The empirical study of automotive telematics acceptance in ... › 5cf9 › 7fd0b4fc4c7c... · the automotive telematics market is poised for explosive growth (Zhao, 2002). Automotive

50 Int. J. Mobile Communications, Vol. 7, No. 1, 2009

Copyright © 2009 Inderscience Enterprises Ltd.

The empirical study of automotive telematics acceptance in Taiwan: comparing three Technology Acceptance Models

Huei-Huang Chen* Department of Information Management, Tatung University, 40 Chungshan North Road, Section 3, Taipei 104, Taiwan, ROC E-mail: [email protected] *Corresponding author

Shih-Chih ChenDepartment of Computer Science and Engineering, Tatung University, 40 Chungshan North Road, Section 3, Taipei 104, Taiwan, ROC E-mail: [email protected]

Abstract: Automotive telematics comprises the applications of Global Positioning System (GPS) navigation, multimedia entertainment, wireless communications and automatic driving assistance systems. This study examined users’ acceptance of automotive telematics. Following the theory comparison approach, we evaluated including the Technology Acceptance Model (TAM), the Theory of Planned Behaviour (TPB) and Combined Technology Acceptance Model (TAM)-TPB model (C-TAM-TPB), that could explain users’ automotive telematics acceptance decisions. The respective models were evaluated using web-based survey data collected from 345 users about their perceptions of automotive telematics. Overall, the results revealed that the effect of Perceived Ease Of Use, Attitude and Perceived Behavioural Control were very important but that usefulness and Subjective Norm did not influence an individual’s Behavioural Intention. The implications of this study were also discussed.

Keywords: automotive telematics; Combined Technology Acceptance Model; CTAM; Combined TAM-TPB; C-TAM-TPB; mobile communications; Technology Acceptance Model; TAM; telematics; Theory of Planned Behaviour; TPB.

Reference to this paper should be made as follows: Chen, H-H. and Chen, S-C. (2009) ‘The empirical study of automotive telematics acceptance in Taiwan: comparing three Technology Acceptance Models’, Int. J. Mobile Communications, Vol. 7, No. 1, pp.50–65.

The empirical study of automotive telematics acceptance in Taiwan 51

Biographical notes: Huei-Huang Chen received his PhD in Computer Science from the University of Illinois at Urbana-Champaign in 1984. He is the Dean of and Professor at the Tatung University’s Management and Design College, Taiwan. He was with the information systems division of Tatung Co. for 14 years. His research interests include data warehousing and business intelligence, information technology applications for e-businesses and m-commerce.

Shih-Chih Chen is a Doctoral Candidate at the Tatung University in Taiwan. His current research interests include quantitative analysis, mobile commerce and managerial issues of emerging technologies.

1 Introduction

It was a luxurious facility only seen in few top cars to control the operation of all systems in the vehicle through computing before 1980s. However, the relationship between the automobile design, production and the technology has been closed. Since 1990s, an automobile had been changed to an intelligent digital vehicle. Information services provided to such a vehicle over wireless communication networks are referred to as automotive telematics. There were two reasons causing such trend. First, due to the automobile manufacturers began greatly relying on the computing for the mechanical structure design of car body. The other was, due to the demand of car drivers and passengers for in-car mobile communications and satellite navigation services were steadily increasing.

Telematics, a union of telecommunications and informatics for the transport sector, is expected to revolutionise automobile industries. With more than 40 million vehicles sold worldwide each year and more than 935 million cellular customers by the end of 2001, the automotive telematics market is poised for explosive growth (Zhao, 2002). Automotive telematics and navigation – including both hardware and services – will experience strong growth in several regions, generating in total revenues of 38.3 billion dollars in 2011(ABI Research, 2006a,b). Strategy Analytics (2001) estimated that by 2007, approximately 55% of all new cars will have a telematics-capable terminal, as compared to approximately 7.5% in 2000 (Telematics Research Group, 2002). For modern people, their cars are almost the place where they take longest time to stay in addition to their homes and offices. Therefore, it seems that one-way information receiving cannot satisfy the users. What the user requests is total information control. When the car becomes a black hole for information transmission, the longer the time is for the user to stay in the car, the longer the time is for him/her being unable to access the information. Telematics integrates such functions as internet, communication, satellite positioning, safety monitoring and entertainment to provide the service of in-car information transmission and exchange to satisfy the demand for in-car mobile computing. One of the most popular cited examples is OnStar, a car navigation system that bases its geographical information on a satellite-based infrastructure (OnStar, 2006; Junglas, 2007).

52 H-H. Chen and S-C. Chen

Safety and convenience is and will always be the core for vehicle design, ubiquitous context information affecting safety and convenience is hidden in three entities: vehicle, people and environment (Zhang, Wang and Hackbarth, 2004). Most telematics research has focused on the technology developments and functional applications essential to its success (Melvin, 1999; Park et al., 2003; Tsou, 2004; Zhang, Wang and Hackbarth, 2004; Tanizaki, Maruyama and Shimada, 2005). On the other hand, lot of recent research has applied Technology Acceptance Model (TAM) to understand mobile commerce related use (Chau and Hu, 2002; Yang, 2005; Han et al., 2006; Hung, Hwang and Hsieh, 2007; Junglas, 2007; Lee and Jun, 2007; Lu, Wang and Yu, 2007; Chen, 2008), but few for automotive telematics. This study examined and compared automotive telematics system acceptance to the TAM (Davis, 1989; Davis, Bagozzi and Warshaw, 1989) the Theory of Planned Behaviour (TPB; Ajzen, 1991) and the integrated model (Taylor and Todd, 1995a). Using the responses from the web-based survey that involved 345 users, this research outcome evaluated and compared the extent to which the respective models could explain individual user’s intention to use automotive telematics technology. The causal paths specified by each model were also examined.

2 Telematics and its technology adoption

2.1 Telematics and automotive telematics

Information Technology (IT) has revolutionised automobiles in many ways. Today, vehicles incorporate a high level of computer technology to meet the needs of consumers for safety, environmental protection, communications and even entertainment needs. In fact, electronic equipment now controls more than 86% of all systems in a typical vehicle (Alliance of Automobile Manufacturers, 2006). The Alliance of Automobile Manufacturers defines it as below:

“integrating the automobile electronics with the communication technology to provide the driver driving guidance and information exchange services to improve personal driving safety and security.”

The automotive industry quickly adopted the term to describe any system that provides location-based services for a vehicle over the wireless telecommunications network. In other words, generally, telematics refers to any automotive system that combines wireless technology with location-based services. One of the most important information services in automotive telematics is the car navigation service, which is a service that calculates a route and guides the user to follow the route (Zhao, 2002; Park et al., 2003).

An automotive telematics platform consists of five key components (Aloi, Alsliety and Akos, 2007):

1 positioning device

2 communication device

3 entertainment device

4 vehicle interface

5 user interface.


Besides, such as in-vehicle displays for drivers, digital maps with dynamic route planning and navigation, are the popular applications for automotive telematics (Finn and Breen, 1996). It connects the On Board Unit (OBU), the carrier and the service content provider with the purpose of providing in-vehicle communication and information services. The method for service provision is to ensure the position of the car through the global positioning system (GPS), to use OBU to collect driving information and accept the user demand and to satisfy the demand of users for information transmission and exchange through the link between the wireless communication network and the service content. Lots of automotive telematics industries in development are focused on either increasing driver safety or giving the driver more pleasurable options, instead of taking the driver’s attention on the road.

2.2 Theoretical background

The development of TPB is originally based on the Theory of Reasoned Action (TRA) (Fishbein & Ajzen, 1975; Ajzen & Fishbein, 1980). TPB hypothesised that intention to perform a behaviour is based on: attitude, Subjective Norms (SN) and Perceived Behavioural Control (PBC) (Ajzen, 1991). Attitude (ATT) towards performing the behaviour is defined as a person’s general feeling that performing that behaviour is a favourable or unfavourable action. SN are perception of whether people who are important to the person think they should or should not perform an activity; this assumes that the more an individual perceives that others think he or she should engage in a behaviour, the more likely it is that the person will do so. PBC is assumed to reflect past experience as well as anticipated obstacles. The more resources and opportunities that individuals think they possess and the fewer obstacles they anticipate, the greater their perceived control over the behaviour. Behavioural Intention (BI) is the individual’s subjective probability that he or she will engage in that behaviour. The immediate determinant of behaviour is the individual’s intention to perform or not perform that behaviour. BI depends on three major factors: attitude towards performing the behaviour, SN and PBC. The relative weights of these three components are expected to vary with the kind of behaviour being predicted and with the conditions under which the behaviour is to be performed.

TAM was initiated by Davis (1989) based on TRA, whose purpose was to generally explain the decisive factor of users for the IT acceptance, to test and verify it by the theory and to interpret most technology use behaviours. It is considered that belief will influence attitude toward use, which will further influence BI to use. TAM theorises two critical beliefs determining a user’s adoption intention and actual usage of IT. The first belief is Perceived Usefulness (PU), reflecting a person’s salient belief in using the technology, will be helpful in improving performance. The second belief is Perceived-Ease-Of-Use (PEOU), explaining a person’s salient beliefs in using the technology, will be free of any effort. And there is obviously positive relation between the BI to use and the system usage. However, the difference from TRA is the exclusion of SN in the model study. In light of the acceptance of users for the information systems, TAM brings up two special beliefs, that is, PU and PEOU.


Figure 1 Technology Acceptance Model, Theory of Planned Behaviour and Combined TAM-TPB model

Source: Davis (1989), Ajzen (1991), Taylor and Todd (1995a).

Taylor and Todd (1995b) believed that the ability of TAM to anticipate the user’s BI to use new technology and the actual use had been supported by a great deal of empirical researches but the model has not included the other two factors (social and control factor) which have been proved by many studies to have the remarkable ability influencing the user’s actual use in using new technology (as shown in Figure 1;Taylor and Todd, 1995a). Combined TAM and TPB (C-TAM-TPB) is a hybrid model which combines the constructs of TPB and TAM. Experience was incorporated into this model in between experienced and inexperienced users. PU, ATT toward behaviour and PBC were all more salient with increasing experience while SN became less salient with increasing experience (Venkatesh et al., 2003).

Furthermore, the explanation for user behaviours to accept new technologies is always a hot issue of the study on the information systems introduction, whose related researches are one of the most mature research fields in modern information management literatures. For this issue, many theoretical models have been developed from information management to sociology and psychology since 1980, including TRA (Fishbein and Ajzen, 1975; Ajzen and Fishbein, 1980), TPB (Ajzen, 1991) and TAM (Davis, 1989), etc. In views that the theory of user acceptance behaviour for new technology equipments is highly valued and continually developed, it is thus clear for the importance of the study on technology acceptance behaviour to the successful introduction of information systems. But the variables and the causal relations in various theoretical models are different. With different theoretical bases and cases, each theoretical model has different influence variables and causal relations. Therefore, for the application of technology acceptance behaviour theory, it still needs more empirical analyses to test and verify the application of theoretical models and their influence variables to different industries.


TAM and TPB have emerged as two dominant models and together provide a theoretical foundation upon which C-TAM-TPB was developed and examined (Taylor and Todd, 1995a; Chau and Hu, 2002). With inclusion of SN and PBC in TAM, they proposed to combine the TPB and the TAM. Such two influence factors are also key variables in TPB. Hence, the cognitive influences specified by TAM may serve as important precedents of attitudinal beliefs in TPB, which reciprocally may enhance the explanatory power of TAM via its potential for adding dimensions essential to individual technology acceptance. Based on the empirical result of (Taylor and Todd, 1995a; Chau and Hu, 2002) it is found that C-TAM-TPB obtained from the combination of TAM and TPB has great Goodness-of-Fit for explaining the user’s behaviour in using new technology.

2.3 Research framework and hypothesis development

Based on these theories and prior research, our basic hypotheses were: For the result of related studies, it shows that the more positive the user’s attitude

toward the system, the higher the user’s intention to use such system (Davis, 1989; Davis, Bagozzi and Warshaw, 1989; Taylor and Todd, 1995a). This research deduces the first hypothesis. H1: Attitude positively influences BI.

Previous research provides evidence of the significant effect of PU on usage intention (Davis, 1989; Davis, Bagozzi and Warshaw, 1989; Jackson, Chow and Leitch, 1997; Venkatesh, 2000; Venkatesh, Morris and Phillip, 2000; Lee and Jun, 2007). For the result of related studies, it shows that the more the user believes his/her work performance will be enhanced after using new system, the more he/she will be willing to use such system. This research deduces the second hypothesis. H2: PU positively influences BI.

Extensive research has consistently discussed that there is a positive relationship between PU and PEOU with acceptance of IT (Agarwal and Prasad, 1999; Dishaw and Strong, 1999; Venkatesh, Morris and Phillip, 2000). For the result of related studies, it shows that the attitude toward adopting the system will be more positive when the user perceives the system is easier to learn and it is more unnecessary to make efforts to use such system (Davis, 1989; Davis, Bagozzi and Warshaw, 1989; Taylor and Todd, 1995a). This research deduces the third hypothesis. H3: PEOU positively influences Attitude.

The result of related studies shows that the attitude toward adopting the system will be more positive when the user believes it will be helpful to enhance the work performance if adopting new IT (Davis, 1989; Taylor and Todd, 1995a; Wu and Chen, 2005). This research deduces the fourth hypothesis. H4: PU positively influences Attitude.

The related study shows that the user will more believe it is helpful to enhance the work performance after adopting the system when he/she perceives such system is easier to learn. This research deduces the fifth hypothesis. H5: Perceived Ease Of Use positively influences PU.

For the result of related studies, it shows that the user will be more willing to use new system when he/she perceives that the important person concerned thinks it shall be more to use such system (Ajzen, 1991; Taylor and Todd, 1995a; Wu and Chen, 2005). This research deduces the sixth hypothesis. H6: Subject Norms positively influence BI.


The related study shows that the user’s BI to use the system will be stronger when he/she believes he/she has the ability to use such system or his/her relevant resources are more (Ajzen, 1991; Taylor and Todd, 1995a; Wu and Chen, 2005). This research deduces the seventh hypothesis. H7: PBC positively influences BI.

3 Research methods

3.1 Measures

To ensure the questionnaire satisfies the content validity, the constructs in the study are modified from relevant prior research, and three command steps are employed to choose items for measurement. Firstly, the items from the existing literature were translated into Chinese. Secondly, two university professors and two programmers of GPS navigation, who are proficient in English and familiar with IT, were asked to provide assistance in examining the appropriateness of the Chinese version of the scale, translated from the original English measurement items. Any inappropriate items were eliminated. Finally, a reexamination for the measurements was repeated throughout the pilot test. In addition, to ensure desired balance and randomness in the questionnaire, some of the items were worded with proper negation and all items were randomly sequenced on the questionnaire in order to reduce the potential ceiling (or floor) effect that induces monotonous responses to the items designed to measure a particular construct.

3.2 Subjects

A survey agency conducted a web-based survey to evaluate the research model for one month. The participants for this study were 386 undergraduate students. After eliminating insincere responses through data filtering, we selected 345 usable responses as the sample finally. The variables and measurement scales are shown in Table 1. We used the value of Cronbach’s for identifying the reliability of the questionnaires, and conducted factor analysis for convergent validity.

The result of our pilot test showed the high reliability of all the questionnaires. All the questionnaires used in this survey have been validated in previous studies. Each item of the questionnaire was measured on a seven-point Likert scale with the end points of ‘strongly disagree (1)’ and ‘strongly agree (7)’. Because of the homogenous nature of our subjects, we did not check for non-response bias.

After data collection, a two-step procedure proposed by Anderson and Gerbing (1988) is applied during the Structural Equation Model (SEM) test. The first step involves developing an effective measurement model with confirmatory factor analysis, while the second step analyses the structural model. Both SPSS 14 and AMOS 6.0 are adopted as the tools for analysing the data. Model estimation was executed through use of the generalised least squares technique.


Table 1 Question items used in the study

Construct Item Measure Source

Behavioural Intention (BI)

BI1 Whenever possible, I intend not to use automotive telematics while driving

Davis (1989), Taylor and Todd (1995a), Venkatesh and Davis (1996), Pavlou (2003) and Wu and Chen (2005)

BI2 I intend to use automotive telematics while driving as often as needed

BI3 To the extent possible, I would use automotive telematics while driving

Attitude (ATT) ATT1 Using automotive telematics is a wise idea

Ajzen and Fishbein (1980), Taylor and Todd, (1995a) and Wu and Chen (2005)

ATT2 Using automotive telematics would be pleasant while driving

ATT3 I dislike the idea of using automotive telematics

Perceived Ease Of Use (PEOU)

PEOU1 My interaction with automotive telematics while driving is clear and understandable

Davis, (1989), Taylor and Todd, (1995a), Pavlou, (2003) and Wu and Chen (2005)

PEOU2 It will be difficult to learn how to use automotive telematics

PEOU3 It will be easy to operate automotive telematics while driving

Perceived Usefulness (PU)

PU1 I would find automotive telematics not useful while driving

Davis, (1989), Taylor and Todd, (1995a), Pavlou (2003) and Wu and Chen (2005)

PU2 Using automotive telematics will enhance my effectiveness while driving

PU3 The advantages of using automotive telematics will outweigh the disadvantages

Subjective Norms (SN)

SN1 People who influence my behaviour would think that I should use automotive telematics

Ajzen and Fishbein (1980), Taylor and Todd (1995a) and Wu and Chen (2005)

SN2 People who are important to me would think that I should use automotive telematics

Perceived Behavioural Control (PBC)

PBC1 I have the resources, knowledge, and ability to use automotive telematics

Ajzen and Fishbein (1980), Taylor and Todd (1995a) and Wu and Chen (2005)

PBC2 Using automotive telematics is entirely within my control


4 Data analysis results

4.1 Reliability and validity of research constructs

Before the structural model analysis, this study used Cronbach’s to test the measurement scale reliability of three models’ components. Reliability can reflect the internal consistency of the indicators measuring a given factor. The item reliability for each scale was examined using Cronbach’s to confirm internal consistency of the measures. Nunnally (1978) suggested that a scale can be considered to have high reliability if Cronbach’s is greater than 0.70 and should be removed if it was lower than 0.35. Results showed that every component of the three clusters’ models had strong reliability with all Cronbach’s greater than 0.73 (shown in Table 2).

Convergent validity is achieved if different indicators used to measure the same construct obtain strongly correlated scores. In SEM, convergent validity can be assessed by reviewing the t-tests for the factor loadings (Hatcher, 1994). For all three models in this study, all factor loadings for indicators measuring the same construct are statistically significant (as shown in Table 2), showing that all indicators effectively measure their corresponding construct and support convergent validity (Anderson and Gerbing, 1988). Hence, this supports convergent validity. In addition, using groups with telematics experience, rather than those without telematics experience, helps facilitate improved external validity of the current study.

Table 2 Summary of measurement scales

Construct Indicators Mean SD Variance extracted

Cronbach’s (reliability)

BI1 6.07 0.66 0.58 0.81 BI2 5.63 0.87 – –

Behavioural intention (BI)

BI3 5.72 0.87 – – ATT1 5.73 0.80 0.68 0.87 ATT2 5.76 0.81 – –

Attitude (ATT)

ATT3 5.75 0.81 – – PEOU1 5.56 0.86 0.52 0.77 PEOU2 5.72 0.72 – –

Perceived Ease Of Use (PEOU)

PEOU3 5.64 0.8 – – PU1 6.08 0.71 0.62 0.83 PU2 6.13 0.68 – –

Perceived Usefulness (PU)

PU3 6.06 0.7 – – SN1 5.1 1.14 0.60 0.73 Subjective

Norms (SN) SN2 4.95 1.24 – – PBC1 5.76 0.98 0.64 0.78 Perceived

Behavioural Control (PBC) PBC2 5.96 0.9 – –


Table 3 Chi-square difference tests for examining discriminant validity

TAM (unconstrained) 2 = 73.44 (d.f. = 48)

TPB (unconstrained) 2 = 64.39 (d.f. = 29)

C-TAM-TPB (unconstrained) 2 = 145.41(d.f. = 90)

Construct pair Constrained 2

(d.f. = 49) 2 differenceConstrained 2

(d.f. = 30) 2 differenceConstrained 2

(d.f. = 91) 2 difference

(BI, ATT) 450.56* 377.12* 500.48* 436.09* 595.26* 449.85* (BI, PEOU) 760.88* 687.44* – – 842.57* 697.16* (BI, PU) 1087.53* 1014.09* – – 1116.72* 971.31* (BI, SN) – – 459.23* 394.84* 679.06* 533.65* (BI, PBC) – – 583.98* 519.59* 700.22* 554.81* (ATT, PEOU) 423.64* 350.20* – – 516.16* 370.75* (ATT, PU) 766.85* 693.41* – – 759.28* 613.87* (ATT, SN) – – 319.34* 254.95* 600.53* 455.12* (ATT, PBC) – – 428.15* 363.76* 471.44* 326.03* (PEOU, PU) 763.87* 690.43* – – 807.61* 662.20* (PEOU, SN) – – – – 551.64* 406.23* (PEOU, PBC) – – – – 450.91* 305.50* (PU, SN) – – – – 664.74* 519.33* (PU, PBC) – – – – 750.43* 605.02* (SN, PBC) – – 357.77* 293.38* 496.79* 351.38*

Note: BI, Behavioural Intention; ATT, Attitude; PEOU, Perceived Ease Of Use; PU, Perceived Usefulness; SN, Subjective Norms; PBC, Perceived Behavioural Control.

*Significant at the 0.001 overall significance level by using the Bonferroni method.

The chi-squared ( 2) difference test can be used to assess the discriminant validity (Hatcher, 1994). Discriminant validity is demonstrated if the 2 difference (with 1 degrees-of-freedom (d.f.)) is significant, meaning that the model in which the two constructs are viewed as distinct (but correlated) factors is superior. By using the Bonferroni method under the overall 0.01 level, the critical value of the 2 test is (1,0.01/15) = 11.58. Since the 2 difference statistics for every two constructs all exceed 11.58 for each model (as shown Table 3), discriminant validity is successfully achieved.

4.2 Empirical results

Every construct in the final measurement models is measured using at least two indicator variables. Seven common model-fit measures were used to assess the model’s overall Goodness-of-Fit: chi-squared/ d.f.; Goodness-of-Fit Index (GFI); Adjusted Goodness-of-Fit Index (AGFI); Normalised Fit Index (NFI); Comparative Fit Index (CFI); Root Mean square Residual (RMR) and Root Mean Square Error of Approximation (RMSEA). As shown in Table 4, Comparison of all fit indices with their corresponding recommended values provided evidence of a good model fit ( 2/ d.f. smaller than 3.0, GFI, AGFI, CFI, NFI all greater than 0.9 and RMR, RMSEA smaller than 0.08), thus demonstrating that the measurement model exhibited a fairly good fit with the data collected (Anderson and Gerbing, 1988; Hinkin, 1995; Hu and Bentler, 1995). However, no statistical differences were found in any included model Goodness-of-Fit measures, suggesting that all the investigated models were of comparable fit.


Based on the entire sample of TAM, TPB and C-TAM-TPB, two paths are not supported (Hypotheses H2 and H6 are not supported), while the remaining paths are all significant at the 0.01 level (H1, H3, H4, H5 and H7 are supported). Properties of the causal paths, including standardised path coefficients and explanation of variance for each equation in the hypothesised model are presented in Figures 2–4. The summary of hypothesis testing results is shown in Table 5.

The explanatory power of each model was also examined. Table 6 summarises the relative strength of each path specified by the respective models. The strength of an individual path was evaluated using the standardised path coefficient. The total effect of one factor on another is obtained by summing up its direct and indirect effects via relevant intervening factors. Bollen (1989) strongly advocates the importance not only of the direct effects, but also of the indirect and the total effects when interpreting results in a SEM. Table 4 Overall fits of models

Fit index Recommended

value TPB TAM C-TAM-TPB 2/ d.f. 3.0 2.07 1.59 1.86

GFI 0.90 0.97 0.96 0.94 AGFI 0.80 0.94 0.94 0.91 NFI 0.90 0.96 0.96 0.93 CFI 0.90 0.97 0.98 0.97 RMR 0.08 0.03 0.02 0.04 RMSEA 0.08 0.06 0.04 0.05

Figure 2 Results of model 1: Technology Acceptance Model

Figure 3 Results of model 2: Theory of Planned Behaviour


Figure 4 Results of model 3: Combined Technology Acceptance Model-Theory of Planned Behaviour

*p < 0.05; **p < 0.01 and ***p < 0.001.

Table 5 Summary of hypothesis testing result

Hypothesis Results H1: Attitude positively influences Behavioural Intention (ATT BI) Supported H2: Perceived Usefulness positively influences Behavioural Intention (PU BI) Not supported H3: Perceived Ease Of Use positively influences Attitude (PEOU ATT) Supported H4: Perceived Usefulness positively influences Attitude (PU ATT) Supported H5: Perceived Ease Of Use Positively influences Perceived Usefulness

(PEOU PU) Supported

H6: Subject Norms positively influence Behavioural Intention (SN BI) Not supported H7: Perceived Behavioural Control positively influences Behavioural Intention

(PBC BI)Supported

Table 6 Strengths of individual factors

Effect on Behavioural Intention TPB TAM C-TAM-TPB Direct effect

PU – 0.04 0.04 ATT 0.94 0.94 0.87 SN 0.10 – 0.025 PBC 0.22 – 0.13

Indirect effect PU – 0.35 0.34 PEOU – 0.64 0.62

Total effect PU – 0.39 0.38 PEOU – 0.64 0.62 ATT 0.94 0.94 0.87 SN 0.10 – 0.025 PBC 0.22 – 0.13


5 Discussions

Several implications can be obtained from the study results. Attitude may be the single most important driver in individual automotive telematics system acceptance decision-making. From the three models of this study, it can verify that the attitude toward use is exactly one of important factors in the adoption of users. Therefore, as promoting telematics service, it shall be provided with various product packing styles to satisfy the fondness of users for in-car decorations and for the visual effect of product shapes. In addition, the service content of Taiwan is mainly navigation, safety and security services. In future, we can increase diversified entertainment services to enhance the fondness of users or passengers for the system service.

In this study, PU appeared to have no significant effects on BI. However, based on the path analysis result of TAM and C-TAM-TPB, we can find that the influence of the PU on the attitude toward use is larger than the PU directly on the BI. This result is different from other researches, but the same with Jackson, Chow and Leitch (1997). However, if we expect users could continue using the system service, a useful service may be a key point for the positive enhancement of attitude toward use.

PEOU positively influences attitude, rather than PU. We can suggest that the user interface is simple to learn, yet incredibly powerful and efficient in use, is what users need. With the automotive telematics system, drivers should establish a direct connection to the dealership through a touch, even a simple voice command. In another word, automotive telematics system should be intuitive and easy to the user’s needs. Enjoyable and hassle-free interactive design is the key element to attracting customers.

SN appeared to have no significant effects on BI. Users are likely to develop independent evaluations and consequently may place less weight on others’ opinions in this study.

PBC appeared to have significant influence on BI but not to an extent comparable to attitude. A plausible explanation for the significant but modest effect is that the operations of telematics technology may not be particularly complicated, especially the considerations of safety, system developers of have to take account that users cannot distract attentions from driving.

6 Conclusions and future work

Creating applications likely to be accepted by target users is critical to harnessing a new technology’s potential. This study investigated factors essential to acceptance of telematics technology by individual users in Taiwan. We took the comparison of TAM related approaches and used BI as a measure of technology acceptance. Findings of the study suggest several potential areas where users of automotive telematics systems or developers might fundamentally and interestingly differ in technology acceptance decision-making, compared with the user populations commonly examined in prior research.

Through the collection of relevant literatures, this research thinks the TAM applicable to automotive telematics systems includes TAM, TPB and C-TAM-TPB under consideration of the characteristics of each theoretical model and the system characteristics in the field of technology acceptance researches. Furthermore, in order to understand the applicability of related technology acceptance theories to the empirical


industry, this research processes the testing and verification and makes the fit analysis as well as the model comparative analysis for those models and their variables. For the analysis of competition indices, we can find that TAM is a better model to apply to the automotive telematics industry. For the path analysis, TAM and TPB has one path that does not only attain to the level of significance, but there are two paths in C-TAM-TPB which is unattained to the level of significance. The explanations of these three models for latent variables are beyond the level except for the explanation for use behaviours.

For future research of this study, we have several suggestions. Firstly, due to closely relation between the numbers or frequency of system uses and the driving frequency of users, we use the frequency of system uses to explain that the actual use may be interfered with the number of driving. And the explanation of actual uses by the time for system uses may be examined with the driving distance of users. Secondly, future researches can use the ratio of the frequency of system uses to the frequency of actual driving to measure in order to reduce the interference by the frequency of driving. Besides, they also can make use of the ratio of the time for system uses to the driving time to measure in order to reduce the interference by the driving distance. Thirdly, this research is designed to study the comparison between models to seek a better model. In future, with the model of this research as the basis, we can increase other theories related to the consumer behaviour model to study the factor influencing the willingness of adoption. Finally, the scope of this research is limited to Taiwan telematics users. We can expand the group of samples to Asian areas and further study the prevalence of the model in the telematics industry in future.

This study suffers from some limitations. The first limitation is the possibility of a common method bias by using a single questionnaire to measure all constructs. Another limitation was that this study used Taiwan areas as its scopes and the automobile forum members as its convenience samples. Additional research across different countries or cultures will be required in order to generalise the findings. The third limitation is that the study uses several different levels of significance (p < 0.001; p < 0.01; p < 0.05). However, the significance level of 0.10 should be used with greater caution during the interpretation of automotive telematics’ BIs.

References ABI Research (2006a) ‘Automotive navigation forecasts’. Available at: http:// www.

abiresearch.com/products/market_data/Automotive_Navigation_Forecasts. Accessed on August 2007.

ABI Research (2006b) ‘Automotive telematics: global consumer telematics markets and technologies’. Available at: http://www.abiresearch.com/ products/market_research/ Automotive_Telematics. Accessed on August 2007.

Agarwal, R. and Prasad, J. (1999) ‘Are individual differences germane to the acceptance of new information technologies?’, Decision Sciences, Vol. 30, pp.361–391.

Ajzen, I. (1991) ‘The theory ofplanned behavior’, Organizational Behavior and Human Decision Process, Vol. 50, pp.179–211.

Ajzen, I. and Fishbein, M. (1980) Understanding Attitudes and Predicting Social Behavior.New Jersy, NJ: Prentice-Hall.

Alliance of Automobile manufacturers (2006) ‘Today’s automobile: a computer on wheels’. Available at: http://www.autoalliance.org/innovation/. Accessed on August 2007.


Aloi, D.N., Alsliety, M. and Akos, D.M. (2007) ‘A methodology for the evaluation of a GPS receiver performance in telematics applications’, IEEE Transactions on Instrumentation and Measurement, Vol. 6, pp.11–24.

Anderson, J.C. and Gerbing, D.W. (1988) ‘Structural equation modeling in practice: a review and recommended two-step approach’, Psychological Bulletin, Vol. 103, pp.411–423.

Bollen, K.A. (1989) Structural Equations with Latent Variables. New York, NY: Wiley. Chau, P.Y.K. and Hu, P.J-H. (2002) ‘Investigating healthcare professionals’ decisions to accept

telemedicine technology: an empirical test of competing theories’, Information and Management, Vol. 39, pp.297–311.

Chen, L.D. (2008) ‘A model of consumer acceptance of mobile payment’, Int. J. Mobile Communications, Vol. 6, pp.32–52.

Davis, F.D. (1989) ‘Perceived usefulness, perceived ease of use, and user acceptance of information technology’, MIS Quarterly, Vol. 13, pp.319–340.

Davis, F.D., Bagozzi, P.R. and Warshaw, P.R. (1989) ‘User acceptance of computer technology: a comparison of two theoretical models’, Management Science, Vol. 35, pp.982–1003.

Dishaw, M.T. and Strong, D.M. (1999) ‘Extending the technology acceptance model with task-technology fit constructs’, Information and Management, Vol. 36, pp.9–21.

Finn, B. and Breen, P. (1996) ‘The use of transport telematics in inter-urban and rural bus service’, IEEE Transactions on Public Transport Electronic Systems, pp.5–10.

Fishbein, M. and Ajzen, I. (1975) Belief, Attitude, Intention and Behavior: An Introduction to Theory and Research Reading. MA: Addison-Wesley.

Han, S., Mustonen, P., Seppanen, M. and Kallio, M. (2006) ‘Physicians’ acceptance of mobile communication technology: an exploratory study’, Int. J. Mobile Communications, Vol. 4, pp.210–230.

Hatcher, L. (1994) Step-by-step Approach to using the SAS System for Factor Analysis and Structural Equation Modeling. Cary, NC: SAS Institute Inc.

Hinkin, T.R. (1995) ‘A review of scale development practices in the study of organizations’, Journal of Management, Vol. 21, pp.967–989.

Hu, L. and Bentler, P.M. (1995) ‘Evaluating model fit’, in R.H. Hoyle (Ed.), Structural Equation Modeling: Concepts, Issues, and Applications. Thousand Oaks, CA: Sagex.

Hung, M.C., Hwang, H.G. and Hsieh, T.C. (2007) ‘An exploratory study on the continuance of mobile commerce: an extended expectation-confirmation model of information system use’, Int. J. Mobile Communications, Vol. 5, pp.409–422.

Jackson, C.M., Chow, S. and Leitch, R.A. (1997) ‘Toward an understanding of the behavioral intention to use an information system’, Decision Sciences, Vol. 28, pp.357–389.

Junglas, I. (2007) ‘On the usefulness and ease of use of location-based services: insights into the information system innovator's dilemma’, Int. J. Mobile Communications, Vol. 5, pp.389–408.

Lee, T.M. and Jun, J.K. (2007) ‘The role of contextual marketing offer in mobile commerce acceptance: comparison between mobile commerce users and nonusers’, Int. J. Mobile Communications, Vol. 5, pp.339–356.

Lu, J., Wang, L-Z. and Yu, C-S. (2007) ‘Is TAM for wireless mobile data services applicable in China? A survey report from Zhejiang, China’, Int. J. Mobile Communications, Vol. 5, pp.11–31.

Melvin, D. (1999) ‘Integrating GPS receivers into consumer mobile electronics’, IEEE MultiMedia,Vol. 6, pp.88–90.

Nunnally, J. (1978) Psychometric Theory (2nd ed.). New York, NY: McGraw-Hill. OnStar (2006) ‘GM to offer simplest, smartest navigation system in North America’. Available at:

http://209.235.195.221/releases_detail.php?ItemID=343. Accessed on February 2007. Park, C-W., Joe, M-J., Byun, S-C., Choi, H-W., Jung, M-H. and Kim, Y-I. (2003) ‘A turn-by-turn

navigation system for automotive telematics terminals’, IEEE Proceedings, Intelligent Vehicles Symposium, pp.21–24.


Paul, J., Hu, P.Y.K.C., Olivia, R. and Liu Sheng, K.Y.T. (1999) ‘Examining the technology acceptance model using physician acceptance of telemedicine technology’, Journal of Management Information Systems, Vol. 16, pp.91–112.

Pavlou, P.A. (2003) ‘Consumer acceptance of electronic commerce-integrating trust and risk with the technology acceptance model’, Int. J. Electronic Commerce, Vol. 7, pp.69–103.

Strategy Analytics (2001) ‘In-car telematics terminals market 2000–2007’, In-Car Telematics Terminals Forecast 2001–2007 Report, January 2002.

Tanizaki, M., Maruyama, K. and Shimada, S. (2005) ‘Acceleration technique of snake-shaped regions retrieval method for telematics navigation service system’, International Conference on Data Engineering, pp.949–957.

Taylor, S. and Todd, P. (1995a) ‘Assessing IT usage: the role of prior experience’, MIS Quarterly,Vol. 19, pp.561–570.

Taylor, S. and Todd, P. (1995b) ‘Understanding Information technology usage: a test of competing models’, Information Systems Research, Vol. 6, pp.144–176.

Telematics Research Group (2002) Auto Manufacturers Seek Younger Buyers for Telematics Systems and Services. Telematics Research Group Inc.

Tsou, M-H. (2004) ‘Integrated mobile GIS and wireless internet map servers for environmental monitoring and management’, Cartography and Geographic Information Science, Vol. 31, pp.153–165.

Venkatesh, V. (2000) ‘Determinants of perceived ease of use: integrating control, intrinsic motivation, and emotion into the technology acceptance model’, Information Systems Research, Vol. 11, pp.265–342.

Venkatesh, V. and Davis, F.D. (1996) ‘A model of the antecedents of perceived ease of use: development and test’, Decision Sciences, Vol. 27, pp.451–481.

Venkatesh, V., Morris, G.M. and Phillip, L.A. (2000) ‘A longitudinal field investigation of gender difference in individual technology adoption decision-making process’, Organizational Behavior and Human Decision Process, Vol. 83, pp.33–60.

Venkatesh, V., Morris, M.G., Davis, G.B. and Davis, F.D. (2003) ‘User acceptance of information technology: toward a unified view’, MIS Quarterly, Vol. 27, pp.425–478.

Wu, I-L. and Chen, J-L. (2005) ‘An extension of trust and TAM model with TPB in the initial adoption of online tax: an empirical study’, Int. J. Human Computer studies, Vol. 62, pp.784–808.

Yang, K. (2005) ‘Exploring factors affecting the adoption of mobile commerce in Singapore’, Telematics and Informatics, Vol. 22, pp.257–277.

Zhang, D., Wang, X-H. and Hackbarth, K. (2004) ‘OSGi based service infrastructure for context aware automotive telematics’, IEEE Transactions on Vehicular Technology, pp.2957–2961.

Zhao, Y. (2002) ‘Telematics: safe and fun driving’, IEEE Intelligent Systems, Vol. 17, pp.10–14.

Documents

The empirical study of automotive telematics acceptance in ... › 5cf9 › 7fd0b4fc4c7c... · the automotive telematics market is poised for explosive growth (Zhao, 2002). Automotive