Internet-based Instruments to Increase Civic Li-

teracy and Voter Turnout

Paper presented at the 5th ECPR General Conference;

10-12 September 2009; Potsdam, Germany

August 2009

Giorgio Nadig, NCCR Democracy, Centre for Democracy, University of Zurich

(nadig@nccr-democracy.uzh.ch)

Jan Fivaz, NCCR Democracy, Centre for Democracy, University of Zurich

(fivaz@nccr-democracy.uzh.ch)

ii

Abstract

Switzerland is one of the countries with the lowest voter turnout rates among advanced de-

mocracies. Also the level of civic literacy is often considered as poor – especially among stu-

dents and younger citizens. In the run up to the Swiss parliamentary elections of October

2007 several online tools and small campaigns were developed to increase the level of in-

formation about political parties, candidates and their political programs as well as to make

voting more accessible and attractive. Some of these tools were especially designed for

young and first time voters. This paper will describe from a practical point of view the devel-

opment, the marketing, the distribution as well as the use of two of these tools. One of them

is called “Parteienkompass” (party compass) and designed to inform students about the most

important political parties, the other one is “myVote”, an online voting assistance application

(VAA) based on www.smartvote.ch. We will evaluate how useful such tools are in increasing

the knowledge about political parties and candidates as well as their electoral programs. In a

second step we will argue that such tools fit also perfectly for civic education purposes, since

they draw upon real life decision-making processes. Finally, we will address to the question

whether such tools might have an effect on voter turnout and whether they are even likely to

improve democratic decision-making.

Contents

1 Introduction 1

2 Political participation and interest in Switzerland 2

2.1 Political participation 2 2.2 Political interest 4

3 smartvote and smartvote-based websites 6

3.1 smartvote 6 3.2 Parteienkompass (party compass) 8 3.3 myVote 9

4 The typical user of smartvote and smartvote-based websites 10

5 Impact on voter turnout an democratic decision-making 13

5.1 Impact on voter turnout 13 5.2 Impact on decision-making 14

6 Conclusions 17

7 References 19

1

1 Introduction

Compared with other advanced democracies voter turnout in Switzerland is on a notably low

level. During the 20th century electoral turnout in Swiss parliamentary elections was constant-

ly decreasing. In 1995 turnout reached its lowest score with only 42%. Since then turnout has

slightly increased from 42% to 49% (Lutz 2008). Still, it remains one of the lowest among

advanced democracies. The voter turnout is especially low among young voters.

But not only the level of political participation is considered as low, the same is also true re-

garding the level of political knowledge among younger citizens. In 2001 an IEA-study on

knowledge about democracy and politics as well as on political participation among students

ranked Switzerland near the end of the compared 27 countries (Oser/Biedermann 2003).

On the other hand so-called online Voting Advice Applications (VAAs), which support voters

in their decision-making process in the forefront of elections, have become very popular in

Switzerland during the last couple of years. Today, in almost every European country is at

least one VAA in service. Despite some differences in design and details all VAAs share a

common key element: the issue-matching system. They produce for each voter an individual

voting recommendation based on a comparison of her or his own political preferences and

values with the ones of candidates or political parties. As a result the voter gets a voting rec-

ommendation in form of a list ranking candidates or political parties according to their close-

ness with the voter’s preferences.

In Switzerland a VAA called smartvote (www.smartvote.ch) was offered for the first time in

2003 on occasion of the elections for the national parliament. Since then it has become an

important element of electoral campaigns – not only on national level but also on cantonal

(regional) as well as local level. During the election campaign for the 2007 national elections

smartvote has generated over 963’000 voting recommendations – compared with about 2.3

million voters, which casted their ballot, a quite intense use.

Against this background questions about the potential impact of smartvote and VAAs in gen-

eral on voter turnout and voting decisions themselves arose:

Do VAAs increase the interest in politics and thereby lead to higher voter turnout?

Do VAAs have other impacts on voters’ information-processing and decision-making

processes?

Can VAAs also be used for the purpose of civic education in general? If so, which

standards and requirements do VAA’s have to meet?

This paper will address to these questions. In order to do so the next section will provide

some basic information about political participation and political interest in Switzerland. Sec-

tion 3 will describe the features and the functionality of smartvote and two spin-off projects:

the party compass and myVote. Both projects are using the key components of smartvote

but are addressing specific user groups: the party compass is a tool for civic education pur-

poses and myVote was the website of campaign in 2007 addressing to young and first time

2

voters. Besides a description section 3 contains also some information about the “marketing”

efforts of the three tools/projects. Section 4 will present some statistics about the use of

smartvote and its spin-offs and section 5 will address to the questions mentioned above. Fi-

nally, section 6 will present some conclusions and suggestions regarding further develop-

ment of VAA’s for the use in civic education.

2 Political participation and interest in Switzerland

2.1 Political participation

According to the IDEA database on voter turnout (see Table 1) Switzerland has one of the

lowest rates for political participation among advanced democracies.

Table 1 Voter turnout in selected countries, 1945-2008

Rank Country (number of elections)

Voter turnout

Rank Country (number of elections)

Voter turnout

1 Italy (14) 92.5 29 Germany (13) 80.6

2 Seychelles (2) 90.2 37 Norway (14) 79.5

3 Cambodia (2) 90.5 40 Finland (15) 79.0

8 New Zealand (18) 86.2 54 Ireland (16) 74.9

12 Austria (16) 85.1 55 United Kingdom (15) 74.9

13 Belgium (17) 84.9 77 Canada (17) 68.4

15 Netherlands (15) 84.8 81 France (15) 67.3

16 Australia (21) 84.4 138 Switzerland (13) 49.3

17 Denmark (22) 83.6 139 USA (26) 48.3

18 Sweden (17) 83.3 172 Mali (2) 21.7

Source: IDEA voter turnout database (www.idea.int/vt/survey/voter_turnout_pop2.cfm); voter turnout in percent.

Generally, this low voter turnout is explained by two factors: First it is argued that the exis-

tence of direct democratic instruments reduces the importance of elections to citizens. Swiss

citizens can vote on popular initiatives and referendums several times per year. These addi-

tional possibilities to influence the general direction of politics between two elections de-

creases the importance citizens attach to elections (Linder 1999). Second, from 1959 to 2003

the four largest political parties formed the Swiss government. The distribution of seats within

this coalition was arranged by the so-called “magic formula”1 and did not change for more

than 40 years. In many ways the four big parties worked as a cartel without any real political

competition. The fact that the outcome of the elections did not have any direct impact on the

1 Two seats for the social democrats, two for the liberal party, two for the Christian-democrats and one

seat for the conservatives.

3

composition of the government reduced the relevance of elections further. This situation

changed at the end of the 1990s. There is now more competition between the political parties

and there was also a change of government composition after the last two elections (2003,

2007). With regard to this development elections have become more interesting again and

also voter turnout has started to increase (see Table 2). But nevertheless, political participa-

tion remains still on a very low level compared to other countries.

Since 1995 the Swiss Electoral Studies (Selects) conducts a representative voter survey af-

ter all elections for the Swiss parliament. This data allows a closer look on the socio-

demographic characteristics and attitudes of voters and non-voters in Switzerland. Table 2

contains a summary of the most important findings regarding political participation based on

the data of the four Selects surveys conducted since 1995. It reveals more or less a pattern

which is not new to electoral researchers (Bühlmann et al. 2003): political participation is

higher among men, older voters, voters with higher education and income.2

With regard to the topic of this paper the participation rate of the 18-24 years old voters is of

particular interest. Whereas the general voter turnout between 1995 and 2007 increased

from 42% to 48%, the voter turnout of this specific voter group increased to a much higher

degree from 21% to 35%. Nevertheless the rate is still clearly below average, but it is a very

positive effect.

2 The big difference in participation rates between women and men is likely to be a long-term effect of

the late introduction of voting rights for women in Switzerland (it was introduced in 1971 on national

level and the last canton introduced it in 1991).

4

Table 2 Voter turnout by specific groups at elections in Switzerland, 1995-2007

1995 1999 2003 2007

Overall voter turnout 42 44 45 48

Gender

Men 46 51 52 55

Women 39 37 40 43

Age

18-24 years old 21 28 33 35

25-34 years old 30 29 31 34

35-44 years old 43 40 37 39

45-54 years old 51 52 50 53

55-64 years old 52 52 55 60

65-74 years old 61 57 61 58

75+ years old 58 56 53 62

Level of education

Low 39 31 33 38

Medium 37 39 43 43

High 53 59 56 61

Household income (in Swiss Francs)

-3’000 39 33 34 35

3’001-5’000 36 39 41 44

5’001-9’000 45 44 48 47

9’001+ 52 59 55 58

Political orientation

Left wing 52 52 55 57

Centre 41 42 41 44

Right wing 57 51 59 58

Participation at popular votes

0-3 out of 10 votes 9 6 6 8

4-6 out of 10 votes 25 29 31 34

7-10 out of 10 votes 67 72 72 76

Source: Lutz (2008:8ff.); turnout in percent.

2.2 Political interest

Comparative studies about political knowledge and political interest of younger citizens show

rather poor results for Switzerland. Swiss 14 to 15 year old students have a below average

political knowledge and only an average capacity for interpretation of political information

(Kersten 2003). Also only one third of them is generally interested in politics, what is clearly

below the average political interest in most other countries (Maiello 2003). Some of the stud-

ies even found a lack of understanding and acceptance of fundamental democratic rights

(Haenni-Hoti 2003; Reichenbach 1998: 29). On the other hand there are some positive find-

ings too. Swiss students have not grown distrustful to politics and they even have a high con-

fidence level in the state and its political institutions, which lies clearly over average com-

5

pared to other countries. Furthermore, it seems that at least some of these students develop

a stronger interest in politics as soon as they receive the right to vote (see Figure 1).

Figure 1 reveals that there is a clear relation between the political interest and specific age

groups. A first increase in political interest takes place at the age of 18 as soon as the young

men and women receive the right to vote. Then, until the age of 35, the level of political inter-

est remains more or less unchanged, but after 35 it increases again until the age of 65.

Figure 1: Political interest for different age groups

0

1

2

3

4

5

6

7

8

9

10

-17 18 - 25 26 - 35 36 - 45 46 - 55 56-65 66 - 75 76 -

Source: Swiss Household Panel, 2007

Remarks: 0 = not interested; 10 = strongly interested; N=5’699

The question is how these low rates of political interest, knowledge and participation among

students and young citizens can be explained. For the most part we cannot explain these

particular low rates in Switzerland. But one reason might be the fact that civic education is

much less developed in Switzerland than in many other countries (Milner 2008: 7). For ex-

ample there are often no specific lectures for civic education in the curricula. Mostly civic

education is integrated into lectures on history but it is up to the particular teacher whether it

will be taught or not to a class. Due to the strong federalism within the Swiss educational

system – each of the 26 cantons can define its particular curricla – there are many different

ways how civic education is organised and taught (Quesel/Allenspach 2007:12ff.). This struc-

tural problem is not very helpful in order to establish a reasonable level of civic education in

Switzerland.

6

3 smartvote and smartvote-based websites

3.1 smartvote

Starting point for our e-tools for civic education is the already mentioned Swiss VAA smart-

vote that was developed and offered for the first time in the forefront of the Swiss parliament

elections in 2003. It can be described as the by far most sophisticated voting assistance tool

in western democracies.3 The smartvote website consists of three main elements: the smart-

vote module with the issue-matching system, a comprehensive database providing informa-

tion on all candidates running for office, and an analysis module with elaborated analytical

tools for the visualisation of political positions.

The core of the smartvote website is the issue-matching module. In a first phase, a few of

months before election day, all candidates receive the smartvote questionnaire, and they are

asked to answer the questionnaire completely and to return it (online or by postal mail). The

questionnaire consists of up to 70 questions on the most important political issues4. Thus it

permits to generate a comprehensive political profile.

About six weeks before election day the second, operational phase starts. The smartvote

website is now accessible to voters and leads them in three steps to their individual voting

recommendation:

1. Voters have to specify their political profile: To do so they are asked to answer the

same questionnaire as the candidates before. Voters can choose between two differ-

ent questionnaire sizes. In 2007, the “deluxe version” consisted of 73 questions

whereas the “rapid version” consisted of 36 questions only. Although the questions

are formulated as easily understandable as possible, additional background informa-

tion and explanations including pros and cons to every question is provided.

2. Voters have to customize their voting recommendation: After defining their political

profile voters have to specify in which constituency (electoral district) they live, re-

spectively for which constituency they want to receive a voting recommendation. De-

pending on the electoral system they also have to decide whether they wish to re-

ceive a voting recommendation for lists/parties or for individual candidates.

3. Calculation and presentation of the voting recommendation: Based on this informa-

tion smartvote calculates an individual voting recommendation for each voter. First,

smartvote compares the answers of the voter with the answers of a candidate. The

3 In a green paper on the future of democracy in Europe the whole group of these voting assistance

tools was labelled “smart voting” tools (see Schmitter and Trechsel 2004).

4 Examples for Questions in the questionnaire: “Do you support the introduction of a minimum wage of

3,500 francs (for a full-time position) for all employees?” or “In some foreign cities (e.g.. London, Oslo

or Stockholm) a fee must be paid by motorists wishing to drive into the city centre. Should similar road

pricing also be introduced in Swiss towns?”

7

higher the congruence of the answers between voter and candidate, the more “con-

gruence points” a candidate gets. This process is repeated over all questions and for

every candidate in the selected constituency. Second, smartvote generates the voting

recommendation in form of a list with a decreasing ranking of the candidates accord-

ing to the over-all sum of their congruence points.

Besides the smartvote module the website features some additional services:

The website contains a database with all candidates, including extensive portraits

with political profile (smartvote questionnaire, information about their political career,

their political agenda, and so on), and information about their personal profile (e.g.,

educational, professional and family background). The database also offers links to

personal websites or to video files.

The website provides tools for visual analysis of political preferences: the so-called

smartspider and smartmap charts (for examples see Figure 2). Both analytical graphs

are based on the candidates’ answers to the smartvote questionnaire.

Figure 2: Examples for smartspider (left graph) and smartmap (right graph)

Source: www.smartvote.ch

The growing popularity of smartvote can be attributed to a certain extent to the deployed

“marketing”- instruments to promote the website. In the forefront of every election, before the

website goes online, media partner are aspired. Media coverage helps to increase the

awareness of smartvote and its benefit for voters as well as for candidates and their parties.

In particular latter have an important bearing on the popularity of smartvote. Parties often

urge their candidates to use smartvote, fill in the questionnaire and create a profile of their

own. smartvote has become a valuable campaigning instrument for parties and candidates,

especially for young candidates. Having a personal profile on the smartvote website is a sim-

ple way to reach and being perceived by a large amount of voters. Candidates running a

personal website often link it to their personal smartvote profile. The spread among the par-

8

ties and candidates helped smartvote to gain popularity and become Switzerland’s most

used VAA.

3.2 Parteienkompass (party compass)

The main objective of the party compass is to present Switzerland’s main political parties and

political positions in a nonpartisan and neutral way detached from upcoming elections.

In collaboration with one of the largest educational textbook publisher5 in Switzerland an

adapted smartvote version for civic education, the so-called “party compass”6, was imple-

mented in summer 2006. The party compass addresses itself primarily to teachers and pupils

(upper secondary level) all over Switzerland7.

The party compass website consists of three main elements: an adapted version of the is-

sue-matching system, a comprehensive database providing information on the basic political

positions and values of the five most important Swiss political parties and the party system in

general and visualisations of political positions. Additionally, a download section with further

information about the tool itself and the political parties is provided.

Unlike the smartvote issue-matching system the party compass doesn’t compare the voter’s

answer to those of single candidates but only to those of the five included parties. The issue-

matching system used for the party compass has the same design and operates very similar-

ly to the one used by smartvote. But the used questionnaire contains fewer questions and

the questions are linguistically adapted. These adaptations were primarily made because the

party compass addresses itself to users with less political knowledge and interest. Simple

wording and less complexity within the questionnaire are geared to the knowledge of people

rather uninterested in politics.

The background information provided in the database and the download section are party

and not candidates oriented (i.e. general information about the party, a short summary of the

party history, statistics with the strength of the party and the smartmap and the smartspider

of the party, based on their answers to the questionnaire).

The collaboration with the textbook publisher “hep” turned out to be an important factor of

success. Due to the publication possibilities offered by “hep” teachers all over Switzerland

shortly knew about the party compass. Advanced training course for teachers on the subject

5 See (http://www.hep-verlag.ch)

6 See (http://www.parteienkompass.ch) only available in German

7 In 2007, by order of the Federal Office for Education, Arts and Culture in Austria, in collaboration with

the University of Krems, Politools (the association owning and running smartvote) developed an

adapted Version of the party compass for Austria called Politikkabine. Functioning and content are

very similar. In addition the Politikkabine offers the pupils the possibility to create a VAA of their own.

See (http://www.politikkabine.at) only available in German.

9

“politics in class” were organized where the party compass was presented, information flyers

were distributed during workshops and the going online of the website was announced in

different medias and newsletters. The weakness of the project is its low impact on young

people not attending an institution of learning.

3.3 myVote

The last Swiss parliamentary elections in October 2007 were a perfect moment to point out

the importance of elections to young citizens as future voters and to motivate them to partici-

pate.

Therefore the project myVote8 was initiated. The main focus was to develop a website tai-

lored to the needs of young citizens and first time voters in the forefront of the 2007 elec-

tions. Again the smartvote website served as a model. The myVote website also consists of

three main elements: the smartvote module with the issue-matching system, a comprehen-

sive database providing information on all candidates running for office, and an analysis

module with elaborated analytical tools for visualisation of political positions.

By offering an uncomplicated introduction to the themes "policy" and "elections" the website

is specifically designed for young people and first time voters. The website should not only

provide guidance and stimulating further thoughts it should also help to spark interest in this

democratic act and as a result stimulate users to put the gained political knowledge into ac-

tion. By using myVote the user should be able to answer the following questions: Where do I

stand politically and which politicians or party running for office represents my interests best?

The myVote website distinguishes itself in several points from the smartvote website. A new

design was created for the website and all texts were adapted by using a more common and

understandable language. The visualization possibilities offered by smartspider and smart-

map were placed more prominently and the amount of texts was reduced to the necessary

minimum.

Based on the experience made with the party compass and in order to reach young people,

especially new voters who did not attend an institution of learning, we had to find partners

helping to promote the website and the project ideas in general. In addition to the collabora-

tion with the textbook publisher, teacher’s associations and several youth organisations a

media partner which appeals to young people had to be found.

With 20 Minuten9 we found an ideal media partner for our project. 20 Minuten is Switzer-

land’s strongest and most read free daily newspaper with a daily print run of almost 795’000

copies and over 1’800’000 daily readers, having a German and a French issue10. Besides 20

8 See (http:// www.my-vote.ch) only in German, French and Italian

9 See (http://www.20min.ch)

10 See (http://www.mytamedia.ch/index.php?id=108&L=0%202%2053059)

10

Minuten runs also the largest online community information platform in Switzerland with more

then 1’000’000 visitors per day11 and most readers and users have the age of the target

group 14 to 34 years.

What started as a project became a campaign due to the collaboration with 20 Minuten. Two

and a half months before the election the platform went online and the campaign started. The

platform consisted of a large numbers of contents such as:

the smartvote module with the issue-matching system;

a problem-barometer where users could indicate their worries;

portraits of famous Swiss people including their smartspider graph;

two kind of blogs, one where candidates running for office, the second were a former

politician and a former editor-in-chief wrote their article and the user had the possibili-

ty to comment the articles;

an e-talk section; two candidates from opposite parties were invited to a discussion

and the users could ask them question in real time online;

a public opinion poll whose results were also published in the print edition;

a political quiz, where users could test their political know-how;

a wide variety of information about political parties and the political and voting system

in Switzerland

During the campaign period flyers were distributed at chosen youth events and advertise-

ment were placed to promote the 20 Minuten online platform and the myVote website. The

impact of the campaign and the promotion was remarkable, slightly over 50’000 recommen-

dation were generated on the 20 Minuten platform and 27’000 on the myVote website which

makes 8% of the total of all the recommendation generated in the forefront of the Swiss par-

liamentary elections by smartvote.

4 The typical user of smartvote and smartvote-based

websites12

During the electoral campaign from July to October 2007 smartvote (including the myVote

spin-off) was used more than 963’000 times to generate a voting recommendation. Of course

this number does not equal the number of real voters using smartvote. These figures include

11

See (http://www.20min.ch/intern/mediadaten/soziodemo_09.pdf)

12 Since 2005 the National Centre of Competence in Research (NCCR) “Challenges to Democracy in

the 21st Century” (short NCCR “Democracy”; www.nccr-democracy.uzh.ch) is running a research

project about smartvote. The project is called IP16 “smart-voting” (www.nccr-

democracy.uzh.ch/nccr/knowledge_transfer/ip16) and is run by the Institut de hautes etudes en ad-

ministration publique (IDHEAP) at the University of Lausanne, at the Universities of Zurich and Berne

and at the European University Institute (EUI) at Florence, Italy. The findings presented in this section

as well as in the following section are coming from this research project.

11

a large number of multiple countings because during their visit on the website most of the

voters have generated more than one voting recommendation. Relying on data from the web

server-statistics we are able to adjust these multiple countings and are therefore able to es-

timate the real number of voters using smartvote in 2007: around 375’000. In absolute num-

bers this is a rather low figure compared to VAAs in other countries. But in a small country as

Switzerland - with only 2.4 million voters that actually casted their ballots in 2007 – this

represents about 15% of the voters participating in the elections.

After having presented some information on how many voters are using smartvote, we will

now turn towards the question: how does the typical smartvote user look like? To expect that

the use of smartvote be spread equally among voter groups would be entirely unrealistic.

Several studies about users of VAAs in Germany, the Netherlands and Belgium showed evi-

dence, that there are clear-cut differences between voter groups using services like VAAs

and voter groups not using such instruments (Marschall 2005 and 2008, Trechsel 2007). Ta-

ble 3 contains a comparison of the most important socio-demographic aspects (gender, age,

education and income) between voters, which casted their ballot, and smartvote users.

Table 3: Socio-demographic profiles of voters and smartvote users

Voters (%) smartvote users (%) Difference

Gender

Women 45.9* 43.5 -2.4

Men 54.1* 56.5 2.4

Age

18-24 7.9 20.5 12.6

25-34 13.0 25.6 12.6

35-44 20.0 22.0 2.0

45-54 16.5 13.9 -2.6

55-64 16.7 11.7 -5.0

65-74 14.1 4.2 -9.9

75+ 11.7 2.1 -9.6

Education

Low level of education 14.1 9.0 -5.1

Medium level of education 58.3 45.3 -13.0

High level of education 27.6 45.7 18.1

Household income (in CHF)

-3’000 11.2 3.7 -7.5

3'001-5'001 22.9 12.7 -10.2

5'001-9'001 41.1 37.2 -3.9

9'001+ 24.8 46.3 21.5

N = 4'392 for voters and 333 for smartvote users; source: Swiss Electoral Studies.

* Due to an over-representation of women within the Swiss Electoral Studies data the gender related

figures had to be estimated on the basis of the Swiss Federal Statistics Office’s population statistics

and the voter turnout among women as determined by Lutz 2008: 8.

12

The results presented in Table 3 bear no surprises. They show as expected that the typical

smartvote user is very similar to the typical Internet user: male, young, with a good education

and with an income clearly above average. The comparison with the average voter shows

that there are only very small differences regarding the gender. The differences are a bit lar-

ger with regard to the age groups, but we find the largest disparities within the educational

and income levels.

In Table 4 we will now address to the characteristics of smartvote users regarding political

aspects like interest in politics, political knowledge and voting behaviour.

Table 4: Interest in politics, political knowledge and voting behaviour of voters and smartvote users

Voters (%) smartvote users (%) Difference

Interest in politics

High 16.5 30.5 14.0

Rather high 45.3 48.1 2.8

Rather low 26.3 19.0 -7.3

Low 11.9 2.5 -9.4

Political knowledge

High 6.6 16.2 9.6

Rather high 19.0 26.4 7.4

Medium 26.9 25.2 -1.7

Rather low 29.9 16.8 -13.1

Low 17.6 15.3 -2.3

Voting behaviour

Loyal party voters 77.7 67.0 -10.7

Swing voters 22.3 33.0 10.7

Voters with modified ballots/lists 54.5 65.4 10.9

Voters with vote splitting* 64.6 70.5 5.9

N = 4'392 for voters and 333 for smartvote users; source: Swiss Electoral Studies.

* Interpretation: “64.6% out of the 54.5% of the voters, who modified their ballots/lists, split their votes between two or more than two parties”.

From previous studies (e.g. Bühlmann et al. 2003) we know that citizens who participate po-

litically have a clearly higher political interest and knowledge than the average citizen. Keep-

ing this in mind we can say that even stronger interest in politics and higher level of political

knowledge of smartvote users compared to voters is quite outstanding.

But also some other aspects show interesting results:

Among the smartvote users there are clearly more swing voters than among the vot-

ers in general.

13

Additionally, smartvote users adapt their ballot more often; a possibility the Swiss

electoral system offers. For example they cancel or cumulate single candidates on

their ballot more often.

In doing so, particularly popular is the alternative to split the votes among candidates

of different political parties.

Altogether, Table 4 presents a central quality of a typical smartvote user, who has a very

high interest in politics and is already very well informed. Generally, he does not feel close to

one particular political party. He is a typical swing voter and distributes his votes to different

political parties. For that reason he fits into the general development of the last years, in

which party affiliations became looser – not only in Switzerland but in all advanced democra-

cies (e.g. Dalton 2007).

The findings presented in this section show that smartvote is by far not used equally by all

the different groups of voters. In fact, only a very specific group of voters – regarding both

socio-demographic as well as political aspects – is using smartvote.

5 Impact on voter turnout an democratic decision-making

The impact of the party compass and myVote can not be measured directly but is included in

the evaluation of smartvote’s impact.

5.1 Impact on voter turnout

In order to go more into details, we will try to assess the impact of smartvote on political par-

ticipation. As already mentioned Switzerland belongs to the countries among the advanced

democracies with the lowest political participation rates: only 48% of the electorate partici-

pated in the 2007 elections (Lutz 2008: 8). Among non-voters women, younger people, peo-

ple with lower education and income as well as people with a low political interest and

knowledge are clearly over-represented (Bühlmann et al. 2003). With regard to these charac-

teristics plus the characteristics of the smartvote users presented in the previous section, it is

arguable, whether the use of smartvote could increase voter turnout. It seems, that smartvote

is just not addressing the appropriate 'target groups' – with one exception: the young voters.

But in general it is true, that the specific voter groups, to whom smartvote is an attractive in-

strument, are not the voter groups which show a non-satisfying rate of political participation.

In this regard smartvote is preaching to the converted and will probably not lead to a remark-

able higher voter turnout.

But if smartvote users are asked directly, whether using smartvote motivated or detained

them to go to polls, one receives a different picture: almost 40% declared that the use of

smartvote had a decisive or at least slight influence on their decision to go to the polls. Based

14

on the estimation of a total of about 375’000 smartvote users this would mean that about

120’000 voters or 5% of Swiss voters were positively influenced by smartvote in their deci-

sion to participate in the 2007 elections. Of course these numbers should be treated with

caution, but they indicate that smartvote has a clear potential for improving voter turnout.

This impact is stronger among the voter groups with a usual turnout rate below the average.

Especially women and young voters seem to be stronger influenced by smartvote than other

voter groups (see Table 5).

Table 5 Impact of using smartvote on the decision to vote or not

Decisively motivated to

vote

Slightly mo-tivated to

vote

No influence Slightly de-tained to

vote

Decisively detained to

vote

N

Total 15.4 23.6 60.3 0.5 0.2 17'147

Gender

Men 12.5 23.8 63.1 0.5 0.2 11'890

Women 22.0 23.3 54.0 0.6 0.1 5'226

Age Groups

18-24 20.3 25.4 53.6 0.5 0.1 3'600

25-34 17.4 24.1 57.7 0.6 0.2 5'043

35-44 13.6 23.3 62.4 0.5 0.2 3'602

45-54 11.9 22.7 64.7 0.6 0.1 2'500

55-64 10.1 21.8 67.4 0.4 0.2 1'667

65-74 8.7 20.5 70.3 0.3 0.2 644

75+ 15.4 19.8 63.7 1.1 0.0 91

Source: NCCR “Democracy”, IP16 “smart-voting”, post electoral survey among voters.

We can summarise, that the conclusions of the impact of smartvote on political participation

are ambivalent. On one hand the socio-demographic characteristics of smartvote users tend

to indicate that even a widespread use of smartvote will not boost voter turnout. On the other

hand the smartvote users themselves say that there is a rather strong positive impact regard-

ing their political participation.

5.2 Impact on decision-making

The findings about the impact of smartvote on the voting decisions of voters are much

clearer. Two thirds of smartvote users say that the voting recommendation had a clear im-

pact on their decision for which parties and/or candidates they voted. And again this impact

is clearest among women and younger voters (see Table 6).

15

Table 6 Impact of using smartvote on the voting decision

Did the smartvote voting recommendation have an impact on your voting decision?

Yes No N

Total 66.4 33.6 16'335

Gender

Men 65.0 35.0 11'354

Women 69.9 30.1 4'949

Age Groups

18-24 71.1 28.9 3'352

25-34 73.3 26.7 4'764

35-44 68.0 32.0 3'462

45-54 58.6 41.4 2'408

55-64 53.2 46.8 1'638

65-74 48.8 51.2 625

75+ 41.9 58.1 86

Source: NCCR “Democracy”, IP16 “smart-voting”, post electoral survey among voters.

The findings in Table 5 are more suited to generate interest within the field of electoral re-

search. For civic education they are of minor importance. But smartvote unfolded not only

this direct impact on the decision-making process. It also changed and enriched the whole

information processing and opinion formation of its users (see Table 7).

16

Table 7 Impact of using smartvote on different aspects of opinion formation

18-24 25-34 35-44 45-54 55-64 65-74 74+ Average

Using "smartvote" helped to vote for candidates/parties on an improved information basis.

Agree 56.4 57.5 56.6 49.8 47.6 42.9 46.4 54.5

Rather agree 31.4 30.3 28.9 30.5 28.6 30.9 23.8 30.1

Rather disagree 7.3 7.4 7.5 8.3 10.8 13.2 22.6 8.1

Disagree 4.8 4.8 7.0 11.5 12.9 13 22.6 7.4

Using "smartvote" led my attention to issues, which I otherwise would have neglected.

Agree 23.2 18.6 16.5 15.1 14.6 14.3 7.1 18.0

Rather agree 33.1 30.9 30.1 30.3 32.1 31.5 28.6 31.2

Rather disagree 26.5 29.0 30.2 30.3 30.3 29.2 33.3 29.1

Disagree 17.2 21.6 23.1 24.2 23.1 25.1 31 21.7

Using "smartvote" motivated me to look for additional information about particular issues.

Agree 18.4 14.8 14.6 16.3 16.4 16.1 17.6 15.9

Rather agree 33.1 30.8 32.0 32.2 34.4 36.1 31.8 32.3

Rather disagree 29.3 32.0 29.6 29.2 27.0 27.6 23.5 29.8

Disagree 19.2 22.5 23.8 22.3 22.2 20.2 27.1 22.0

Using "smartvote" motivated me to look for additional information about particular candi-dates/parties.

Agree 23.9 21.8 19.9 17.7 16.5 13.2 15.3 20.4

Rather agree 37.2 36.4 36.4 34.6 31.4 33.0 24.7 35.6

Rather disagree 23.1 23.5 22.9 25.4 28.2 31.4 28.2 24.3

Disagree 15.9 18.3 20.8 22.3 23.9 22.4 31.8 19.7

Using "smartvote" motivated me to engage in discussions about particular issues with other people.

Agree 35.1 30.5 24.8 23.3 20.8 16.9 16.9 27.7

Rather agree 35.6 38.0 37.3 38.5 37.5 36.8 28.9 37.3

Rather disagree 17.6 18.0 21.3 20.4 21.4 24.2 28.9 19.6

Disagree 11.7 13.6 16.6 17.7 20.2 22.1 25.3 15.4

Using "smartvote" motivated me to engage in discussions about particular candi-dates/parties.

Agree 38.0 34.2 28.4 25.7 21.7 18.8 15.9 30.7

Rather agree 35.2 37.1 37.7 38.6 36.6 37.8 39.0 37.0

Rather disagree 15.2 17.0 18.0 19.6 21.7 20.8 19.5 17.8

Disagree 11.5 11.7 15.8 16.2 19.9 22.6 25.6 14.4

Source: NCCR “Democracy”, IP16 “smart-voting”, post electoral survey among voters; N = 16’683-16’731.

Since the beginning of the rapid spread of the Internet there was the fear that the introduction

of e-voting systems as well as other online tools like VAAs could lead to new form of 'instant

democracy' with 'junk votes' and voters as 'couch potatoes'. In such a democracy voters

would make decisions in a hasty and superficial way with only a few or no social contacts or

exchange of different opinions (e.g. Buchstein 2004). Our findings in Table 7 point in a con-

tradictory direction.

17

Table 7 contains some remarkable findings:

smartvote improved the information basis for not less than 85% of its users (despite

their generally high level of political knowledge).

For 49%, answering the 70-item smartvote-questionnaire led their attention to political

issues, which they would have neglected otherwise.

48% respectively 56% of the smartvote users searched due to smartvote for further

information about political issues and political parties.

smartvote motivated its users to engage in discussions about political issues (65%)

as well as about political parties and candidates (68%) with friends and family mem-

bers.

Finally, voters between 18 and 34 years as well as women were stronger influenced

by these effects than the average smartvote user.

All these effects together with the limited but positive effect on voter turnout lead us to the

conclusion, that e-democracy tools like smartvote are delivering over-all an added value to

their users by motivating them to become politically more active, interested and better in-

formed citizens.

6 Conclusions

Civic literacy – general knowledge about democracy and political institutions as well as

knowledge about democratic and civil rights and the ability to develop and compare an own

political position – is a fundamental aim of any civic educational effort. But civic literacy and

political knowledge alone do not guarantee political participation. Above all political participa-

tion requires also political interest. In order to foster political active citizens civic educational

projects and tools have not only the task to impart knowledge, but also the task of sparking

interest in politics.

We argue that tools like smartvote, the party compass or myVote can foster both, knowledge

and interest. Since they draw upon real life decision-making processes with real actors (po-

litical parties and candidates), they offer a potentially better access to students and young

voters as a standard textbook can do. But of course these pilot projects have still their defi-

cits:

Despite the made efforts to use an adequate language and complexity in order to

reach people with low interest and knowledge, the tools were still to complex.

As we could show, using smartvote encouraged searching for additional information

about political parties and candidates as well as about political issues. Future ver-

sions of these tools should therefore pay more attention to integrate such information

(or at least offer links to other websites or textbooks).

The effectiveness of these tools could be increased, if they would be better integrated

into the Swiss curricula. Today the curricula concentrate on offering information about

the electoral system. Political actors and their political positions and values play only

a minor role.

18

Finally, it would necessary to introduce such tools into the training of teachers for civic

education before integrating them into curricula. Today, it is a not uncommon situa-

tion, that the students are better informed about such online tools than the teachers.

As soon as these problems and deficits are solved, we are convinced that such online tools

offer a valuable extension to already existing tools for civic education – not only in Switzer-

land but in many other countries.

19

7 References

Buchstein, Hubertus (2004): Online Democracy, Is it Viable? Is it Desirable? Internet Voting

and Normative Democratic Theory; in: Kersting, Norbert; Baldersheim Harald

(eds.): Electronic Voting and Democracy. A Comparative Analysis; New York; pp.

39-60.

Bühlmann, Marc; Freitag, Markus; Vatter, Adrian (2003): Die schweigende Mehrheit. Eine

Typologie der Schweizer Nichtwählerschaft; in: Sciarini, Pascal; Hardmeier, Si-

bylle; Vatter, Adrian (Eds.): Schweizer Wahlen 1999; Bern; pp. 27-58.

Dalton, Russell J. (2007): Democratic Challenges – Democratic Choices. The Erosion of Po-

litical Support in Advanced Industrial Democracies; Oxford.

Heanni Hoti, Andrea (2003) „Chancengleichheit und Gleichstellung von Migrantinnen und

Migranten – Chancengleichheit und Gleichstellung von Frauen“. In Jugend und

Politik. Ergebnisse der IEA Studie zu politischem Wissen, Demokratieverständnis

und gesellschaftlichem Engagement von Jugendlichen in der Schweiz im Ver-

gleich mit 27 Ländern, ed. Fritz Oser and Horst Biedermann, 101-127. Zürich,

Chur: Verlag Rüegger.

Kersten, Bernd (2003): „Politisches Verstehen der Schweizer Schülerinnen und Schüler“. In

Jugend und Politik. Ergebnisse der IEA Studie zu politischem Wissen, Demokra-

tieverständnis und gesellschaftlichem Engagement von Jugendlichen in der

Schweiz im Vergleich mit 27 Ländern, ed. Fritz Oser and Horst Biedermann, 39-

54. Zürich, Chur: Verlag Rüegger.

Linder, Wolf (1999): Schweizerische Demokratie. Institutionen, Prozesse, Perspektiven;

Bern.

Lutz, Georg (2008): Eidgenössische Wahlen 2007. Wahlteilnahme und Wahlentscheid; Lau-

sanne.

Maiello, Carmine (2003): „Politisches Engagement und politische Aktivität“. In Jugend und

Politik. Ergebnisse der IEA Studie zu politischem Wissen, Demokratieverständnis

und gesellschaftlichem Engagement von Jugendlichen in der Schweiz im Ver-

gleich mit 27 Ländern, ed. Fritz Oser, Fritz und Horst Biedermann, 129-153. Zü-

rich, Chur: Verlag Rüegger.

Marschall, Stefan (2005): Idee und Wirkung des Wahl-o-Mat; in: Aus Politik und Zeitge-

schichte; 51; pp. 41-46.

Marschall, Stefan (2008): Wahl-o-Mat Hamburg 2008. Erste Ergebnisse der Online-

Befragung; online publication from the University of Siegen; Siegen

(http://www.uni-siegen.de/wahlomat/materialen/wom_hh_2008_online-eval.pdf).

Milner, Henry; Nguyen, Chi; Boylston, Frances (2008) : Variations in Civic Education: The

IDEA Civic Education Database ; Montréal

20

Oser, Fritz; Biedermann, Horst (2003): Jugend ohne Politik. Ergebnisse der IEA Studie zu

politischem Wissen, Demokratieverständnis und gesellschaftlichem Engagement

von Jugendlichen in der Schweiz im Vergleich mit 27 anderen Ländern; Zürich.

Quesel Carsten und Dominik Allenspach (2007): Rahmenkonzept zur politischen Bildung in

der Volksschule. Erarbeitet zuhanden des Departements für Bildung und Kultur

des Kantons Solothurn. Solothurn, März 2007

Reichenbach, Roland 1998. „Zwischen Polisidyll und massendemokratischem Realismus.

Bemerkungen anlässlich der Ergebnisse einer Expertenbefragung zur Politische

Bildung in der Schweiz“. In Politische Bildung und staatsbürgerliche Erziehung in

der Schweiz. Perspektiven aus der Deutschschweiz und der Westschweiz, ed.

Fritz Oser and Roland Reichbach, 15-35. Freiburg: Universitätsverlag Feiburg

Schweiz.

Trechsel, Alexander H.; Mendez, Fernando (Eds.) (2004): The European Union and E-

Voting. Adressing the European Parliament's Internet Voting Challenge; London.

Trechsel, Alexander H. (2007): Inclusiveness of Old and New Forms of Citizens’ Electoral

Participation; in: Representation; 43; pp. 111-121.