basis for the zoning process. Because of that, for the customised zones, patterns and clusters are already incorporated in the zones. In addition, for all levels of analysis the number of zones was below the number of what is said to deliver reliable results. The very irregular size of the different zones would have contributed to questionable results.

93


The process of analysis can be summarised as follows: The data were stored, manipulated and visualised in a GIS environment. Techniques of geovisualisation supported by descriptive statistics showed the distribution of German LCCP and their characteristics and delivered ideas of differences, relations and causes. Traditional statistical testing confirmed discovered differences. All these methods were descriptive and exploratory in nature. They helped describe and confirm the distribution of the German LCCP consumer groups in Mallorca. However, they cannot explain causes. To do so, a good domain knowledge and cautious interpretation of discovered phenomena are necessary. It can be supported by classical statistical regression methods. Because of the lack of confirmatory instruments GOODCHILD & JANELLE (2004b) call spatial analysis in general an exploratory technique that can generate insight and hypotheses (cp. as well GATRELL 2008; OPENSHAW.S. 1991). This study confirms this view as the tourist consumer research would not have shown any reliable results without classical statistical methods. So for spatial tourist consumer research a solid basis in classical statistical theory and methods is necessary. Therefore, it cannot be spoken of pure spatial analysis here. Especially since some sophisticated methods of spatial analysis are not fully applicable, it is more a spatially aware analysis and thus, an extension to classical consumer research rather than a replacement. This spatial enhancement of the consumer research process has produced an insight in the profiles of the LCCP that would not have been accessible without spatial information and analysis. The characteristics of the LCCP consumer groups have been organised in tables. They were described by descriptive statistics and compared with the assistance of statistical tests. One could argue against geospatial analysis because in consumer research many instruments are available for consumer group definition and analysis. Nevertheless, the geospatial consumer analysis shown in this thesis produced added value for tourist consumer research. Firstly, the consumers in this case were clustered by spatial characteristics. In that way differences between the consumer groups around the island could be detected. In addition, the geocoding of the visitors made it possible to set them into relation to environmental information that helped discovering new phenomena and explanation for distribution pattern. Those have been the distinction of the distribution of LCCP and German tourists in general, the different characteristics between LCCP consumer groups and the relations to road access, coast access, or the supply of infrastructure. All that would not have been seen without the geospatial extension of tourist consumer research. Here is to mention that this is a complete different approach to consumer analysis. A comparison of classical statistical cluster

visitors’ characteristics discovered by geospatial analysis

94


analysis and spatial aggregation of consumer groups would be interesting for future research. In order to analyse the spatial behaviour and profile within spatially enhanced consumer research two main kinds of data are mandatory: data about the tourists including the spatial information asked for by the research purpose and data about the tourism spaces within the research area. In a GIS environment, the latter can then be used to set the information derived from the tourist data into a spatial context and gain new insight. To bring the data from different sources and with different content together, a spatial reference to be useable in a geospatial analysis context is needed.

data requirements

Most of the additional statistics that were used to find relationships and explain phenomena were available on municipality level, some on regional level. Since there were only 7 regions, correlation or regression calculations did not work on regional level. For the fine scaled zonal analysis information about the tourism supply has not been available at all. Therefore, the explanations that are based on statistical calculations are only valid for the municipality level. Conclusions from this spatial scale to the zonal level are not valid (see ecological fallacy, chapter 3.3.2). Since the municipalities are not the best areal units for tourism consumer analysis, as remarked earlier, the analysis is somewhat incomplete. For the completeness of such analysis the additional data beyond the survey should be available for the smallest spatial unit of analysis. This study could discover statistical relationships of the LCCP distribution and accessibility and the tourist infrastructure (accommodation, gastronomy). To some extent even the natural attractions (coast, beaches) could be incorporated into explanatory approach. But tourism (a holiday trip) is a special product that the consumer cannot touch, look at or compare directly. The tourist supply of a destination is composed of attractions (natural and artificial), tourist infrastructure and accessibility (FREYER 1998). The composition of these three factors builds together with the reputation the image of a destination (FREYER 1998). This image is dependent on the quality of the supply. That is why that qualitative information would be necessary for a more complete approach to explain detected phenomena. Those data have not been available and are very difficult to collect. Therefore, the explanations for discovered phenomena are based on assumption to some extent. Through the self-administered interviews, the poll used in this thesis achieved a sample size that could not have been achieved in most other destination polls. Still, it appears to be too small for several analysis levels and techniques. On municipality level some areas had too few or even no respondents. Within the spatial units subgroups

95


frequently had too few cases to make statistically sound statements. For fine-scale analysis even the almost 1,700 interviews of this survey did not suffice for all analysis techniques. For consumer analysis a sufficient sample size is necessary that is determined by the research question, the scale of analysis derived from it and the distribution of the tourists within the research area. Spatially enhanced tourist consumer research will need frequently high numbers of respondents (cp. chapter 3.2.2.2.2). In spatially enhanced consumer research the results of analysis are only valid for their actual spatial level of analysis and the shape of the chosen spatial units of aggregation. This is known as the modifiable areal unit problem (MAUP, cp. chapter 3.3.2) Conclusions from the regional level for small parts of this region are not possible. E.g. the user group of the region Costa de Llevant seems to consist of the average LCCP. But that does not mean that the visitor groups in the zones within this region have all characteristic values close to the average. On the contrary, the consumer research on the zonal level discovered very different consumer groups along the Costa de Llevant. In the same way it would be an ecological fallacy to conclude that all the visitors in one zone are similar in their characteristics. Some attributes are highly distinct to other regions and thus define partly the characteristics of the consumer group. Since some of them, however, are represented by less than half of the visitors to a zone, they indicate that the visitors within a zonal consumer group are probably still heterogeneous. But for a finer classification more data would be necessary.

risks of geospatial analysis

More uncertainties were introduced by geocoding the tourists. As not all could be assigned to places with complete reliability (cp. Chapter 5.2.1), this uncertainty is incorporated in all the results and should be kept in mind by the reader. In the same way, the reader should be aware that the mentioned figures are only the middle of confidence intervals spanning around that value in an amount depending on the confidence level (in this thesis most often 90%). The combination of spatial and classical statistical analysis introduces uncertainties and propagates errors on different levels. This, the user should be aware of. Therefore, it is mandatory to deliver fault tolerance tables and additional statistical measure helping the user to interpret the results and question their reliability. The risk of blind trust in computer generated outcome can be mitigated by documenting the results and possible errors and uncertainties. Thus, the user will be aware of the spatial level of analysis and the error margins where the results are reliable within.

96

8. Outlook

97

8. Outlook Since it is not to be expected that the technical development and the legal privacy issues will make it easier in the future to collect data about tourists, the requirement for a high number of cases and the difficulty to obtain additional qualitative statistics will hinder the further development of spatial analysis in tourist consumer research. But the now widely available and affordable GIS technology together with growing possibilities of coupling software could be a driving force for spatially aware consumer research. Spatial profiling of consumer groups and the analysis of their distribution will then be a useful and easy to achieve extension to classical tourist consumer research and consumer profiling. The results will assist participants in the tourism industry, especially destinations, to adjust their marketing and management. For Mallorca the advent of the LCCP might have some significant influences on the characteristics of tourism. The individual tourism will grow and thus the hotels’ dependence from the big European tour operators will decrease. This individual travelling is sign of a growing residential tourism. With cheap and frequent flights owners of holiday properties will be able to travel more often, but for shorter periods to Mallorca. All in all, the sojourn time of the tourists will decrease when the price of transportation stays low. Spatially, the different areas will develop more pronounced visitor groups in some areas. Especially the Playa de Palma might be again flooded by party seeking, young Germans, which are satisfied with beach and nightlife. With LCC even a weekend party in Mallorca is affordable and timely manageable when well-connected to the airport. All these developments depend on the future of the LCC. The market is sending ambiguous signals right now. The fuel prices rise and thus complicate the operations on low costs for the airlines. On the other side even new players enter the market as recent developments show (SOBIK 2007). Additional research is necessary to further explore the changes in the tourism structure of Mallorca caused by the impact of the low-cost carrier passengers.

References

References

AEA Association of European Airlines (2005): Yearbook 2004.

AENA. (2008a): Estadisticas Trafico: Trafico por ano Palma de Mallorca 2005.- Available at: http://estadisticas.aena.es/. [accessed 20-8-2008a] Spain, AENA Aeroportes Espanoles y Navigacion Aera.

AENA. (2008b): Estadisticas Trafico: Trafico por ano Palma de Mallorca 2007.- Available at: http://estadisticas.aena.es/. [accessed 20-8-2008b] Spain, AENA Aeroportes Espanoles y Navigacion Aera.

Ahas, R., Aasa, A., Mark, Ü., Pae, T., & Kull, A. (2007): "Seasonal tourism spaces in Estonia: Case study with mobile positioning data", Tourism Management, vol. 28, no. 3, pp. 898-910.

Albers, S. & Skiera, B. (1999): "Regressionsanalyse," in Marktforschung. Grundlagen - Methoden - Anwendungen, A. Herman & C. Homburg, eds., Wiesbaden, pp. 205-236.

Andrienko, N. & Andrienko, G. (2006): Exploratory Analysis of Spatial and Temporal Data. A Systematic Approach Springer, Berlin, Heidelberg.

Anselin, L. (1995): "Local indicators of spatial association LISA", Geographical Analysis, vol. 27, pp. 93-115.

Arrowsmith, C. & Itami, R. (2004): "Recreational and behaviour simulation validation", in 30th Congress of the International Geographic Union.-.

Arrowsmith, C. & Ntuwah, W. (2002): "Mapping potential for tourist segments", Journal of the Mapping Science Institute, vol. 31, no. 1, pp. 55-61.

Ashworth, G. & Goodall, B. (1990): Marketing Tourism Places Routledge, London, New York.

Bahaire, T. & Elliot-White, M. (1999): "The application of geographical information systems (GIS) in sustainable tourism planning", Journal of Sustainable Tourism, vol. 7, no. 2, pp. 159-174.

Bahrenberg, G., Giese, E., & Nipper, J. (1990): Statistische Methoden in der Geographie. Band 1: Univariate und bivariate Statistik.- 3. edn, B.G. Teubner, Stuttgart.

Bahrenberg, G., Giese, E., & Nipper, J. (1992): Statistische Methoden in der Geographie. Band 2: Multivariate Statistik.- 3. edn, B.G. Teubner, Stuttgart.

Bardolet, E. (1992): "Der Tourismus auf den Balearen: Bilanz eines Pionierziels des populären europäischen Tourismus im Mittelmeerraum," in Tourismus auf Mallorca.

i

http://estadisticas.aena.es/

http://estadisticas.aena.es/

References

Bilanz, Gefahren, Rettungsversuche, Perspektiven. Zu den Grenzen touristischen Wachstums, vol. 77. W. Isenberg & S. Lennartz, eds., Thomas-Morus-Akademie Bensberg. Bergisch Gladbach, pp. 33-62.

Becker, C. (2004): "Destinationsmanagment," in Geographie der Freizeit und des Tourismus: Bilanz und Ausblick, 2 edn, C. Becker, H. Hopfinger, & A. Steinecke, eds., Oldenbourg. München, Wien, pp. 464-474.

Belland, Greg and Boss, Erin. (2008): Cultural and Historical Sites: Assessing the Tourism Potential.- Available at: http://www.icomos.org/icomosca/bulletin/vol3_no3_belland_e.html. [accessed 20-7-2008] , ICOMOS International Council on Monuments and Sites.

Breuer, T. (1992): "Mallorca - Eine Bestandsaufnahme aus geographischer Sicht," in Tourismus auf Mallorca. Bilanz, Gefahren, Rettungsversuche, Perspektiven. Zu den Grenzen touristischen Wachstums, vol. 77. W. Isenberg & S. Lennartz, eds., Thomas-Morus-Akademie Bensberg. Bergisch Gladbach, pp. 9-32.

Brewer, C. A. (2005): Designing better Maps. A Guide for GIS USers ESRI Press, Redlands.

Brimicombe, A. (2003): GIS; environmental Modelling and Engineering Taylor & Francis, London, New York.

Brosius, F. (2007): SPSS für Dummies: statistische Datenanalyse stat Datenchaos - Daten aufbereiten und analysieren, grafische Darstellung der Ergebnisse, Export in Excel, Word & Co.- 1 edn, WILEY-VCH, Weinheim.

Burrough, P. A. (1994): "Principles of Modern Geographical Information Systems," in Remote Sensing for Environment and Forest Management, A. Mehrotra & R. K. Suri, eds., Indus Publishing Company. New Delhi, pp. 159-198.

Busch, H. & Leitner, B. (1992): "Urlauberbefragungen in Zielgebieten," in Erhebungsmethoden und ihre Umsetzung in Tourismus und Freizeit, C. Becker, ed., Trier, pp. 164-179.

Buswell, R. J. (1996): "Tourism in the Balearic Islands," in Tourism in Spain. Critical Issues., M. Barke, J. Towner, & M. Newton, eds., CAB International. Oxon, pp. 309-339.

Chen, R. J. C. (2007): "Geographic information systems (GIS) applications in retail tourism and teaching curriculum", Journal of Retailing and Consumer Services, vol. 14, no. 4, pp. 289-295.

CITTIB. (2003a): German tourist profile 2002.- Available at: http://www.inestur.es/documentos/document.en.377.pdf. [accessed 24-6-2007a] Palma de Mallorca, INESTUR Instituto de Estrategia Turística de las Illes Balears.

CITTIB. (2003b): Tourist profile in Mallorca 2002.- Available at: http://www.inestur.es/documentos/document.en.368.pdf. [accessed 24-6-2007b] , INESTUR Instituto de Estrategia Turística de las Illes Balears.

ii

http://www.icomos.org/icomosca/bulletin/vol3_no3_belland_e.html

http://www.inestur.es/documentos/document.en.377.pdf

http://www.inestur.es/documentos/document.en.368.pdf

References

Conseller de Turisme (2003): Dades Informatives 2002. El turisme a les Illes Balears. Mallorca.




Conseller de Turisme, INESTUR, CITTIB, and Jerez, Rafael Solivellas. (2007): Tourist Accommodation of Mallorca. Geographical Information System of Tourist Resources and Attractions - Mallorca.- Available at: http://sig.inestur.es:8080/index_mapa.php?lang=en. [accessed 22-5-2008] , INESTUR.

Coors, V., Huch, T., & Kretschmer, U. (2000): "TouriSTAR - ein mobiles AR-System in einer urbanen Umgebung," in Visualisierung raumbezogener Daten: Methoden und Anwendungen, Bd. II, vol. 8. B. Schmidt & C. Uhlenküken, eds., pp. 191-208.

Cuberos, R., Molina, N., Indriago, J. A., & Caldera, N. (2000): "Sigtur-Zulia: An Application of GIS Technologies for Tourism Planning", in Twentieth Annual ESRI User Conference.- ESRI.

de Smith, M. J., Goodchild, M. F., & Longley, P. A. (2007): Geospatial Analysis - A Comprehensive Guide to Principles, Techniques and Software Tools Matador, Leicester.

Dickmann, F. (2005): "Effectiveness and efficiency of tourism maps int the World Wide Web and their potential for mobile map services," in Map-based Mobile Services. Theories, Methods and Implementations, L. Meng, A. Zipf, & T. Reichenbacher, eds., Springer. Berlin, Heidelberg, pp. 45-56.

Dietvorst, A. G. J. (1995): "Tourist Behaviour and the Importance of Time-Space Analysis," in Tourism and spatial transformations. Implications for policy and planning, G. Ashworth & A. G. J. Dietvorst, eds., CAB International. Wallingford, pp. 163-181.

Diggle, P. J. (2003): Statistical analysis of spatial point patterns.- 2. edn, Arnold, Oxford University Press, London, New York.

DuMont Reisekartografie. (2007): Regionalkarte Mallorca.- 3. Edition.- Ostfildern, MairDumont. DUMONTdirekt. - Map

Dye, A. S. & Shaw, S.-L. (2007): "A GIS-based spatial decision support system for tourists of Great Smoky Mountains National Park", Journal of Retailing and Consumer Services, vol. 14, no. 4, pp. 269-278.

ELFAA. (2008): Background of ELFAA.- Available at: http://www.elfaa.com/background.htm. [accessed 10-2-2008] , ELFAA European Low Fares Airline Association.

iii

http://sig.inestur.es:8080/index_mapa.php?lang=en

http://www.elfaa.com/background.htm

References

Elliot-White, M. & Finn, M. (1998): "Growing in sophistication: The application of GIS in post-modern marketing", Journal of Travel and Tourism Marketing, vol. 7, no. 1, pp. 65-84.

Erkin, E. & Usul, N. (2005): "Site Selection for New Tourism Types in Bodrum Peninsula", in Twenty-Fifth Annual ESRI User Conference.- ESRI.

ESRI Environmental Systems Research Institute (1990): Understanding GIS. The Arc/Info® Method ESRI Press, Redlands.

ESRI Environmental Systems Research Institute. (2003): ESRI Data & Maps.- ESRI Environmental Systems Research Institute. [2003]. Redlands, ESRI. Ref Type: Data File

ESRI Environmental Systems Research Institute. (2007): MDA_Satellite_World_2D.- Redlands, ESRI. - Map

Farsari, Y. & Pratacos, P. (2004): "GIS applications in the planning and management of tourism," in A Companion to Tourism, A. A. Lew, C. M. Hall, & A. M. Williams, eds., Blackwell Publishing. Malden, Mass, pp. 596-607.

Feilmayr, Wolfgang. (2007): Regionale Tourismusplanung und -analyse.- Available at: http://www.srf.tuwien.ac.at/feil/tur.pdf. [accessed 15-8-2008] , Technische Universitaet Wien.

Ferrer, M. (2007): Mallorca.- 3 edn, DuMont Reise Verlag, Ostfildern.

Fichtner, U. (2004): "Evaluation in Freizeit und Tourismus: Theoretische und methodische Aspekte," in Geographie der Freizeit und des Tourismus: Bilanz und Ausblick, 2. edn, C. Becker, H. Hopfinger, & A. Steinecke, eds., München, Wien, pp. 502-514.

Fischer, M. M. (2006): Spatial Analysis and GeoComputation Springer, Berlin, Heidelberg.

Freyer, W. (1998): Tourismus: Eine Einführung in die Fremdenverkehrsökonomie.- 6. edn, Oldenbourg, München, Wien.

Freyer, W. (2004): Tourismus-Marketing. Marktorientiertes Management im Mikro- und Makrobereich der Tourismuswirtschaft.- 4. edn, Oldenbourg, München, Wien.

Fründt, H. J., Nahm, P. M., Seeler-Herzog, B., Strüber, R., & Goetz, R. (2006): Mallorca. Baedeker Allianz Reiseführer.- 12 edn, Karl Baedeker Verlag, Ostfildern.

Gabler-Mieck, R. & Duttmann, R. (2007): "Application of Geovisualisation Techniques in Costal-zone Management", in AGILE 2007.- M. Wachowicz & L. Bodum, eds., AGILE Association Geographic Information Laboratories Europe, Aalborg.

Gatrell, Anthony. (2008): Environmental Epidemiology and GIS.- Available at: http://geog.queensu.ca/h_and_e/healthandenvir/Finland%20Workshop%20Papers/HJALMARS.DOC. [accessed 20-8-0008] , Queens University Department of Geography.

iv

http://www.srf.tuwien.ac.at/feil/tur.pdf

http://geog.queensu.ca/h_and_e/healthandenvir/Finland%20Workshop%20Papers/HJALMARS.DOC

http://geog.queensu.ca/h_and_e/healthandenvir/Finland%20Workshop%20Papers/HJALMARS.DOC

References

Gimblett, H. R., Richards, M., & Itami, R. (2001): "RBSim: geographic simulation of wilderness recreation behavior", Journal of Forestry, vol. 99, no. 4, pp. 36-42.

Goodchild, M. F. (2005): "GIS and Modeling Overview," in GIS, Spatial Analysis, and Modeling, 1. edn, D. J. Maguire, M. Batty, & M. F. Goodchild, eds., Redlands, pp. 1-18.

Goodchild, M. F. & Janelle, D. G. (2004a): Spatially integrated social science Oxford University Press, New York.

Goodchild, M. F. & Janelle, D. G. (2004b): "Thinking Spatially in the Social Sciences," in Spatially integrated social science, M. F. Goodchild & D. G. Janelle, eds., Oxford University Press. New York, pp. 3-17.

Haak, E. (2009): "Image Analysis of the island of Mallorca from the Perspective of German Travellers to the Holiday Destination - An Empirical Study," in German Tourists on Mallorca. Travellers Perspectives and Descriptive Consumer Profiles, vol. 1. H. Wachowiak, ed., K. H. Bock Verlag. Bad Honnef, pp. 27-76.

Haggett, P. (2001): Geography: a global synthesis.- 1. edn, Pearson Education Limited, Essex.

Haining, R. (2003): Spatial Data Analysis: Theory and practice Press Syndicate of the University of Cambridge, Cambridge.

Haining, R., Wise, S., & Ma, J. (2000): "Designing and impelemting software for spatial statistical analysis in a GIS environment", Journal of Geographical Systems, vol. 2, no. 3, pp. 257-286.

Hall, C. M. & Page, S. J. (2006): The geography of tourism and recreation: environment, place, and space.- 3. edn, Routledge, New York.

Hallerbach, B. (2004): "Marktsegmentierung und der Trend zum hybriden Urlauber," in Geographie der Freizeit und des Tourismus: Bilanz und Ausblick, 2 edn, C. Becker, H. Hopfinger, & A. Steinecke, eds., Oldenbourg. München, Wien, pp. 171-180.

Heiberger, R. M. & Holland, B. (2006): Statistical Analysis and Data Display. An Intermediate Course with Examples in S-PLUS, R, and SAS.- 3 edn, Springer.

Hofmann-Göttig, Joachim. (2005): Meinungsforschung und Meinungsmache am Beispiel des Bundestagswahlkampfes 2005. Was taugt die "Sonntagsfrage"?- Available at: http://www.mbwjk.rlp.de/fileadmin/Dateien/Downloads/Ministerium/hofmann-goettig/Meinungsforschung_Meinungsmache.pdf. [accessed 12-9-2007] , Ministerium für Bildung, Wissenschaft, Jugend und Kultur Rheinland Pfalz.

Hopfinger, H. (2004): "Theorien im Tourismus. Anmerkungen zur theoriebezogenen Anschlussfähigkeit der Freizeit- und Tourismusgeographie," in Von Erholungsräumen zu Tourismusdestinationen: Facetten der Fremdenverkehrsgeographie, vol. 27. A. Brittner-Widmann, H.-D. Quack, & H. Wachowiak, eds., Selbstverlag der Geographischen Gesellschaft Trier. Trier, pp. 29-48.

v

http://www.mbwjk.rlp.de/fileadmin/Dateien/Downloads/Ministerium/hofmann-goettig/Meinungsforschung_Meinungsmache.pdf

http://www.mbwjk.rlp.de/fileadmin/Dateien/Downloads/Ministerium/hofmann-goettig/Meinungsforschung_Meinungsmache.pdf

References

Hudson-Smith, A. & Evans, S. (2003): "Virtual cities: from CAD to 3-D GIS," in Advanced Spatial Analysis. The CASA book of GIS, P. A. Longley & M. Batty, eds., ESRI Press. Redlands, pp. 41-60.

Huizingh, E. (2007): Applied Statistics with SPSS.- 1 edn, SAGE Publications, London, Thousand Oaks, New Delhi.

IDEE. (2007a): BDLL 25 - Municipios.- . IDEE La Infrastructura de Datos Espaciales de España. 10-8-2007a. Ref Type: Data File

IDEE. (2007b): EGM - Construcciones (Puntos).- IDEE La Infrastructura de Datos Espaciales de España. IDEE La Infrastructura de Datos Espaciales de España. Ref Type: Data File

IDEE. (2007c): EGM - Vias de Communicacion.- IDEE La Infrastructura de Datos Espaciales de España. IDEE La Infrastructura de Datos Espaciales de España. Ref Type: Data File

INE. (2004): Censos de Población y Viviendas 2001. Resultados definitivos.- Available at: http://www.ine.es/censo/es/inicio.jsp. [accessed 26-2-2008] , INE Instituto Nacional de Estadistica.

Institut Balear d' Estadística. (2006a): Evolució de la població dels municipis. Des de l'any 1996.- Available at: http://www.caib.es/ibae/dades/municipals/dadesmunicipals_archivos/sheet003.htm#RANGE!A1. [accessed 29-6-2007a] , CAIB Comunitat Auònoma Illes Balears.

Institut Balear d' Estadística. (2006b): Fitxes Municipals 2005. Municipis de Mallorca.- Available at: http://www.caib.es/ibae/fitxesmunicipals/mmallorca.htm. [accessed 8-11-2007b] , CAIB Comunitat Auònoma Illes Balears.

Institut Balear d' Estadística. (2007): Capacitat d'allotjament dels municipis i tipus d'establiment. Nombre d'establiments, unitats i places. 2005.- Available at: http://www.caib.es/ibae/dades/municipals/dadesm.htm. [accessed 29-6-2007] , CAIB Comunitat Auònoma Illes Balears.

Isaacson, M. & Shoval, N. (2005): "Tracking Spatial Behavior of Visitors - Old Acre", in Twenty-Fifth Annual ESRI User Conference.- ESRI.

Jambu, M. (1991): Exploratory and Multivariate Data Analysis Academic Press, Boston et al.

Joun, H. & Ryu, K.-H. (2004): "Using GIS for Middle or Low Rate Hotel Reservation", in The Twenty-Fourth Annual ESRI User Conference.- ESRI.

Katiyar, P. & Jain, N. (2007): "Tourist Sites Need Attention in Maharashtra: Shore RS and GIS - An Ultimate Tool to Explore and Develop Beaches as New Tourist Spots", in Map World Forum.-.

vi

http://www.ine.es/censo/es/inicio.jsp

http://www.caib.es/ibae/dades/municipals/dadesmunicipals_archivos/sheet003.htm#RANGE!A1

http://www.caib.es/ibae/dades/municipals/dadesmunicipals_archivos/sheet003.htm#RANGE!A1

http://www.caib.es/ibae/fitxesmunicipals/mmallorca.htm

http://www.caib.es/ibae/dades/municipals/dadesm.htm

References

Klemm, K., Lund-Durlacher, D., & Wolf, A. (2004): "Erhebnungs- und Auswertungsmethoden im Naherholungsverkehr: das Ausflugsverhalten der Berliner Bevölkerung," in Geographie der Freizeit und des Tourismus: Bilanz und Ausblick, 3. edn, C. Becker, H. Hopfinger, & A. Steinecke, eds., Oldenbourg. München, Wien, pp. 181-192.

Koch, S. & Wachowiak, H. (2007): Tourism on the Balearic Islands - A Commentary Literature Bibliography.

Krivoruchko, K. & Gotway Crawford, C. A. (2005): "Assesing the Uncertainty Resulting from Geoprocessing Operations," in GIS, Spatial Analysis, and Modeling, 1. edn, D. J. Maguire, M. Batty, & M. F. Goodchild, eds., Redlands, pp. 67-92.

Kumar, P. & Datta, D. (2007): "Finding Different Suitable Sites for Specific Tourism Purpose Using Geospatial Techniques", in Map World Forum.-.

Kunze, G. (1992): "Das touristische Mallorca," in Tourismus auf Mallorca. Bilanz, Gefahren, Rettungsversuche, Perspektiven. Zu den Grenzen touristischen Wachstums, vol. 77. W. Isenberg & S. Lennartz, eds., Thomas-Morus-Akademie Bensberg. Bergisch Gladbach, pp. 105-110.

Kwan, M.-P. (2000): "Analysis of Human Spatial Behavior in a GIS Environment: Recent Developments and Future Prospects", Journal of Geographical Systems, vol. 2, no. 1, pp. 85-90.

Larbig, C., Kämpf, R., & Keller, F. (2004): "Graubündner Destinationsbenchmarking. Der Einsatz eines Geographischen Informationssystem (GIS) als Planungs- und Managementinstrument in touristischen Zielgebieten", Tourismus Journal, vol. 8, no. 1, pp. 77-86.

Levy, R. M. & Dickson, E. E. (2006): "GIS, Remote Sensing, and Tourism Destination Management," in Community Destination Management in Developing Economies, W. Jamieson, ed., New York, London, Oxford, pp. 101-111.

Longley, Paul A. Position Statement.- Available at: http://www.ncgia.ucsb.edu/conf/sa_workshop/papers/longley.html. [accessed 28-2-2008] , NCGIA National Center for Geographic Information & Analysis.

Longley, P. A., Goodchild, M. F., Maguire, D. J., & Rhind, D. W. (2001): Geographic Information System and Science John Wiley & Sons, Chichester, New York, Weinheim, Brisbane, Singapore, Toronto.

Maasdam, R. (2000): Exploratory Data Analysis in Data Quality Monitoring Systems, MSc, University of Salford / UNIGIS.

Maguire, D. J. (2005): "Tools and Techniques: Introduction," in GIS, Spatial Analysis, and Modeling, D. J. Maguire, M. Batty, & M. F. Goodchild, eds., ESRI Press. Redlands, pp. 63-66.

vii

http://www.ncgia.ucsb.edu/conf/sa_workshop/papers/longley.html

References

Maguire, D. J., Batty, M., & Goodchild, M. F. (2005): "GIS, Spatial Analysis, and Modeling: Current Status and Future Prospects," in GIS, Spatial Analysis, and Modeling, 1. edn, D. J. Maguire, M. Batty, & M. F. Goodchild, eds., Redlands, pp. 445-456.

McAdam, D. (1999): "The value and scope of Geographical Information Systems in tourism management", Journal of Sustainable Tourism, vol. 7, no. 1, pp. 77-92.

Mitchell, A. (1999): The ESRI Guide to GIS Analysis. Volume 1: Geographic Patterns & Relationships ESRI Press, Redlands.

Mitchell, A. (2005): The ESRI Guide to GIS Analysis. Volume 2: Spatial Measurement & Statistics ESRI Press, Redlands.

Monmonier, M. (1996): Eins zu einer Million. Die Tricks und Lügen der Kartographen Birkhäuser Verlag, Basel, Boston, Berlin.

Myatt, G. J. (2007): Making Sense of Data. A Practical Guide to Exploratory Data Analysis and Data Mining John Wileys & Sons Inc., Hoboken.

Neumann, J. & Wachowiak, H. (2009): "The Daytrip Behaviour of German Tourists on Mallorca," in German Tourists on Mallorca. Travellers Perspectives and Descriptive Consumer Profiles, vol. Bachelor Dissertation. H. Wachowiak, ed., K.H. Bock Verlag. Bad Honnef, pp. 121-142.

O'Looney, J. (2004): "GIS and Enlightened Location-Based Tourism: An Innovation Whose Time Has Come", in The Twenty-Fourth Annual ESRI User Conference.- ESRI.

O'Sullivan, D. (2004): "Too Much of the Wrong Kind of Data. Implications for the Practice of Micro-Scale Spatial Modeling," in Spatially Integrated Social Science, M. F. Goodchild & D. G. Janelle, eds., Oxford University Press. Oxford, pp. 95-108.

O'Sullivan, D. & Unwin, D. J. (2003): Geographic Information Analysis John Wiley & Sons, Inc., Hoboken, New Jersey.

Öcal, S. & Usul, N. (2006): "Developing a GIS for Kars Sarikamis Winter Tourism Center", in The Twenty-Sixth Annual ESRI User Conference.- ESRI.

Openshaw.S. (1991): "Developing appropriate spatial analysis methods for GIS," in Geographic Information Systems. Volume 1: Principles, 1 edn, vol. 1. P. A. Longley et al., eds., Wiley & Sons. pp. 389-402.

Openshaw.S. and Alvanides, S. (1996): Designing Zoning Systems for representation of socio-economic data.- Available at: http://www.geog.leeds.ac.uk/software/zdes/. [accessed 19-2-2008] , University of Leeds - School of Geography.

Ord, J. K. & Getis, A. (1995): "Local Spatial Autocorrelation Statistics: Distributional Issues and an Application", Geographical Analysis, vol. 27, no. 4, pp. 287-306.

viii

http://www.geog.leeds.ac.uk/software/zdes/

References

Palladini, S. (2004): "ArcObjects Development in Zone Design Using Visual Basic for Application", in ICCSA 2004.- A. Laganà et al., eds., Springer, Berlin, pp. 1057-1068.

Piyathamrongchai, Kampanart and Tripathi, Nitin K. (2000): Dynamic Spatial Modeling using ROS and Carrying Capacity for Ecotourism Management.- Available at: http://www.gisdevelopment.net/aars/acrs/2000/ts7/gdi009.asp. [accessed 13-7-2007] Pathumthani, Thailand, GIS Development.

Ross, S. M. (2005): Introductory Statistics.- 2 edn, Elsevier Academic Press, Amsterdam et al.

RRZK Universität zu Köln. Regressionsanalyse.- Available at: http://www.uni-koeln.de/rrzk/kurse/unterlagen/statistik/regression.pdf. [accessed 15-4-2009] , RRZK Universität zu Köln.

Rutow, S. & Wachowiak, H. (2009): "Kultur als wertvolle Resource auf dem Weg zum Qualitätstourismus", TourHP, vol. 2009, no. 02, pp. 12-18.

Schilling, A., Coors, V., & Laakso, K. (2005): "Dynamic 3D Maps for Mobile Tourism Applications," in Map-based Mobile Services. Theories, Methods and Implementations, L. Meng, A. Zipf, & T. Reichenbacher, eds., Springer. Berlin, Heidelberg, pp. 233-246.

Schloemer, A. (1999): Nachhaltiger Tourismus?- 1 edn, Academia Verlag, Sankt Augustin.

Schmitt, T. (1999): Ökologische Landschaftsanalyse und -bewertung in ausgewählten Raumeinheiten Mallorcas als Grundlage einer umweltverträglichen Tourismusentwicklung Franz Steiner Verlag, Stuttgart.

Schmitt, T. (2000): ""Qualitätstourismus" - eine umweltverträgliche Alternative der touristischen Entwicklung auf Mallorca", Geographische Zeitschrift pp. 53-65.

Schmitt, T. (2007): "Qualitätstourismus auf Mallorca: "Ballermann" wäre besser", Rubin, vol. Special Issue, no. Special Issue, pp. 20-27.

Schmitt, Thomas and Blazquez i Salmon, Macia. (2004): Der dritte Tourismusboom auf Mallorca (1991 - 2000) - zukunftsweisender Trend oder überschrittener Zenit?- Available at: http://www.gobmallorca.com/deutsch/text_tourismusboom.pdf. [accessed 24-6-2007] , GOB - Grup Belear d'Ornitologia i Defensa de la Naturalesa.

Scholz, W. (2002): "Nachhaltiger (Massen-)Tourismus auf Mallorca? Die lokale Agenda 21 in Calvià/Mallorca," in ReiseRäume. Touristische Entwicklung und räumliche Planung, vol. 109. J. Borghardt et al., eds., IRPUD Institut für Raumplanung Universität Dortmund. Dortmund, pp. 238-254.

Schreiber, M. (2006): Low-Cost Airlines in Europa, PhD, University of Lüneburg.

Schürgers, P. (2005): Lokale Tourismus-Systeme Mallorcas. Integrative Analyse des Tourismus und seiner Raumrelevanzen als Grundlage ganzheitlicher lokaler Zukunftsentwürfe an der Badia de Pollenca, Badia d'Alcudia und Costa de Ponent Selbstverlag des Geographischen Institutes der Heinrich-Heine-Universität, Düsseldorf.

ix

http://www.gisdevelopment.net/aars/acrs/2000/ts7/gdi009.asp

http://www.uni-koeln.de/rrzk/kurse/unterlagen/statistik/regression.pdf

http://www.uni-koeln.de/rrzk/kurse/unterlagen/statistik/regression.pdf

http://www.gobmallorca.com/deutsch/text_tourismusboom.pdf

References

Seitz, E. & Meyer, W. (2006): Tourismusmarktforschung. Ein praxisorientierter Leitfaden für Touristik und Fremdenverkehr.- 2 edn, Verlag Franz Vahlen, München.

Silva, Oriana and Correia, Antonio. (2008): Facilitators and Constraints in Leisure Travel Participation: The Case of South East Portugal.- Available at: http://ertr.tamu.edu/conferenceabstracts.cfm?abstractid=2675. [accessed 20-7-0008] .

Sobik, Helge. (2007): Asiens Billigflieger: Für 100 Euro nach Hongkong.- Available at: http://www.focus.de/reisen/fliegen/billigflieger/tid-7329/asiens-billigflieger_aid_131850.html. [accessed 20-3-2008] .

Steinecke, A. (1988): "Urlaubserwartungen und Urlaubertypen - Möglichkeiten und Probleme der soziologischen und psychologischen Zielgruppenbestimmung und Marktsegmentierung," in Moderner Tourismus - Tendenzen und Aussichten, D. Storbeck, ed., Trier, pp. 325-344.

Steinecke, A. (1992): "Methoden der Marktsegmentierung und Zielgruppenanalyse: Möglichkeiten - Probleme - Perspektiven," in Erhebungsmethoden und ihre Umsetzung in Tourismus und Freizeit, C. Becker, ed., Trier, pp. 180-192.

Steinecke, A. (2006): Tourismus. Eine Geographische Einführung.- 1 edn, Bildungshaus Schulbuchverlag, Braunschweig.

Steingrube, W. (2004): "Erhebungsmethoden der Geographie der Freizeit und des Tourismus," in Geographie der Freizeit und des Tourismus. Bilanz und Ausblick, 2 edn, C. Becker, H. Hopfinger, & A. Steinecke, eds., Oldenbourg. München, Wien, pp. 138-148.

Stenau, S. (2009): "A Profile Analysis of Selected Quality-Tourism Segments on Mallorca - At the Example of German Golf and Nautical Tourists," in German Tourists on Mallorca. Travellers Perspectives and Descriptive Consumer Profiles, vol. 1. H. Wachowiak, ed., K. H. Bock Verlag. Bad Honnef, pp. 143-188.

Storbeck, D. (1992): "Einleitung: Der moderne Tourismus als Gegenstand interdisziplinärer Forschung," in Moderner Tourismus - Tendenzen und Aussichten, 3 edn, vol. 17. D. Storbeck, ed., Selbstverlag der Geographischen Gesellschaft Trier. Trier, pp. 11-38.

Stosch, Stefan. (2007): Jedem sein Mallorca.- HAZ Hannoversche Allgemeine Zeitung from 18-8-2007. vol. 2007, p.R/4. Madsack, Hannover.

Sweeney, S. H. & Feser, E. J. (2004): "Business Location and Spatial Externalities," in Spatially integrated social science, M. F. Goodchild & D. G. Janelle, eds., New York, pp. 239-262.

The Free Dictionary. (2008): consumer research.- Available at: http://www.thefreedictionary.com/consumer+research. [accessed 22-2-0008] , FARLEX.

Triola, M. F. (2008): Essentials of Statistics Pearson, Boston et al.

TUI (2008): "TUI Schöne Ferien. Mallorca. Menorca, Ibiza, Formentera", TUI Schöne Ferien.Mallorca.Menorca, Ibiza, Formentera, vol. März-Nov, no. 2008.

x

http://ertr.tamu.edu/conferenceabstracts.cfm?abstractid=2675

http://www.focus.de/reisen/fliegen/billigflieger/tid-7329/asiens-billigflieger_aid_131850.html

http://www.focus.de/reisen/fliegen/billigflieger/tid-7329/asiens-billigflieger_aid_131850.html

http://www.thefreedictionary.com/consumer+research

References

Tukey, J. W. (1977): Exploratory Data Analysis.- 1 edn, Addison-Wesley, Reading, Menlo Park, London, Amsterdam, Don Mills, Sydney.

Universität Rostock: Professur für Geodäsie und Geoinformatik. (2008): Kartogramm.- Available at: http://www.geoinformatik.uni-rostock.de/einzel.asp?ID=689587679. [accessed 1-8-2008] , Universität Rostock.

UNWTO. (2000): System of Tourism Statistics (STS) and its Links with the General Statistical System - Basic Statistics.- Available at: http://www.unwto.org/statistics/basic_references/index-en.htm. [accessed 22-9-2007] , UN World Tourism Organisation.

Vadell, J. B. G. (2005): El turismo de golf en Balears. 2005 INESTUR; CAEB, Palma de Mallorca.

Wachowiak, Helmut. (2005a): Forschungsvorhaben Tourismus auf den Balearen. Teilmodul: Aktuelle Konsumentenprofile deutscher Urlauber auf Mallorca. - Face-to-face-Befragung.- Wachowiak, Helmut. Bad Honnef, Bonn, FH Bad Honnef: Fachbereich Tourismusmanagement. Ref Type: Data File

Wachowiak, Helmut. (2005b): Forschungsvorhaben Tourismus auf den Balearen. Teilmodul: Aktuelle Konsumentenprofile deutscher Urlauber auf Mallorca. - Fluggastbefragung.- Wachowiak, Helmut. Bad Honnef, Bonn, FH Bad Honnef: Fachbereich Tourismusmanagement. Ref Type: Data File

Wachowiak, H. & Neumann, J. (2008): "The daytrip behaviour of German tourists on Majorca", Advances in hospitality and leisure, vol. in press..

Walch, S. (1999): Implementierung von Marketingstrategien in Tourismusregionen Haupt, Bern, Stuttgart, Wien.

Weeks, J. R. (2004): "The Role of Spatial Analysis in Demographic Research," in Spatially integrated social science, M. F. Goodchild & D. G. Janelle, eds., New York, pp. 381-399.

WELT ONLINE. (2007): Familienferien: Bringen Sie doch Ihre Kinder mit.- Available at: http://www.welt.de/reise/article721188/Bringen_Sie_doch_Ihre_Kinder_mit_.html. [accessed 24-5-2008] , Welt Online (Springer).

Wikipedia. Finca.- Available at: http://de.wikipedia.org/wiki/Finca. [accessed 20-3-2009] , Wikipedia.

Wong, D. W. S. & Lee, J. (2005): Statistical Analysis of Geographic Information with ArcView GIS® and ArcGIS® Wiley & Sons, Hoboken, New Jersey.

Worboys, M. F. & Duckham, M. (2004): GIS: a computing perspective.- 2. edn, CRC Press, Boca Raton (Florida).

xi

http://www.geoinformatik.uni-rostock.de/einzel.asp?ID=689587679

http://www.unwto.org/statistics/basic_references/index-en.htm

http://www.welt.de/reise/article721188/Bringen_Sie_doch_Ihre_Kinder_mit_.html

http://de.wikipedia.org/wiki/Finca

References

xii

Wright, D. J., Goodchild, M. F., & Proctor, J. D. (1997): "Demystifying the Persistent Ambiguity of GIS as "Tool" Versus "Science"", The Annals of the Association of American Geographers, vol. 87, no. 2, pp. 346-362.

Zipf, A. (2002): "Projektabschlußbericht Deep Map/GIS", HGG-Journal, vol. 17.

Zwan, J. van der. (2007): Low-Cost Carriers - Europe.- Available at: http://www.jvdz.net/. [accessed 20-8-2008] .

http://www.jvdz.net/

Attachments Attachment 1 ........................................................................................................................ A-2 1. Implemented Data......................................................................................................... A-2

1.1. Attribute Data ......................................................................................................... A-2 1.2. Geographic Data.................................................................................................... A-3

Attachment 2 ........................................................................................................................ A-4 2. Statistical Significance .................................................................................................. A-4

2.1. Confidence Interval ................................................................................................ A-4 2.2. Regression Analysis .............................................................................................. A-7

Attachment 3 ...................................................................................................................... A-14 3. Assign tourists to locations with multiple occurrences ................................................ A-14

3.1. Cala Figuera ........................................................................................................ A-14 3.2. Sa Coma .............................................................................................................. A-14

Attachment 4 ...................................................................................................................... A-17 4. Interview forms............................................................................................................ A-17

4.1. Outbound Flight Interview Form........................................................................... A-18 4.1.1. German Interview Form ................................................................................ A-18 4.1.2. English Interview Form ................................................................................. A-22

4.2. Return Flight Interview Form................................................................................ A-26 4.2.1. German Interview Form ................................................................................ A-26 4.2.2. English Interview Form ................................................................................. A-30

4.3. Face-to-face Interview Form ................................................................................ A-34 4.3.1. German Interview Form ................................................................................ A-34 4.3.2. English Interview Form ................................................................................. A-39

A-1

Attachment 1

1. Implemented Data

1.1. Attribute Data

Title Content Spatial level Source of attribute data Year

LCC passengers LCC passengers characteristics

Location University of Applied Science Bad Honnef/ Bonn: passenger questionnaire

2005

German tourists Tourists characteristics Location University of Applied Science Bad Honnef/ Bonn: face-to-face questionnaire

2005

Municipality of destination

Relative distribution of tourists (subdivided by nationality)

Municipality (CONSELLER DE TURISME 2007)

2006

Region of destination Count of visitors (subdivided by nationality)

Tourist regions (CONSELLER DE TURISME 2003;CONSELLER DE TURISME 2008)

2002/2007

Accommodation Places and beds Municipality (CONSELLER DE TURISME 2006)

2005

Restaurants and Bars

Number of restaurants and bars

Municipality (INSTITUT BALEAR D' ESTADÍSTICA 2006)

2005

Secondary homes Number of secondary homes

Municipality (INE 2004) 2001

Table A-1.1: List of available tourism statistics with spatial differentiation

A-2

1.2. Geographic Data Title Content Repre-

sentation Source of geographic data Creation Maximum

Scale

Tourism locations, POI

Places on Mallorca visited by tourists

Point “Regionalkarte Mallorca”, (DUMONT REISEKARTOGRAFIE 2007)

Scanned and digitsed

1:650,000

Built areas Built areas on Mallorca

Polygon MDA Satellite Images (ESRI ENVIRONMENTAL SYSTEMS RESEARCH INSTITUTE 2007)

Digitised 1:650,000

Municipalities Municipalities of Mallorca

Polygon (IDEE LA INFRAESTRUCTURA DE DATOS ESPACIALES DE ESPAÑA 2007)

Data downloaded 1:25,000

Tourism regions

8 aggregated regions on Mallorca

Polygon (CONSELLER DE TURISME 2008)

Vectorised raster image

1:25,000

Tourism sectors

5 aggregated sectors on Mallorca

Polygon (CONSELLER DE TURISME 2006)


1:25,000

Tourism zones

20 zones accommodation best the results of the LCCP survey 2005

Polygon Created by Johannes Luberichs (for details see chapter 5.1.3)

Digitised 1:650,000

Average temperature July

Temperature classes for July in °C

Polygon (SCHÜRGERS 2005) Vectorised raster image

1:650,000

Average temperature January

Temperature classes for January in °C


1:650,000

Precipitation Annual precipitation


1:650,000

Coast Coastal type Line (SCHMITT 1999; SCHÜRGERS 2005)


1:25,000

Roads Line (DUMONT REISEKARTOGRAFIE 2007; ESRI ENVIRONMENTAL SYSTEMS RESEARCH INSTITUTE 2003; IDEE LA INFRASTRUCTURA DE DATOS ESPACIALES DE ESPAÑA 2007)

Spatially combined datasets complemented by data from a scanned map

1:10,000

Secondary homes

Proportion of secondary homes in Mallorca

Polygon (SCHMITT 2007) Scanned map 1:650,000

Table A1.2: List of implemented geographic data

A-3

Attachment 2

2. Statistical Significance

2.1. Confidence Interval

Even if the samples are representative (see chapter 3.2.2.2) they are only a small portion of the whole population. That is why single statistics like the mean or a proportion calculated from the sample only near the true values. These single values would not reflect the amount of confidence in the value, meaning how far the real value could differ from the calculated one. But it is possible to quantify the range the observed values might deviate from the true value. This range is called the confidence interval. The size of this interval is dependent on the risk the researcher is willing to take of being wrong. The more he wants to be confident about his values, the bigger the confidence interval will be. So there will always be a trade-off between a close estimation of the right value and a low risk of being wrong, as both are not possible. The risk is measured by the so called confidence coefficient and its common values are 90 %, 95 % or 99 % (HEIBERGER & HOLLAND 2006, p. 41ff; ROSS 2005, p. 348f). Confidence ranges are differently calculated for categorical and continuous variables. Continuous variables are often summarised by single statistics as, for example, the mean. But the sample mean is only an estimate of the population mean. Therefore, the confidence range is to be calculated to indicate the variation of the sample mean from the population mean. The average age, for example, of the LCC passengers is 39.04 years. With a 95% confidence level the average age is between 38.37 and 39.72 (with 90% probability the average age is between 38.48 and 39.61 years).

A-4

p 1% 2,5% 5% 10% 25% 50% 90%

n cl 90% 95% 90% 95% 90% 95% 90% 95% 90% 95% 90% 95% 90% 95%

500 (0.73) (0.87) 1.15 (1.37) 1.60 1.91 2.20 2.63 3.18 3.80 3.67 4.38 2.20 2.63 750 (0.60) (0.71) 0.93 1.12 1.31 1.56 1.80 2.15 2.59 3.10 2.99 3.58 1.80 2.15 1000 (0.52) (0.62) 0.81 0.97 1.13 1.35 1.56 1.86 2.25 2.68 2.59 3.10 1.56 1.86 1100 0.49 (0.59) 0.77 0.92 1.08 1.29 1.48 1.77 2.14 2.56 2.47 2.95 1.48 1.77 1500 0.42 0.50 0.66 0.79 0.92 1.10 1.27 1.52 1.83 2.19 2.12 2.53 1.27 1.52 1696 0.40 0.48 0.62 0.75 0.87 1.04 1.20 1.43 1.73 2.07 2.00 2.39 1.20 1.43 1750 0.39 0.47 0.61 0.73 0.85 1.02 1.18 1.41 1.70 2.03 1.96 2.34 1.18 1.41 2000 0.36 0.44 0.57 0.68 0.80 0.96 1.10 1.31 1.59 1.90 1.83 2.19 1.10 1.31

P = Proportion of characteristic in population (%) p = proportion of characteristic in sample n = sample size t = 1.64 (confidence level 90%) or t = 1.96 (confidence level 95%)

Table A-2.1: Error margins

Categorical variables are often analysed by computing the proportion of the sample with a particular value. To make a statement about the whole population the confidence interval under a certain confidence coefficient has to be indicated. Some margins of error defining the confidence intervals are presented in table A-2.1. The table is to be read as in the following example: Roughly 3% from the LCCP travel to the area defined by INESTUR as the interior of Mallorca. The sample consists of 1696 valid cases. With a probability of 90% the value for the whole population will be in the interval between 2.3% and 3.7%. The confidence interval estimate always has to be calculated for the proportion of all valid cases of the sample. This is important to keep in mind when working with proportions of subgroups. The spatial analysis of the data often results in grouping the cases of the sample by a certain spatial unit like to the regions in the former example. In this inland region, 24% of visitors stay for five to seven days in Mallorca. Due to the fact that only 3% of the visitors stay in the inland, the proportion of people who stayed for five to seven days in the inland in relation to the whole population is even smaller. In whole numbers, those are 13 out of 1612 respondents who answered this question, which makes a share of 0.8%. The confidence interval estimate is 0.45% to 1.15%. If the error margin exceeds the half of the proportion the results are said to be suspicious and thus not interpretable (cp. HOFMANN-GÖTTIG 2005, p. 13). In table A-2.1 these values are shown in brackets (in the detailed table they are printed red). The example is close to the acceptable value and would have been omitted when working with a 95% confidence coefficient. An extended table is attached on the following side.

A-5

pn t 0.50% 0.75% 1% 2.50% 3% 5 10% 15% 20% 25% 30% 40% 50% 60% 70% 80% 90%

50 1.64 - - - - - - 6.96 8.28 9.28 10.04 10.63 11.36 11.60 11.36 10.63 9.28 6.96100 1.64 - - - - 2.80 3.57 4.92 5.86 6.56 7.10 7.52 8.03 8.20 8.03 7.52 6.56 4.92200 1.64 - - - 1.81 1.98 2.53 3.48 4.14 4.64 5.02 5.31 5.68 5.80 5.68 5.31 4.64 3.48300 1.64 - - 0.94 1.48 1.62 2.06 2.84 3.38 3.79 4.10 4.34 4.64 4.73 4.64 4.34 3.79 2.84400 1.64 - 0.71 0.82 1.28 1.40 1.79 2.46 2.93 3.28 3.55 3.76 4.02 4.10 4.02 3.76 3.28 2.46500 1.64 - 0.63 0.73 1.15 1.25 1.60 2.20 2.62 2.93 3.18 3.36 3.59 3.67 3.59 3.36 2.93 2.20600 1.64 0.47 0.58 0.67 1.05 1.14 1.46 2.01 2.39 2.68 2.90 3.07 3.28 3.35 3.28 3.07 2.68 2.01700 1.64 0.44 0.53 0.62 0.97 1.06 1.35 1.86 2.21 2.48 2.68 2.84 3.04 3.10 3.04 2.84 2.48 1.86750 1.64 0.42 0.52 0.60 0.93 1.02 1.31 1.80 2.14 2.40 2.59 2.74 2.93 2.99 2.93 2.74 2.40 1.80800 1.64 0.41 0.50 0.58 0.91 0.99 1.26 1.74 2.07 2.32 2.51 2.66 2.84 2.90 2.84 2.66 2.32 1.74900 1.64 0.39 0.47 0.54 0.85 0.93 1.19 1.64 1.95 2.19 2.37 2.51 2.68 2.73 2.68 2.51 2.19 1.64

1000 1.64 0.37 0.45 0.52 0.81 0.88 1.13 1.56 1.85 2.07 2.25 2.38 2.54 2.59 2.54 2.38 2.07 1.561100 1.64 0.35 0.43 0.49 0.77 0.84 1.08 1.48 1.77 1.98 2.14 2.27 2.42 2.47 2.42 2.27 1.98 1.481200 1.64 0.33 0.41 0.47 0.74 0.81 1.03 1.42 1.69 1.89 2.05 2.17 2.32 2.37 2.32 2.17 1.89 1.421300 1.64 0.32 0.39 0.45 0.71 0.78 0.99 1.36 1.62 1.82 1.97 2.08 2.23 2.27 2.23 2.08 1.82 1.361400 1.64 0.31 0.38 0.44 0.68 0.75 0.96 1.31 1.57 1.75 1.90 2.01 2.15 2.19 2.15 2.01 1.75 1.311500 1.64 0.30 0.37 0.42 0.66 0.72 0.92 1.27 1.51 1.69 1.83 1.94 2.07 2.12 2.07 1.94 1.69 1.271600 1.64 0.29 0.35 0.41 0.64 0.70 0.89 1.23 1.46 1.64 1.78 1.88 2.01 2.05 2.01 1.88 1.64 1.231680 1.64 0.28 0.35 0.40 0.62 0.68 0.87 1.20 1.43 1.60 1.73 1.83 1.96 2.00 1.96 1.83 1.60 1.201750 1.64 0.28 0.34 0.39 0.61 0.67 0.85 1.18 1.40 1.57 1.70 1.80 1.92 1.96 1.92 1.80 1.57 1.18

P = Proportion of characteristic in population (%)Formular p = proportion of characteristic in sample

n = sample sizet = 1.64 (confidence level 90%) or t = 1.96 (confidence level 95%)

If the error margin exceeds the half of the proportion the results are said to be suspicious and thus not interpretable. These figures are printed red.

2.2. Regression Analysis

The goal of a regression analysis is the description of the relationship between a dependent variable (also called response or outcome variable) and one or more independent variables (also called predictors) via a mathematical function. This function should be able to calculate the best possible estimate for the dependent variable (BROSIUS 2007, p. 255f). Amongst others, this study wants to find out the strength of the influence of the variables like accessibility, supply and natural factors on the distribution of the LCCP. Which type of function might best suit depends on the relationship between the response variable and the predictors. Examples of these are the linear function, the parabola function or the exponential function. When working with more than one predictor the regression is a multiple regression. To model them all in one function they must have the same type of relationship to the outcome variable. If the variables have different scales, then they even have to be standardised to be comparable. If the outcome variable is likely to be different from zero even if the predictors are zero the regression has to be modelled with a constant that accounts for this effect (BROSIUS 2007, p. 258). The regression analysis is one of the most applied statistical analysis methods, especially the linear regression (ALBERS & SKIERA 1999, p. 206). It is easy to use and is chosen often even if the relationship between outcome and predictor is non-linear for some variables (RRZK UNIVERSITÄT ZU KÖLN). This method of course contains in itself the danger of misleading or no results. Several iterations with different combinations of independent variables proved that for the LCCP in Mallorca by providing no significant results. The following tables only display the results of the linear regression with the predictors that could be linear modelled in relation to the dependent variable: the number of respondents per municipality (see tables A-2.3 to A-2.8). It is calculated with a constant as it is very likely to have tourists even if all the selected predictors would be zero.

R R² Adjusted R² Std. Error of the Estimate Significance of Regression

0.934 0.873 0.846 46.369 0.000 Predictors: (Constant), no. of gastronomy establishments, no. of holiday homes, no of hotel beds Dependent variable: no. of respondents

Table A-2.3: Model summary multiple linear regression (all linear related predictors)

Unstandardised coefficients Standardised coefficients

Model B Std. Error Beta

t Sig.

(Constant) 7.066 16.678 0.424 0.678 No. of gastronomy establishments

0.112 0.020 0.714 5.510 0.000

No. of holiday homes 0.002 0.005 0.057 0.367 0.719 No. of hotel beds 0.002 0.001 0.268 1.933 0.074

Table A-2.4: Model coefficients multiple linear regression (all linear related predictors)

A-7

In order to find out which factors have what impact on the distribution of the German LCCP in Mallorca all available information that could be a pull factor for tourists was considered (see table A-2.9). Since most of it was only available on a municipality level it was chosen as the level for analysis. All possible curves modelling the relationship between the dependent variable and each independent variable have been estimated, but only the most significant are displayed in table A-2.9. In addition, the linear function is always displayed to highlight the problems encountered with the multiple linear regression.


0.916 0.839 0.818 50.419 0.000 Predictors: (Constant), no. of gastronomy establishments, no. of holiday homes Dependent variable: no. of respondents

Table A-2.5: Model summary double linear regression



t Sig.

(Constant) 94.222 11.884 7.929 0.000 No. of gastronomy establishments (z-score)

88.543 16.471 0.749 5.376 0.000

No. of holiday homes (z-score) 26.684 16.471 0.226 1.620 0.126 Table A-2.6: Model coefficients double linear regression

Although all the listed factors probably influence in some way the distribution patterns of the LCCP in Mallorca, they cannot be all modelled in one equation. For once, this is owed to the different types of statistical relationships between the distribution of LCCP and the independent variables. On the other side, a high multicollinearity exists between some of the variables. The incorporation of the strong correlated variables hotel beds, holiday homes and gastronomy (see table 3 in chapter 6.1) would lead to a function that might be a good estimation for the response variable, but could not tell the weight a single variable for the LCCP distribution. One could not reliably make the statement that the change in one predictor causes the response variable to change a certain amount (cp. BAHRENBERG, GIESE, & NIPPER 1992, p. 41/42). That is why after a series of try outs even the resulting best fitting regression models (shown in the tables A-2.3 to A-2.8) could not help explain the distribution of the LCCP better as it could be done without it.

A-8


0.934 0.872 0.855 45.012 0.000 Predictors: (Constant), no. of gastronomy establishments, no. of hotel beds Dependent variable: no. of respondents

Table A-2.7: Model summary double linear regression



t Sig.

(Constant) 9.290 15.082 0.616 0.547 No. of gastronomy establishments

88.543 16.471 0.749 5.376 0.000

No. of hotel beds 0.002 0.001 0.296 2.667 0.018 Table A-2.8: Model coefficients double linear regression

The shown regressions do not suffer only from the multicollinearity but in addition cannot provide significant results. On the first glance, all of them seem to be a good model because they have a high R² value close to one and are statistically significant (see tables A-2.3, A-2.5, A-2.7). Looking at the single coefficients, however, at least one of them is insignificant. In the multiple regression with the three linear related predictors (table A-2.4) even the two the constant and the number of holiday homes are insignificant. Leaving out the hotel beds, the holiday homes coefficient remains insignificant even within a 90% confidence range. Only the constant becomes significant (table A-2.6). Somewhat better but not finally satisfying is the regression modelled with the predictors ‘number of gastronomy establishments’ and ‘number of hotel beds’ (see table A-2.8). Nevertheless, the constant here is again insignificant. Even from the logical point of view these two last models are not satisfying. They both show a high unstandardised coefficient for the gastronomy factor. On the one hand, that cannot be considered right because of the multicollinearity (mentioned above). On the other hand, although even other less well fitting regression models suggested a huge influence of the gastronomy it is probably their image as a whole that attracts the visitors. The opening of one single establishment would not attract a measurable amount of new visitors. In addition the correlation might partly be the other way around. Maybe not the tourists follow the gastronomy supply but the gastronomy follow the tourists. Due to different types of statistical relationships between the distribution of LCCP and explaining variable as well as correlations between those variables regression analyses were not able to show the strength of the influence of the forenamed influencing factors.

A-9

A-10

Independent variable

Description Curve Estimation

Men-made tourist infrastructure No. of hotel beds

It is very likely that the supply of accommodation has a direct, strong influence on the distribution of the LCCP around Mallorca. They correlate positively (see chapter 6.1). Their relationship however is best expressed by a cubic or compound function. A linear relationship is weak but has a sufficient significance for using this variable in multiple linear regression. The beds in hotel and hostel accommodation are only a part of the available accommodation for tourists. That is why for explanation reasons other types of overnight accommodations play a vital role, too.

Equation R² Sig.

Linear 0.498 0.001 Cubic 0.609 0.004 Compound 0.514 0.001

No. of holiday homes

The residential tourism has a high market share in the LCCP tourism. Therefore, its influences on the distribution of the LCCP might be significant. There is a positive correlation between the number of holiday homes per municipality and the number of LCCP (see chapter 6.1) that is stronger than that of LCCP and hotel beds. Even if a quadratic modelling of the relationship would fit better, a linear representation is still significant.

Equation R² Sig.

Linear 0.530 0.001 Quadratic 0.545 0.003 Cubic 0.782 0.000



No. of gastronomy establishments

Gastronomy is a very important side business to tourism. A lot of LCCP rated the visiting of a restaurant as important or even very important. In places like the Playa de Palma or in Cala Rajada (in Capdepera) pubs and clubs make the reputation for nightlife and thus attract visitors. Although probably not a single restaurant makes the difference but the whole scene, a high positive correlation exists between the number of LCCP and the number of gastronomy establishments (see chapter 6.1). The relationship can be modelled as linear, but as well shows signs of a quadratic or cubic function.

Equation R² Sig.

Linear 0.811 0.000 Quadratic 0.811 0.000 Cubic 0.812 0.000

Distance to motorway

The distance to the motorway was averaged for all visitors per municipality. The geospatial analysis of the distribution of LCCP in relation to the motorway connection suggested that there is some influence to the distribution regarding the accessibility of different places (see chapter 6.1, map 14). But since there are in distant places like Capdepera a high number of people, the relationship to the number of LCCP is inverse and cannot be reliably expressed with a linear function.

Equation R² Sig.

Linear 0.082 0.251 Inverse 0.732 0.000

A-11



Distance to airport

As a more simple measure of the accessibility the distance to the airport has been considered for analysis. But for the same reasons (a lot of people close to the airport, few people in middle distance, and again many people far from the airport) there is no liner relationship between the number of LCCP and the distance to the airport. It probably has influences on the distribution. But in some places the pull factors are much stronger and outweigh the accessibility constraints.

Equation R² Sig.

Linear 0.187 0.073 Inverse 0.640 0.000 Cubic 0.660 0.001

Natural factors Distance to coast

The distance to the coast was calculated as weighted average of all LCCP in a municipality. The coast has a huge attraction to the tourists. 4 out of 5 LCCP stay close to the coast (see chapter 6.1). Nevertheless, no function can describe mathematically the relationship of the number of LCCP and the distance to the coast on municipality level with significant results.

Equation R² Sig.

Linear 0.019 0.584

A-12



Yearly Precipitation

The yearly precipitation is very different around the island. And there are only few people in the area with the highest precipitation. Although it is certain that a lot of tourists chose their holiday destination after considering the weather, in this case reasons for avoiding the wet places are very likely different. The area of high precipitation is in the rough and inaccessible part of the Tramuntana Mountains. There exists no infrastructure for tourists. It is probably more the landscape and the sparse accommodation supply than the weather that keeps LCCP from visiting the Costa de Tramuntana. The distribution of LCCP in relation to the rain cannot be expressed by a reliable function.

Equation R² Sig.

Linear 0.029 0.497

Average temperature in July

The hottest and the coldest places are avoided by the LCCP. However, that still does not make a calculation of the relation possible. In addition are there other reasons that probably outweigh the temperature as main factor. Those reasons are the same as for the precipitation.

Equation R² Sig.

Linear 0.050 0.373

Table A-2.9: Possible predictors for the distribution of LCCP in Mallorca

A-13

A-14

Attachment 3

3. Assign tourists to locations with multiple occurrences Since there are some places on Mallorca that share the same name, it was problematic to assign visitors to the right place. At a first glance, that seems to pertain to Cala Figuera und Sa Coma. Both occur three times. By comparing the local situations of all these places by the Mallorca Accommodation website, which offers localisation of accommodation with areal photos, Google Earth and the Map Viewer of the IDEE, as well as other available data a rough key for assigning the tourist to different places has been generated.

3.1. Cala Figuera

One Cala Figuera is an unpopulated beach in a bay near of Cap Formentor at the southern tip of the island, belonging to the municipality of Pollenca. In a German guidebook it is said to be a day trip destination for people who really love about solitariness1 (FRÜNDT et al. 2006). Nevertheless, no tourists will have stayed there overnight. Another Cala Figuera is situated at the western side of the Bay of Palma in the municipality of Calvia. The nearby cape is named after it Cap de Cala Figuera. On some maps Cala Figuera is somewhat misleading placed next to the small village Portals Vells and seems to be populated. However, closer investigation showed that Cala Figuera is only a small bay approximately one kilometre away from Potals Vells where no buildings exist. So no tourists will have stayed there as well. That means all tourists said to have stayed in Cala Figuera did so in the one situated at the south eastern coast in the municipality of Santanyi.

3.2. Sa Coma

The place name Sa Coma also occurs three times. But all places are populated. So the people might not have been all to the same Sa Coma. Because of the different sampling techniques of the two polls the assigning is done in different ways. Four tourists from the face-to-face poll have stayed in Sa Coma. They have been assigned all to the Sa Coma situated at the west coast in Sant Llorenc des Cardassar because the interviews have been done in the near Cala Millor.

1 The German word used is „Einsamkeitsfanatiker”.

[_

#*

[_

#*

[_

#*


Recurring Places#* Cala Figuera[_ Sa ComaTourist Beds 2005

253 - 12351236 - 74467447 - 1875118752 - 2949629497 - 59750

0 5 10 15Kilometers

Map A-3.1: Recurring place names in Mallorca

There where 54 LCC passengers who said to have been in Sa Coma. For the geocoding of all these tourists the characteristics of the locations are important. The Sa Coma in Sant Llorenc de Cardassar is the only place where tourist accommodations are offered. So all the visitors to Sa Coma staying in a hotel, hostel or rented holiday flat were assigned to that place, with the number of 41 respondents. Two of the remaining did not answer the question on their type of accommodation, eleven owned a holiday home or flat or stayed in the house or flat of friends or relatives. Sa Coma at Andratx has approximately ten percent the size of Sa Coma in Sant Llorenc. It is not really a tourist area, located 5 km away from the sea. But the municipality of Andratx has one of the highest proportions of second homes in Mallorca. That is why ten percent of the remaining 13 respondents were rounded up to 2 and assigned to Sa Coma in Andratx; the others to the one in Sant Llorenc de Cardassar. The Sa Coma at Valldemossa lies in an even less touristic area and is a very small settlement. Therefore, it is highly improbable that one of these houses has been visited by one of the respondents.

A-15

Municipality % 2nd Homes

Distance to sea

Coastal type Hotels Houses Assigned LCCP

Andratx > 60 % 5 km Mostly steep coast, only some small

beaches

0 Ca. 30 52


50 – 60 % 0 m Beach Ca. 20 Ca. 300 2

Valldemossa 40 – 50 % 3.5 km Steep coast 0 Ca. 5 0 Table A-3.1: Places named Sa Coma on Mallorca

Of course, this is all a form of speculation. That most of the respondents stay in the Sa Coma on the western coast is highly probable due to the development in that village. Leaving out the entire tourists from Sa Coma because of geocoding difficulties would have distorted the results even more as they might be now.

A-16

Attachment 4

4. Interview forms

4.1 Outbound Flight Interview Form 4.1.1 German Outbound Flight Interview Form 4.1.2 English Outbound Flight Interview Form

4.2 Return Flight Interview Form 4.2.1 German Return Flight Interview Form 4.2.2 English Return Flight Interview Form 4.3 Face-to-Face Interview Form 4.2.1 German Face-to-Face Interview Form 4.2.1 German Face-to-Face Interview Form

A-17

��

��

��

� �� !�� "��!��"��

��#�� $ ��

��

%�� #�� &� � � � �' � �(��)� ��*�� +,�

-�� #��)� �,�� &�� .�.��

/�� )� �,��/�' ��&� � � .�.��

+�� 0 ��#�� &� � .�.�� 0 ��

�

��

1�� #��&�)�� ,�� 2 �� (�� 3��

�� #��..............� � �� 4�

5�� #��&�� 6�� "�� 0�� #��

7��8 ��#�� "��9��&�

�� 3�� :�� 6�� ;�� <��

=��#�� !��&�� ' �� (�� < ��

>��#��#�� &�)��

�� ?�� ,�

�� @�� 8 $ ��9�A�� 8 $ ��9��A��;B$�� C�� C�

�� 903�� C��9��A�� D��E0�� D�� #��.....................................�

%F�� #��"��G��6��&��

�� H%�� %I/�� /I5�� J5��

%%�� #��"��8�� &��8��8��*��.�.�.�;��

��8�I��6�� *��.�.�.�;��

%-��#��2��&�� 2��8��*��.........�)��%I1,�

��2��8�I��6�� *.........��)��%I1,�� K��?�!K��

�

"� ��#� $%� �#��

%/��(��#��L�� "��3��*........................�

%+��"��L�� #��"�� )� �� ,&�

��........................................................................................�

%1�� ,��#��C��"��3�� *�

�� 8�� ED�� 8�� E2�� EI�� EC��

�� 8 ��E�� 8 �� E�� E:�� C� ��

��,��8��6��< ��&�� M� �MM� �MMM� �MMMM� �MMMMM�� C� ��

��,�C��:��$��*�� #�� $�� 82�� :2� �� C��

�

Ich

Schreibmaschinentext

Outbound Flight Interview

&� ��

%5��8 ��#�� #�� &�� ' � �(��

��'�� #�� &��.�.��0 ��

�

%7�� #��"�� )��"��,&�� "�� 6��

�� :�� :�� C�� :�� #��

�(��)��

%=�� #��' ��)-FF+,�3�� &��

�� ,�6��1�0 ��!�� .�.�� C� ��

��,�6�� 1�0 �� .�.�� C� ��

��,�"��9 �E9!�� #��"��3�� &�....................................�

��,�� "��' �� $�� 2 �� &�

�� 2 �� 3�� C� ��

%>��#�� ' ��)-FF1,�3�� &�

�� ,�6��1�0 ��!�� .�.��

��,�6�� 1�0 �� .�.��

��&��

-F�� !��"�� &�)�� ,�

�� ;�� ;�� 9 �� E( �� <�� #$� �� #�!��

�� ;�� #$��

�� 2��E��6��

�-%��#�� "��C�� !��3�� "��G��

�� )��%I1,�� K��?�!K��

� � �� @��3�� )��%I1,�

�� )��%I1,�

N<��C� ��O�

:�� #� ��

C�� #$��

<� ��;��

;�� E�#��$$��)��,� ��

;�� E#��$$��)�� ,� ��

;�� <��I��9��

@��P�� 9 �� ( ��

C��E�� A$��@ ��

��

<��C�� !��)��

��

��,�

��

C��;�$ ��

C��EO��@��O� ��

C��C��9��

��

Ich



--��#��C��G�� #��

�� *� �

��*�� +�� $�� , �� *��(��+�

��%,�;�� )2 �� I#Q��,*�� G �� G ��

��-,�9��E� �� *�� G �� G ��

��/,��*� � � �� G �� G ��

��+,�#��EI ��*� � �� G �� G ��

��1,��!��*� � �� G �� G ��

��5,�6�� 2��)��R��,* �G �� G ��

��7,�C�� 2 �� *� � �� G �� G ��

��-/�� I��#��&�

�� >1�=�� "�� ;��C ��

�� C��*�.......................................�

-+�� I�� #��E��E��&�

�� >1�=�� "�� ;��C ��

�� C��*�.......................................�

-1��C��@��#�� $�!�� &�)��(��,�

�� ;�� @��!�� #$�� @�� <�� ( ��

�� "�� 3�� C��*�.......................................�

-5�� #�� )#��,&.............................................�

��

-��(��+��+� �� .��/��0�� (��

+��

�

-7��@��&�� !�� C� ��

-=��C��&� ��.�.��' �� C� ��

->�� 2��"��8 �� )#��,&��.�.�2��)�,�

/F��#��#�� #��EC�� )� ��I,�C�� C�� #��!��E��!��

�� "��6�� 8 �� EI� �� #��EC�� #��*�................�

/%��#��2�� "�� 29�*�..�..�..�..�..�� C� ��

/-�� ,�� #�� C�� &� �C�� 2 ��

�� @��!��$ ��E<�� C� ��

��,�@��P��@��$$��*��...�2��)�,�� ....�<��)��,��%+�' ��

�

1��+��#��0��%� �� 0 ,��0��2��(��

&��333�

Ich



��

Ich




Outbound Flight Dear air passenger,

the International University of Applied Sciences Bad Honnef-Bonn conducts a tourist survey in

collaboration with the administration of the Balearic Islands and Germanwings. The intention is

to even better oblige the wishes and desires of Mallorca tourists in the future. Therefore we

would love to obtain your opinion. We would appreciate if you could take a couple of minutes.

__________________________________________________________________________

1. Have you been in Mallorca before? Yes No (If no, continue with question 4)

2. How often have you been in Mallorca? ____ times 3. How often in the last 3 years? ____ times 4. How many days are you going to stay in Mallorca this time? ______ days

Booking behaviour:

5. What did you book (check as many as apply)? Package tour only flight

Accomodation (if package holiday, please continue with question 8)

6. How did you book this flight? Travel agency Internet By telephone

Directly over the counter

7. Did you book one or more of the following services via the internet? Accomodation Rental car Insurance Travel guide Admission tickets

None of these

8. Do you fly more frequently since there are cheaper flights to Mallorca? Yes No Cannot say

9. Have you been flying with one of the following airlines to Mallorca before? (Please check as many as apply)

Germanwings Hapag Lloyd Flugdienst Hapag Lloyd Express Air Berlin Deutsche

BA LTU Aero Llyod Condor/Thomas Cook Miscellaneous __________

10. When did you book your current holiday? Less than 1 month ago 1-3 months ago 3-6 months ago

more than 6 months ago

11. What did you pay for your outbound flight today? ______ Euro Roundtrip (outbound flight and return flight together) ______ Euro

12. How satisfied are you with this price? Price for outbound flight: ____, price for roundtrip: ___

(scale from 1 to 5, 1=very satisfied, 5=very unsatisfied)

Locality/Accomodation: 13. Where in Mallorca are you going to stay (village)? ___________________________________ 14. Where did you stay during your last Mallorca holiday (if you have been here before)? _____________________________________________________________________________


15. a) Which type of accommodation are you staying in?

Hotel Club Hostal/Guesthouse Holiday home/apartment

Finca/Agroturismo Own house/apartment

Friends’/Relatives’ house/apartment Not applicable

b) If hotel or club, which category? * ** *** **** ***** Not applicable

c) Kind of board: Self-catering Breakfast Half-board Full-board

All inclusive

Rental car:

16. Have you rented a car or are you going to rent a car? Yes no

If yes, for how long have you rented the car? ____ days

17. How did you book your rental car (or will book)? Internet Travel agency

Rental at holiday place Rental at airport Relatives’ car

Miscellaneous

General travel behaviour: 18. How often have you been on holiday last year (2004)?

a) Holidays 5 days or more ____ times not applicable

b) Holidays less than 5 days ____ times not applicable

c) In what country/countries did you spend your holiday? ___________

c) Was your last core holiday a package holiday or did you put it together yourself?

Package tour Flight, accommodation etc. self-arranged not applicable

19. How often are you probably going to be on holiday this year (2005)?

a) Holidays 5 days or more ____ times cannot say yet

b) Holidays less than 5 days ____ times cannot say yet

Mallorca 20. What do you particularly like in Mallorca? (Check as many as apply)

Accessibility Food Nature People Culture

Language Beaches Leisure time facilities Shopping facilities

Sports facilities Sightseeing Price/low priced holiday destination


21. Please indicate how important the following activities are for you generally during holidays and

especially during your current Mallorca holiday. (scale of 1 to 5, 1=very important, 5=of no

importance)

Generally during

holidays (scale of 1 to 5)

Current Mallorca holiday

(scale of 1 to 5)

Not

applicable

Be at the beach/bathing Go in for sports Get in contact with locals Shopping (generally) Shopping (products that are characteristic

of the region)

Get to know local lifestyle and culture Enjoy scenery and nature Go to local/typical restaurants and bars Cultural activities (e.g. museums, concerts,

exhibitions, visit historic buildings)

Relax Trips/Excursions/Explore the area Go out at night, meet people

22. Please indicate for the following choice of products if you know them and if you have

used/visited them before? Knowledge Have used/visited Not applicable

1) Train trip (Palma-Soller) yes no yes no

2) Leather manufacture/factory outlet

yes no yes no

3) Museums yes no yes no

4) Boat trips yes no yes no

5) Weekly markets yes no yes no

6) Products that are characteristic for the region (e.g. oil, wine)

yes no yes no

7) Historic city of Palma yes no yes no

23. What Mallorca media do you know?

Mallorca 95.8 Das Inselradio Mallorca Zeitung Mallorca Magazin El Aviso

www.mallorca.de Others:

24. What Mallorca media have you used yourself?




25. For which reason do you mainly fly to Mallorca this time (please only one response)?

Recreation Business Sports Health Culture Nature

Visit friends Have own accommodation in Mallorca Miscellaneous

26. What do you like best in Mallorca? ________________________________________

The following information is of course optional, however exceptionally important for the statistical evaluation:

27. Sex female male

28. Age _____ years

29. How many persons live in your household (including yourself)? ___ persons

30. Are you student/apprentice, labourer, clerk, own business/freelancer, retired,

house wife/husband, university/college student, miscellaneous

31. Your postal code? __________

32. a) Who are you in Mallorca with? By yourself family friends partner

co-worker n.a.

b) Size of the group altogether: ____ persons, thereof ___ children under 14.

Thank you very much for your assistance and have a nice stay in Mallorca!

��

�� !��"��!��"��#�� $��

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!%��#��"��%�� &� �'�� (�� )��*'�+��#�� #��,-�!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

.��"�� &� � �'�� (��)��/�0��1-�

2�� #��)��-�� &�� 3�3��

4�� 4�'��&� � � � 3�3��

1�� 5��#�� &� 3�3�� 5��

�

��"� ��#

6��#��&�)��-�� 7�� (�� 8��

�# ��333333333333333333��)��$�� %,-�

9��#��&�� :�� "�� 5�� #��

;��%��#��"�� <��&�

�� 8�� =�� :�� >�� ?��

@��#�� !��&� �'�� (�� ?��

A��#��#�� &�)��

��B��-�

�� C�� %�$��<�D �� %�$��<� D��>E$�� F�� F�

�� <58�� F�� <� D�� G � �H5� ��G �� # ��333333333333333333333333333333333333�

.I��#��"��/��:��&��

�� J�.�� .K4�� 4K9�� L�9��

..��#��"��:��&��%��:��0�3�3�3�>��

��%�K��:��0�3�3�3�>��

.2��#��7��&��7��:��0�33333333�)��.K6-�

��7��%�K��:��0��3333333333�)��.K6-��&M��B��'M��

( ��)��*�� )��#

.4��(��#��N�� "��8��&�3333333333333333333333333�

.1��-��#��F��"��8�� 0�

�� % �� HG�� % ��H7�� HK� �� HF��

��%��H�� %�� H��H=�� F��

��-��% �� :�� ?�� &�� O� �OO� �OOO� �OOOO� �OOOOO�� F��

��-�F��=��$��0�� #�� $�� %7�� =7� �� F��

.6��#�� .K6��#��"��8��&��333�( �� ?��F��

+�� #

.9��%��#��&�� '�� (��,��(��.A�

��-��#��&�� 3�3��5��

Ich


Return Flight Interview

.;��#��"��&� �"�� :�� =��

�� =�� F� � �=�� # ��33333333333333333333333333�

.@�� #��)��!��-��"��&��3�3�3�3�3��

.��"� ��#

.A�� #��'��)2II1-�8��&��

��-�:��6�5��!�� 3�3�� F��

��-�:��6�5�� 3�3�� F��

�

.��/�"� ��#

2I�� #��)0 �$� ��-��"��/�� &�

��-�? ��)��F�� 8��F��

��N�-0� � � ��33�33�33�33�>�� +I+�>8:� � ��F��

��-��"��F��&�)��#�� ?��

��*I+�>8:�� !��F��-�

7�� +I+�>8:��33�33�33�33�>�� ?��F�� +I+�>8:� �33�33�33�33�>��

� (�� +I+�>8:��33�33�33�33�>�� #$ �� !�� +I+�>8:� �33�33�33�33�>��

� (��8�� +I+�>8:��33�33�33�33�>�� >��H#� $$�� +I+�>8:� �33�33�33�33�>��

� GP��H:�� +I+�>8:��33�33�33�33�>�� +I+�>8:� �33�33�33�33�>��

� <��H�� +I+�>8:��33�33�33�33�>�� # ��033333333 �+I+�>8:�� 33�33�33�33�>�� $��=��$��+�� #

2.��%��$��"�� &�)��(��-�

�� 8�� ?�� (�� #$�� C��!��

�� >�� #$ ��

22��#�� F�� /�� #��

��0�

��# ��/��*��1�# ��.��/�

��.-�>��)7��K#Q��-0�� /�� /��

��2-�<��H�� 0�� /�� /��

��4-��0� � � �� /�� /��

��1-�#�� HK��0� � �� /�� /��

��6-�� !��0� � �� /�� /��

��9-�:�� 7� ��)��R��-0 �/�� /��

��;-�F�� 7��0� � �� /�� /��

�

24��"�� &�)��-�

�� >�� >�� <��H(�� ?�� #$�� #�!��

�� 7��H��:�� # ��33333333333333333333333333333333333333�

�

26��F�� "�� &�333333333333333333333333333333333333333333333333333�

�

29��%!��#��F�� !�� &� '�� (��-��3333333333333333�

��333333333333333333333333333333333333333333333333333333333333333333333333333333333333333�

Ich



2;�� K��#��&�

�� A6�@��"�� >��F ��

�� F��0�333333333333333333333333333333333333333�

�

2@�� K��#��H��H��&�

�� A6�@��"�� >��F ��

�� F��0�333333333333333333333333333333333333333�

�

2A��?��#�� !��'�� &�

�� '�� (�� ?��

4I��#�� #��"�� &�� ( ��

�

4.��#��/�� 1�� #�� C�� 0��

��)��.K6-��&M��B�'M��

� ��).�S�6-� ?��F��=�� 7�� C�� #�� ?�� #�!�� N�� <�� # ��03333333333333� � �

2��.��/��/��"� ��3�� 4��5 ��.��

/�� #

42��C��&�� !�� F��

44��F��&� ��3�3��'�� F��

41�� 7�� "��%��)#�� -&��33�33�7�� )�-�

46��#��#�� #��HF�� )��K-�F�� F�� #��!��H��!��

�"��:�� %��HK�� #��HF�� # ��0�3333333333333�

49��#��7 ��"�� 7<�0�33�33�33�33�33�� F��

4;��-��#��F�� &� �F�� 7��

�� C��!��$��H? �� F��

��-�C��T��C��$$��0��333�7�� )�-�� 3333�?��)��-��.1�'��

6� /��)��"��5 �� 5�1�� 5��

�� 7

Ich



��

Ich




Return Flight Dear air passenger,

the International University of Applied Sciences Bad Honnef-Bonn conducts a tourist survey in

collaboration with the administration of the Balearic Islands and Germanwings. The intention is

to even better oblige the wishes and desires of Mallorca tourists in the future. Therefore we

would love to obtain your opinion. We would appreciate if you could take a couple of minutes.

Have you already participated in this survey during your outbound flight? Yes no (If yes, please fill in

this questionnaire anyhow, it is different from the one you already know).

1. Was this your first stay in Mallorca? Yes No (If no, continue with question 4)

2. How often have you been in Mallorca? ____ times 3. How often in the last 3 years? ____ times 4. How many days are you going to stay in Mallorca this time? ______ days

Booking behaviour:

5. What did you book (check as many as apply)? Package tour only flight

Accomodation (if package holiday, please continue with question 8)

6. How did you book this flight? Travel agency Internet By telephone

Directly over the counter


None of these





10. When did you book your current holiday? Less than 1 month ago 1-3 months ago 3-6 months ago


11. What did you pay for your outbound flight today? ______ Euro Roundtrip (outbound flight and return flight together) ______ Euro

12. How satisfied are you with this price? Price for outbound flight: ____, price for roundtrip: ___

(scale from 1 to 5, 1=very satisfied, 5=very unsatisfied)

Locality/Accomodation: 13. Where in Mallorca are you going to stay (village)? ___________________________________


14. a) Which type of accommodation are you staying in?

Hotel Club Hostal/Guesthouse Holiday home/apartment

Finca/Agroturismo Own house/apartment

Friends’/Relatives’ house/apartment Not applicable

b) If hotel or club, which category? * ** *** **** ***** Not applicable

c) Kind of board: Self-catering Breakfast Half-board Full-board

All inclusive

15. On a scale from 1 to 5, how satisfied have you been with your accommodation (1=very satisfied,

5=very unsatisfied)? ____ n.a.

Rental car:

16. Did you rent a car? Yes no (If no, please continue with question 19)

If yes, for how long have you had the rental car? ____ days

17. How did you book your rental car? Internet Travel agency

Rental at holiday place Rental at airport Relatives’ car Miscellaneous

18. How many kilometers did you drive with your rental car (approximately)? ___________ km




Money spending behaviour: 20. How much did you spend approximately altogether (per person) for your current Mallorca holiday?

a) Expenses altogether (all expenses including flight, accommodation, additional costs at holiday

place): appr. _______ Euro “0” Euro not applicable

b) If applicable, how do your expenses divide up approximately (if you have not spent any money in

one or more categories please check 0 Euro, otherwise please fill in your estimation)

Package holiday Only flight Only accommodation Café/Restaurant Groceries

Cultural offers Sports Shopping Car rental Miscellaneous

Mallorca 21. For which reason did you mainly fly to Mallorca (please only one response)?

Recreation Wellness Culture Nature Language course

business shopping sightseeing Sports


22. Please indicate for the following choice of products if you know them and if you have used/visited

them before? Knowledge Have used/visited Not applicable



yes no yes no





yes no yes no


23. What do you particularly like in Mallorca? (Check as many as apply)


Language Beaches Leisure time facilities

Price/low priced holiday destination Miscellaneous

25. What do you miss in Mallorca?

____________________________________________________________________________

26. Are there any additional activities you would have liked to do in Mallorca? Yes no, if

yes what activities? __________________________________________________________







29. Can you imagine coming again to Mallorca within the next 3 years? Yes no

cannot say

30. In terms of school grades, how would you rate your current Mallorca holiday? _______


31. Please indicate how satisfied you were with the following features in Mallorca (scale of 1 to 5, 1=very

satisfied, 5=very unsatisfied)

scale of 1 to 5 Not applicable

Public transport (bus, train) Price level Recreation offers Gastronomy Cleanliness Cultural offers Beaches Cities Scenery Miscellaneous


32. Sex female male

33. Age _____ years




36. Your postal code? ___

37. a) Who did you spend your Mallorca holiday with? By yourself family friends

partner co-worker n.a.


Thank you very much for your assistance and have a nice trip home!

Interviewername: _________________________________ Datum: __ __. Mai 2005 Uhrzeit:__________ Standort:___________________________

��

�� !��"��

"��#�� $��$��%��"��$��$��&��'��

�� "�� (��$��&��)��)��

��(��$��)�� '��*��+��(��

��,��*�� (��-� � � � �.�� /�� 0��!��$��1#��

2�� *��#��$�� (��-�� 3�3�� (��

4�� #��&��4�.��-� � � 3�3�� (��

1��$��*��(�� (��-� � 3�3�� $�� $��5�� 0�3�3��3�3��2667�

�

��

7��8��!��$�� *��$� ��$��-� 3333333333333333333333333333333333333333333333333�

��!��$$�� -� � 3333333333333333333333333333333333333333333333333�

9� ��*�� 5��$��-��(�� $��)$��#�

�� /��!��$��

�� *��$��3333333333333333333333333333333333333333333333333333333333333333333333333333�

:� ��*�� #�!��$;��$��-�

�� <�� *��

=��*�� $��>��$��$��-�

�� (��$�� 8��$�� <�� ?�� @��$��

A��!��$��*��'��$��$��!��$��(��$��'�� $��-� �.�� /�� @��$��

,6��*��*�� $��!��$$�� (��$� ��$��-��"��

��&��B�(�� $��)$��#�

�� C��$�� +�$�>�D��!��$�� +�$�>��D��?E+�� 5��"�� "5�

�� >�� 5��>��D�� F��;��F�� *��$��3333333333333333333333333333333333333�

,,� ��*�� %�&�$��!��$��&�� $��-��

��G,�(�� ,H4�(�� 4H9�(�� I9�(��

�

��

,2��/��*��J�� (��'�� -333333333333333333333333�

,4�� J��*�� &��(�� (��$��#-�

��3333333333333333333333333333333333333333333333333333333333333333333333333333333333333333��

,1��#��*��5�� (��-�

�� !��;F�� ;�� !��;H��$� �!��;5$��

��$��;��$��$� ��!��;"��;8�� 5�$��

��#��<��'��@��$��-�� K� �KK� �KKK� �KKKK� �KKKKK� ��5�$��

��#�5��8��+ ��$��$0�� *��+ ��$��$� �� !�� 8�� 5��

Ich


Face-to-Face Interview

��

,7��*��(��$��$��'��*��-�� .�� /��

��'� ��$��*��(��$��$��-��3�3��$��

,9��*�� (��$��$��&�� #-�� <��

�� 8��!�� 8��!��$�� $��5��8�� *��$��

��

,:�� *��&��.��2661#��$��-��

��#�<��7��$��$�� 3�3�(�� 5�$��

��#�<��&��7��$�� 3�3�(�� 5�$��

��!��"��I7��$�#��03333333333333333333333333333333�

3333333333333333333333333333333333333333333333333333333333333333333333333333333333333333333�

��#�� &��.��+��&��$��-�

�� $�� !��$'�� '��&��$�� 5�$��

,=�� *��.��2667#��-�

��#�<��7��$��$�� 3�3�(�� $��

��#�<��&��7��$�� 3�3�(�� $��

��

,A��*��$��"��#��#� �� %�&�$��(�� $�$��-�

��#�@��$��5�$��&��$��'��!��$'�� '��5�$��

��J�#0� � � ��33�33�33�33�?�� L6L�?�<� � ��5�$��

��#�!��&�� '�� 5�$�� -��*��@��$��

��$�$��'��M6L�?�<��&��'��$� ��5�$��$��#�

�� L6L�?�<��33�33�33�33�?��@��5�$�� L6L�?�<� �33�33�33�33�?��

� /��!��$� � �L6L�?�<��33�33�33�33�?��*+�� L6L�?�<� �33�33�33�33�?��

� /�� L6L�?�<��33�33�33�33�?��?�� ;*��++��$� �L6L�?�<� �33�33�33�33�?��

� FN �;<�� L6L�?�<��33�33�33�33�?��(��$�� L6L�?�<� �33�33�33�33�?��

� >��;�� L6L�?�<��33�33�33�33�?��*��$�033333333 �L6L�?�<�� 33�33�33�33�?��+��8��+ ��$��$�

�

��

26��$�� *��'��$��D+��&��-�

�� $�� $�� $��$�� $�

2,��$��D+��*��-�333333333333333333333333333333333333333333333333�

22��$�� *��'��5�$��&��&��-��

�� $�� $�� $��$�� $�

24��5�$�� (��*��-3333333333333333333333333333333333333333333�

��33333333333333333333333333333333333333333333333333333333333333333333333333333333333333�

21��*��*��O��D��*��O��'�"��$��#�� (��-�

�� $�� $�� $��

27��*��O��D��-�3333333333333333333333333333333333333333333333�

��33333333333333333333333333333333333333333333333333333333333333333333333333333333333333�

Ich



29��"��$��*��'�� $��5��$�� %�&�$��

��(��"��$�,H7#��$P��$B�%P��$�

� � �� C��"��$�,H7#�

��(��"��$�,H7#�

M@��5�$��L�

8��*��$��'�"��

5��*+��

@��?��

?�� ;�*��++��$��#� ��

?�� ;*��++��$�� #� ��

?��@��H��>��

C��O��>�� /��

5� ��;��$��D+��C��

��

@��5��

��

��#�

��

5��'�?�+��$� ��

5� ��$�;L��C�$��L� ��

5��5�$��'�>��

�

2:��#�"��*��J��'��*��<�� %�&�$��(��

��&��*��%�&��'��*��0�3333333333333333333333333333�

��!��@5<�?��(��J�� 0��/��$��>��&��

��5� ��&��%��$��5� ��$��J��#�

��#��*��J��-��(�� $��)$��#��

��!��5�� @�� J��Q�

��*��;*��?�� &�"��$��$��;@��'�*+�&��$��(�*+��

��)�F� R;<��*��$�0�33333333333333333333333333333333333333333333333333333333�

2=��5��C��*��#��+��(��<��&��$��-�

�� ?��$�� C�� *+�� C�� @�� /�� !��

�� $�� (�� 5��0�3333333333333333333333333333333333333�

2A��$� �� $�� (��-��(�� $��)$��#�

�� ?�� ?�� >��;/�� (�� @�� *+�� *��

�� !��&��$�� ?�� )$�� *+��)$�� "��$��$�)$��

46��)$��*��+��)�� (��-�3333333333333333333333333333333333333333333333�

��333333333333333333333333333333333333333333333333333333333333333333333� ��5�$��

4,��5�$�� (��-�333333333333333333333333333333333333333333333333333�

��333333333333333333333333333333333333333333333333333333333333333333333� ��5�$��

42��*��0�5��$��5� ��J�� *��+��)��<��

��(��-��3�3��$��

44��(��H(��*��-�

�� (��A7�=�� (��$�� (��(�$�&�� ?��5��

�� 5��0�333333333333333333333333333333333333333�

Ich



41��(��H(��*��$��&;$��;$��)�-�

�� (��A7�=�� (��$�� (��(�$�&�� ?��5��

�� 5��0�333333333333333333333333333333333333333�

47��"��$��*�� $��5�$��%��'��*��&��

��$��&��0�

��*�� +�� *��

��,#�?�� H*S��#0�� %�� %��

��2#�>��;!�� 0�� %�� %��

��4#�(��0� � � �� %�� %��

��1#�*�� ;H�� $�0� � �� %�� %��

��7#��0� � �� %�� %��

��9#�<�$��&�"�T��'��'��#0 �%�� %��

��:#�5��0� � �� %�� %��

49��"��$��*��%��'��+��,��*�� $��C�$�� (��0��"��$�,H7#��$P��&� ��B�%P��&� ��B�!��

��-��.��

� "��$��,H7#� @��5�$��T ��"�'�"��#� � �� !��&��$�� C��$�� *�� @��$�� *�� J�� >�� *��$�0333333333333� � �

4:�� $�� "�� *��+��&��<��&��-�

�� &�� (�� !�� 5�$��

��"��$�,H7#��$P��'�%P��

4=��@)��*��'��.��(��&��-�

�� .�� /�� @��$��

'��,�� ,�� /�,��0�,��1��,�� ,��

4A��C��-�� 5�$��

16��5��-� ��3�3��.�� 5�$��

1,�� *��$��#-��3�3��#�

12��*��*�� *��;5&�� !��H#�5�� 5�$�� *��$; �� $�

�� <�� ;H�� *��;5��$�� *��$�0�333333333333333�

14��"��*��&�� 0��>�0�33�33�33�33�33�� 5�$��

��#�(��$��*��5� ��-� �5�� !�� !��

�� C�� +��;@��$�� 5�$��

��#�C�)O��C��++��$�0��333��#'��3333�@��#��,1�.��

2�� ,��1��!��1�+��,��1��!��3��4�

!56( 7!(2( ��,��#�� ,�� 1��-*� 6(.��444�

Ich



��

��

��

��

��

�� !�� "�

��#$�� %#��& ��'� $�� %!��#(�� #(��

��)��*%+��#�� ,�--------------------------------------------------------�

�+.$�� ,��.��

��

��'��

Ich




Face-to-Face Interview Dear Mallorca guest,

the University of Trier conducts a tourist survey in collaboration with Prof. Dr. Helmut Wachowiak

(International University of Applied Sciences Bad Honnef-Bonn). This project is supported by the

administration of the Balearic Islands. The intention is to even better oblige the wishes and

desires of Mallorca tourists in the future. Therefore we would love to obtain your opinion. We

would appreciate if you could take a couple of minutes.

__________________________________________________________________________

1. Have you been in Mallorca before? Yes No (If no, continue with question 4) 2. How often have you been in Mallorca? ____ times 3. How often in the last 3 years? ____ times 4. How many days are you going to stay in Mallorca this time? __ days. Day of arrival: __2005

Booking behaviour: 5. From which airport did you depart? ______________________________________________ … and with which airline? _______________________________________________

6. What did you book before your departure? (Check as many as apply) Package holiday Only flight Accomodation

Miscellaneous: _______________________________________________________________

7. How did you book your flight/package holiday? Travel agency Internet By telephone Directly over the counter


None of these





11. When did you book your current flight or package holiday? Less than 1 month ago 1-3 months ago 3-6 months ago


Locality/Accomodation: 12. Where in Mallorca are you going to stay (village)?___________________________________ 13. Where did you stay during your last Mallorca holiday (if you have been here before)?_______ 14. a) Which type of accommodation are you staying in?


Hotel Club Hostal/Guesthouse Holiday home/apartment Finca/Agroturismo

Own house/apartment Friends’/Relatives’ house/apartment Not applicable

b) If hotel or club, which category? * ** *** **** *****…. Not applicable

c) Kind of board: Self-catering Breakfast Half-board Full-board All inclusive

Rental car/Generalities:

15. Have you rented a car or are you going to rent a car? Yes no

If yes, for how long have you rented the car? ____ days

16. How did you book your rental car (or will book)? Internet Travel agency

Rental at holiday place Rental at airport Relatives’ car Miscellaneous




Interview instruction: Please list the countries (>5 days)

____________________________________________________________________________

c) Was your last core holiday a package holiday or did you put it together yourself?

Package tour Flight, accommodation etc. self-arranged not applicable

18. How often are you probably going to be on holiday this year?

a) Holidays 5 days or more ____ times cannot say yet

b) Holidays less than 5 days ____ times cannot say yet

Money spending behaviour: 19. How much did you spend approximately altogether (per person) for your current Mallorca holiday?

a) Expenses altogether (all expenses including flight, accommodation, additional costs at holiday

place): appr. _______ Euro “0” Euro not applicable

b) If applicable, how do your expenses divide up approximately (if you have not spent any money in

one or more categories please check 0 Euro, otherwise please fill in your estimation)

Package holiday Only flight Only accommodation Café/Restaurant Groceries

Cultural offers Sports Shopping Car rental Miscellaneous

Mallorca: 20. How important is it for you to get to know products that are characteristic of the region?

very important important neither…nor not very important


of no importance

21. What products that are characteristic of the region do you know? _______________________

22. How important is it for you to make use of cultural offers?

very important important neither…nor not very important

of no importance

23. What cultural offers in Mallorca do you know? ______________________________________

24. How would you judge the road network (road conditions, signposting etc.)

very good good neither…nor bad very bad cannot say

25. How could the road network be improved?_________________________________________

26. Please indicate how important the following activities are for you generally during holidays and

especially during your current Mallorca holiday. (scale of 1 to 5, 1=very important, 5=of no importance)

Generally during

holidays (scale of 1 to 5)

Current Mallorca holiday

(scale of 1 to 5)

Not

applicable

Be at the beach/bathing Go in for sports Get in contact with locals Shopping (generally) Shopping (products that are characteristic

of the region)

Get to know local lifestyle and culture Enjoy scenery and nature Go to local/typical restaurants and bars Cultural activities (e.g. museums, concerts,

exhibitions, visit historic buildings)

Relax Trips/Excursions/Explore the area Go out at night, meet people

27. a) Please indicate the places you visited during your current Mallorca holiday so far and those that

you plan to visit: _____________________________________________________________

Interview instruction: (Have Mallorca map ready. Draw a line between residence and visited place.

Circle visited place.)

b) What did you do at this place? (Check as many as apply)

(Interview instruction: Note activity down on map next to place.)

A) Bathing/Beach B) Shopping C) Sightseeing/Culture D) Took a walk E) Sports F)

Café/Restaurant S) Miscellaneous ________________________________________________

28. Why did you choose (this time) Mallorca as holiday destination?


Recreation Business Sports Health Culture Nature \

Friends Have own accommodation in Mallorca Miscellaneous

29. What do you particularly like in Mallorca? (Check as many as apply)


Language Beaches Leisure time facilities Shopping facilities

Sports facilities Sightseeing

30. What do you like best in Mallorca? ____________________ n.a.

31. What do you miss in Mallorca? _______________________ n.a.

32. What do you think? How many days minimum stay are necessary to make a Mallorca holiday

worthwhile? ______ days



www.mallorca.de Others: ____________________________



www.mallorca.de Others: ____________________________

35. Please indicate for the following choice of products if you know them and if you have used/visited

them before? Knowledge Have used/visited Not applicable



yes no yes no





yes no yes no



36. Please indicate how satisfied you are with the following features in Mallorca (scale of 1 to 5, 1=very

satisfied, 5=very unsatisfied)

scale of 1 to 5 Not applicable

Public transport (bus, train) Price level Recreation offers Gastronomy Cleanliness Cultural offers Beaches Cities Scenery Miscellaneous

37. How interesting are the following Balearic islands for you as potential holiday destinations (scale from

1 to 5)?

Ibiza Menorca Formentera n.a.

38. Can you imagine coming again to Mallorca within the next 3 years? Yes no cannot say


39. Sex female male

40. Age _____ years




43. Your postal code?

a) Who are you in Mallorca with? By yourself family friends partner

co-worker n.a.


Thank you very much for your assistance and have a nice stay in Mallorca!

Documents

GIS and Tourism Research - Home - SPINlab Vrije ... · PDF fileGIS and Tourist Consumer Research ... Geographical Information Systems ... Only the market research appears to be poorly