Voting for two different parties in mixed-member systems  

By Pedro Riera (University Carlos III, Madrid), and Damien Bol (King’s College London).

What is the story?

In some countries, voters have two votes: (1) a ‘candidate vote’ usually in a single-member district like in the UK, and (2) a ‘party-list vote’ in a multi-member district like in Spain. These are called mixed-member systems.

A remarkable increasing number of democracies around the world have adopted a mixed-member system in the last twenty-five years. They represent a good compromise between majoritarian and proportional representation systems. Emblematic examples include Germany, New Zealand, and Japan.

A substantial portion of voters use these two votes to support different parties. This is typically referred to as a split-ticket. In a recently published paper, we look at this intuitively-odd behavior, and show that even a very subtle difference in the electoral system can have important consequences for vote choice.

Hypothesis

There are two types of mixed-member system. In the first type, called MMP for mixed-member proportional, there is a linkage between the allocation of seats across tiers. The system works in two steps. First, the single-member votes are counted and transformed into seats. Then, some extra seats are allocated to parties, so that the final seat share of each party corresponds to its proportional representation vote share. In other words, proportional representation votes are used to compensate single-member district votes.

In the second type, called MMM for mixed-member majoritarian, there is no linkage between the tiers. Everything works just as if there were two elections organized in parallel: one in single-member districts, and the other under proportional representation.

Our argument is that the type of mixed-system should influence the propensity to cast a split-ticket. In MMP, voters should not care about which party is elected in single-member districts. Regardless of which party is elected, the proportional representation votes will compensate. In other words, single-member district votes have no effect on the final seat share of the parties.

Voters should thus feel to vote for another party than their favorite one in the single-member districts of MMP systems. This vote will not influence the overall success of their favorite party. But this is not the case in MMM systems where both votes count. Thus, split-tickets votes should be more likely in MMP than in MMM.

Results

To test this hypothesis, we gathered survey data across the four available waves of the Comparative Study of Electoral System Project. We also added all the survey data we could find for countries using mixed-member systems. In total, our data is covering 18 mixed-member elections in 7 countries/regions.

In each of these surveys, respondents were asked about their two votes for the latest election.
We consider that they are split-ticket voters if the single-member district vote is different from the proportional representation vote. Table 1 shows the proportion of split-ticket votes in our data. We observe that there are many more split-ticket votes in MMP elections than in MMM elections. The difference is about 15 percentage points.

In the paper, we estimate a series of regressions in which we show that this difference is robust to the inclusion of many controls variables (socio-demographics, incentives to cast a strategic vote, democratic background of the country…) and many specifications (including the exclusion of some countries from the dataset).

pic

Conclusion

With our paper, we would like to argue that it is important to study the intricacies of electoral systems to examine their impact on voting behavior and elections. More importantly, it is crucial to consider even the smallest detail of electoral systems. A difference that many might see as anecdotal, such as the existence of seat-linkage in mixed-member systems, can have important consequences for vote choice

For more details, see Pedro Riera and Damien Bol. 2017. >Ticket-splitting in Mixed-member Systems: On the Importance of Seat Linkage Between Electoral Tiers. West European Politics 40(3): 584-597.

Posted in Blog | Leave a comment

Party Mobilization and Electoral Systems

 By Ignacio Lago (Universitat Pompeu Fabra) & al. 

Research question

Electoral systems scholarship has extensively researched how electoral systems affect voting and parties’ entry decisions. However, we have almost no insight into how electoral systems shape the strategies adopted by political parties in election campaigns. The goal of our study is to explore how district magnitude and the number of districts shape campaign strategies.

Arguments

When there is a single nationwide district, the payoff in seats of any given increment of votes as a consequence of mobilisation efforts is the same everywhere. Accordingly, both large and small parties will invest more heavily in the most populated areas/provinces given that they can win more votes there. When using a districted electoral system, party mobilisation will be driven less by (district-level) population (i.e., district magnitude) as the rate at which a mobilising party gains seats when it gains more votes differs across districts. But this logic only applies to large parties which face incentives to mobilise everywhere. Given that small parties only have good chances of winning a seat in those districts allocating a high number of seats, they will continue investing their resources in the most populated areas.

Second, as parties exert mobilizational effort the higher the probability of that effort being decisive, small national parties will target densely populated areas regardless of the electoral system. On the contrary, large parties’ mobilizational effort will be more driven by population when using a single national district than in a districted electoral system.

Data

The hypotheses are examined through a quantitative analysis of party mobilisation in the 2009 European election and the 2011 Lower House election in Spain. The 50 Spanish European MPs are elected in a single nationwide district, while the 350 members of the Lower House are elected in 52 districts in which magnitude ranges from 1 to 36. The analysis is focused on three national parties, the two largest, the Socialist Party (PSOE), the Popular Party (PP), and a small one, Union, Progress and Democracy (UPyD). The dependent variable is the number of visits to each district made by the candidate that topped the list in the 2009 European election and the candidate that topped the list put out by the district of Madrid in the 2011 Lower House election. We also use qualitative data collected through a series of semi-structured face to face interviews with members of the campaign teams of the three parties.

Findings

We found that varying district magnitudes create different incentives for campaigning. This effect on parties is not universal: smaller parties always target population size or districts with the highest number of seats to be awarded, where their chances of winning are much greater. We find that larger parties alter their strategies depending on the number of districts, mainly by targeting populous provinces in an electoral system with a single national district. When there are multiple districts, large parties will visit more provinces than when there is only one district.

Figure 1. Frequency of visits to each province in the 2009 European election

Figure 1. Frequency of visits to each province in the 2009 European election

Figure 2. Frequency of visits to each province in the 2011 Lower House election

Figure 2. Frequency of visits to each province in the 2011 Lower House election

Implications

We show that there is not always a change in party mobilisation strategies when there is a change in the payoff of votes to seats. The effect of district magnitude and the number of districts on party mobilisation strategies depends on the size of the political party.

For more information, see:

Lago, Ignacio, Sandra Bermúdez, Marc Guinjoan, Kelly Rowe, and Pablo Simón. “Party Mobilization and Electoral Systems.” Government and Opposition, January 2017, 1–24. doi:10.1017/gov.2016.46.

 

Posted in Blog | Comments closed

Anti-Elite and Anti-Corruption Appeals of European Political Parties

By The CHES Team 

What is the story?

While democratic institutions are currently being subverted by populists both in Hungary and Poland, the 5 Star Movement and the National Front are leading the polls in France and Italy respectively, and the Brits opted for a “hard Brexit” after a clearly anti-elite referendum campaign. These are just a few examples of how anti-establishment sentiments, typically depicting the existing political system as broken, and politicians as corrupt, are gaining prominence across the European continent.

In a recent article published in Research and Politics, we tackle this issue, and study the variation of anti-elite and anti-corruption salience in party positioning in European democracies. We show that whereas the salience of anti-elite appeals varies mostly as a function of party ideology, the salience of political corruption depends mostly on the country’s quality of government. Simultaneously, we introduce the most recent 2014 round of the Chapel Hill Expert Survey (CHES), which is the longest-running expert survey on party positioning in European democracies.

Hypotheses

Although some may intuitively assume that anti-elite and anti-corruption political messages are conceptually closely related, we expect that they function according to divergent logics.

Populist anti-elite stances are likely to be associated with ideological extremism. We differentiate two classical ideological axes: economic left-right, and placement on the socio-cultural (GAL-TAN) dimension, which spans from social liberalism to social conservatism. Left-wing extremists tend to stress economic issues, whereas right-wing extremists tend to emphasize their authoritarian and nationalist appeals. Both extremes, the economic left and the cultural right, should be more inclined to criticize the political establishment (in much the same way as they have opposed European integration).

Hypothesis 1: Parties of either the economic left or the socio-cultural right are more likely to emphasize anti-elite, anti-establishment rhetoric.

As regards anti-corruption appeals, they are less likely to be related to ideology. While populists often denounce the alleged corruption of career politicians, a call to reduce political corruption alone does not make a party populist. Instead, anti-corruption appeals are likely to be driven by the prevalence of corrupt practices. The more they are seen as widespread, the more this issue will be salient to voters and, thus, to parties.

Hypothesis 2: Parties in countries with high levels of political corruption are more likely to stress the importance of reducing political corruption.

Data and Methods

To test our hypotheses, we use the data from the 2014 wave of the Chapel Hill Expert Survey (CHES) available at chesdata.eu. Administered in 2014 to 337 political scientists specializing in political parties and European integration, the 2014 CHES, this data provides information about the positioning of 268 parties on political ideology, European integration, and various policy areas. The survey covers political parties in 31 countries, including all European Union member states, plus Norway, Switzerland and Turkey. In addition, the 2014 survey has been combined with prior waves to produce a trend file with five time points from 1999 to 2014, making the CHES the longest-running, most extensive expert survey on political parties in Europe.

The dependent variables in our research correspond to two newly included questions about the “salience of anti-establishment and anti-elite rhetoric” and the “salience of reducing political corruption” for the political parties of Europe. All experts were asked to provide salience scores for all parties in a given party system on these two questions; responses could range from 0 (not at all important) to 10 (very important).

We employ a multi-level regression analysis of the two variables. The predictors at the party-level are economic (left-right) and social (GAL-TAN) placement of political parties (we add also their quadratic terms to allow for curvilinear relationships), party age and incumbency. At the country-level, we include the quality of democracy using the Group’s International Country Risk Guide (ICRG) from the Quality of Government Dataset.

Results

The results of the multi-level analysis are presented graphically in Figures 1 and 2. In conformity with our hypotheses, anti-elite salience is strongly associated with ideological extremes (the economic left and the socio-cultural right, see Figure 1). In contrast, anti-corruption salience varies as a function of quality of government, but it is practically unrelated to ideology (see Figure 2). In addition, both anti-elite and anti-corruption salience tend to be higher among more recent parties and parties outside the government (results not shown in Figures 1 and 2).

Figure 1 Predicted anti-elite salience

Predicted anti-elite salience

Figure 2 Predicted anti-corruption salience

Predicted anti-corruption salience

 

Implications

This research provides a meaningful contribution to a better understanding of party competition in Europe. It unveils a contrasting logic in the functioning of two types of similar party appeals. While anti-elite salience primarily depends on parties’ ideology, anti-corruption salience reflects the environment in which parties operate.

More generally, the anti-elite and anti-corruption questions, newly included in 2014, supplement core items in the CHES that have now been collected over five time points, making the dataset an increasingly useful source of information for dynamic analysis of party positioning across Europe.

For more information, see:

Polk, Jonathan, Jan Rovny, Ryan Bakker, Erica Edwards, Liesbet Hooghe, Seth Jolly, Jelle Koedam, Filip Kostelka, Gary Marks, Gijs Schumacher, Marco Steenbergen, Milada Vachudova and Marko Zilovic (2017). Explaining the salience of anti-elitism and reducing political corruption for political parties in Europe with the 2014 Chapel Hill Expert Survey data. Research & Politics 4 (1): 1-9. doi: 10.1177/2053168016686915. Available also here as an open-access PDF .

 

Posted in Blog | Comments closed

« Negative » personalization: party leaders and party strategy

By 

Scott Pruysers, University of Calgary
William Cross, Carleton University

The Story

Political parties are increasingly going negative in their campaign advertising and electoral messaging. At the same time, party leaders and candidates are becoming increasingly relevant to considerations of vote choice and to the electoral success of political parties. There is an increasing trend in recent decades towards more candidate-centered politics, individualized local campaigns and a personalization of politics more generally. These trends point to changing electoral and political norms in which the centrality of individual actors has increased while emphasis on the political party has declined. In this sense, we are particularly interested in the targets of negative campaigning, especially from the perspective of personalization. Is it opposing political parties or their leaders who are targeted in routine election campaign communication? Our central contribution is an examination of the inter-party dynamics of campaign personalization and the development of a new concept: negative personalization. We define negative personalization as an emphasis on opposing party leaders in campaign communication more so than on the parties that they lead. In adopting this approach we question whether parties play a role in personalization by negatively personalizing their opponents. In particular, we hypothesize that negative campaign personalization is a common feature of election campaigns and that this negativity is targeted at unpopular leaders more so than popular ones.

The Data

Drawing on data from two recent elections (2011 and 2014) in the province of Ontario, Canada’s largest province, we provide a preliminary empirical look at the dynamics of negative personalization in election campaign material. We do so by examining 53 television advertisements as well as more than 350 party press releases in order to gauge the target of negative party messaging. Additionally, we take a closer look at the campaign dynamics that shaped negative personalization during the 2011 and 2014 Ontario provincial elections.

The Results

First, we provide compelling – albeit preliminary – evidence to demonstrate that negative personalization is a common feature of contemporary election campaigns. Parties routinely attack opposing party leaders in addition to the parties that they lead. In fact, our analysis of the 2011 and 2014 provincial elections in Ontario demonstrate that both television advertisements and press releases are significantly more likely to mention an opposing party leader than an opposing party. This is particularly evident in television advertising where more than half of all campaign ads targeted an opposing party leader compared to only 15% that mentioned an opposing party. The predominance of negative personalization in television advertising is consistent with the broader personalization literature, which has noted the ease with which leaders, or opposing leaders in our case, can be pictured on screen and the impact that these images can have on voter perceptions.

Second, we find clear evidence that negative personalization is indeed a calculated decision. Parties make strategic use of their campaign messaging and in this regard attack their opponents where they are the weakest: party leaders who are popular experience the least negative personalization while relatively unpopular leaders experience the most. This helps to explain why the New Democrat leader escaped negative personalization in 2011 and 2014 while the PC leader did not. It also explains why the Liberal leader was subject to more negative personalization in 2011 than his more popular successor in 2014. Furthermore, an examination of the dynamics of negative personalization over the course of the 2014 Ontario election campaign reveals that the Progressive Conservative strategy responded to shifting public opinion and targeted the Liberal leader the most when her approval was lowest.

Conclusion

While the conclusions presented here need to be tested in other cases before they can be generalized beyond this particular analysis, this note does offer the first theoretical and empirical look at the concept of negative personalization as well as provide suggestions for future research.

 

 

 

Posted in Blog, Party Strategies | Comments closed

Call for Papers: Pre-APSA MEDW Workshop, San Francisco, August 30, 2017

André Blais, University of Montreal
Filip Kostelka, University of Montreal

The Event
The Making Electoral Democracy Work (MEDW)  project calls for papers to be presented at a workshop held on August 30, just before the 2017 American Political Science Association (APSA) annual meeting in San Francisco. Proposed papers should deal with the core research themes of the project, which examines how the rules of the game (especially the electoral system) and the electoral context influence the dynamic and reciprocal relationship between voters and parties in democracies. Among other topics, we welcome papers on voter turnout, political participation, vote choice, party competition and evaluations of democratic performance. We will consider with interest all submitted proposals but papers using the MEDW data will be given priority.

How to apply
Send an abstract of not more than 5000 characters with a title (no more than 80 characters), your name and institutional affiliation to andre.blais@umontreal.ca (with filip.kostelka@umontreal.ca in copy) by January 5, 2017.

How to access the MEDW data
The MEDW include pre- and post-electoral surveys from 27 elections held at different levels of government in 5 countries (Canada, France, Germany, Spain and Switzerland) and in 10 regions (Ontario, Quebec, le de France, Provence, Bavaria, Lower Saxony, Catalonia, Madrid, Lucerne and Zürich). For more details, see the codebooks and questionnaires available here. The data are available upon request to André Blais (andre.blais@umontreal.ca). Requests should include an abstract for the proposed study.

Contact
For further information on the workshop, do not hesitate to contact Filip Kostelka (filip.kostelka@umontreal.ca). The PDF version of this call for papers is available here.

Posted in Blog | Comments closed

“Who is the best football player?” Researchers ask football fans

 foot-picture

On December 13, France Football will announce the best performing football player of 2016.  Researchers from the Making Electoral Democracy Work (MEDW) project and our international collaborators would like to learn about football fans. To this end, we are holding a simultaneous vote using three different voting systems.

Who should be, according to football fans, the best football player of the year? In addition to answering this question, this project will contribute to our knowledge of voting behaviour, and football in general.

Everyone is invited to participate here. The survey is available in six languages (English, French, German, Italian, Portuguese and Spanish). Please share this survey with your friends across the globe and help both your favourite football player and science.

For more information, see: https://votefoot.org/en

 

Posted in Blog | Comments closed

Voter Turnout & Emigration: What Affects Transnational Electoral Participation?

Filip Kostelka, Université de Montréal & Sciences Po, Paris

 
What is the story?

Migration flows to, but also within, developed democracies have kept intensifying since the end of the Cold War. Consequently, relatively large segments of democratic electorates currently live abroad. This raises the question of emigrants’ engagement in their motherlands’ politics.

In a recent article published in the Journal of Ethnic and Migration Studies, I tackle the topic of transnational electoral participation that results from emigration. My objective is to show how transnational voting rates differ from the domestic rates (i.e. the voting rate in the sending country), what specific factors affect emigrants’ turnout, and, more generally, what implications emigration has for nationwide turnout in the sending countries.

Hypotheses

The literature on political behaviour mostly focuses on migrants’ electoral participation in the receiving country (i.e. the effect of immigration). In contrast, migrants’ participation in the elections of the sending country (i.e. the effect of emigration) has so far been, to a large extent, an uncharted territory. Drawing on the literature on domestic residential mobility (i.e. mobility within a single country), I expect that:

  1. Transnational voting rates are much lower than domestic ones as emigration increases the costs and reduces the benefits of the voting act.

With regards to the factors that affect transitional voting rates, I hypothesize that:

  1. The less burdensome the procedures for voting abroad, the higher the transnational voting rate.
  1. The larger the size of the diaspora, the higher the transnational voting rate (but the lower the overall voting rate).

This last hypothesis is motivated by the fact that large diasporas mean large pools of potential votes and a stronger motivation for political parties to go campaigning abroad and mobilize emigrants (see this news report covering a spectacular transnational campaign event held in 2007 for Romanians in Spain). At the same time, large diasporas mean a large number of voters for whom voting is costlier and less beneficial when compared to voters who stayed back home. I expect that transnational voting rates will never reach the level of domestic voting rates and, therefore, although a rise in the number of emigrants will lead to higher transnational voting rates, it will however also further reduce the overall (i.e. nationwide) voter turnout.

The empirical case: 10 post-communist democracies 

To test my hypotheses, I study legislative elections in ten Central and East European democracies (CEE-10) between 1990s and 2012. As Figure 1 shows, the number of CEE-10 citizens living abroad strongly increased especially after 2000. Nevertheless, voter registration was automatic and for life in all ten countries (note that this has recently changed in Bulgaria, see p. 6 in this OSCE report). Therefore, emigrants kept the right to vote and were counted in the overall voter turnout rates as long as they maintained the citizenship.

Figure 1 The Evolution of the Number of Emigrants from the CEE-10

figure1Findings

My empirical analyses corroborate Hypothesis 1. Transnational voting rates, measured as the number of votes case abroad divided by the number of emigrants, are much lower than domestic voter rates. After 2000, they never exceeded 32.1 % (Slovenia’s legislative election of 2008). This means that the growing emigration depresses the nationwide voting rates in the CEE-10. Actually, in those countries, nationwide voter turnout fell dramatically in the first two democratic decades (by 25.1 percentage points, see also this article). According to my estimates, emigration accounts for almost 10 % of this fall (2.1 pp).

Hypotheses 2 and 3 are also validated. In some of the CEE-10, citizens need to (re)register to vote from abroad. According to my models, which control for domestic turnout and other factors, each extra day that separate the registration deadline from the actual election, reduces transnational turnout by 0.1 percentage points. Conversely, diaspora size exerts the expected positive effect. Every increase in the number of voters abroad (as a percentage of the total electorate), boosts transnational voter turnout by 0.7 percentage points.

Figure 2 Factors Affecting Transnational Voting Rates (regression coefficients)

Factors Affecting Transnational Voting Rates (regression coefficients)

Implications

This study demonstrates that emigration is an increasingly salient factor for understanding voter turnout variation over time and across countries. Its findings are not of interest only to researchers who study electoral participation but also to policy-makers in the CEE-10 and elsewhere who are concerned by declining voting rates. Transnational electoral participation by emigrants specifically depends on legal provisions for external voting and diaspora size. The former factor is fully at the reach of legislators.

For more information, see:

Kostelka, Filip (2016). Distant souls: post-communist emigration and voter turnout. Journal of Ethnic and Migration Studies. doi:10.1080/1369183X.2016.1227696

 

 

Posted in Blog | Comments closed

Putting parties and voters into the lab (at the same time!)

Damien Bol, King’s College London
André Blais, University of Montreal
Simon Labbé St-Vincent, University of Montreal

What is the story?

Lab experiments are increasingly popular to study elections. In a recent book published at Springer, we present the variety of voting experiments, in the lab and on the field, in showing their respective contributions to research in the domain.

In lab experiments, researchers have the possibility to manipulate the factors that may influence the outcome of an election, such as the electoral system, the distribution of preferences or the party platforms. In doing so, they can to isolate the causes of this outcome and the mechanism behind it.

In a recent paper published in Political Science Research and Methods, we report the results of a one-of-a-kind experiment where we put parties and voters into the lab. We hope this will pave the way for new voting experiments that consider how interactions between multiple actors shape the electoral outcome.

A one-of-a-kind experiment

For each experimental session, we organised 4 series of 5 elections between 17 subjects at experimental lab CIRANO in Montreal. We randomly assigned the 17 subjects to a role: 6 were designated as parties and 11 were designated as voters, and to a position on a scale ranging from 0 to 10 (see Table 1 below, parties are represented in letters, voters are represented Roman numbers). The roles and positions were randomly reshuffled after each series of 5 elections.

Table 1: Positions of voters and parties on the 11-point scale

Scale 0 1 2 3 4 5 6 7 8 9 10
Parties A B C D E F
Voters I II III IV V VI VII VIII IX X XI

The elections had two stages. First, the parties decided to form an alliance or not with their paired party (the pairs are A/B, C/D and D/E). Then, the voters saw on their screen which parties are participating and had to decide for which party to vote among them.

After the election, the party that received the most votes was declared the winner. If this winning party was not in an alliance with her paired-party, she got 40 points. If there was an alliance the 40 points were divided between the two partners. The distribution of gains between them constitutes an experimental condition. Depending on the series, it is either equal (each party receives 20 points) or unequal (the party that enters the election receives 30 points, while the other receives only 10 points).

Findings

One of the advantages of our experiment compared to observational studies is that we know how many parties could have potentially participated in the elections, that is 6 parties. In reality, researchers can hardly determine how many parties envisioned participating in an election before the campaign. We can thus calculate how much the Effective Number of Parties (ENEP) actually observed in our experimental elections deviates from this theoretical maximum, and the part of reduction that is due to parties forming alliances and to voters voting strategically, that is deserting the party closest to their position if this party has no chance of winning.

Table 2 reports these results. It shows that on average voters reduce 38% of the fragmentation they could theoretically by voting strategically, and parties 60% by forming alliances. Also, we observe that the degree of reduction by parties is larger when the distribution of gains within alliances is equal.

Table 2: Degree of ENEP reduction by voters and parties

Gains Equal Gains Unequal Total
Reduction by voters 36% 40% 38%
Reduction by parties 64% 57% 60%
ENEP 3.10 3.18 3.14

Our conclusion is that the contribution made by party strategic exit is greater than that due to strategic voting. We explain this difference by the nature of the coordination problem and the amount of gains at stake (party coordination is much easier to achieve than voter coordination). Also, in our experiment just as in real elections, parties have much more to gain if they win, and have thus more incentives to behave strategically than voters.

For more information, see:

Damien Bol, André Blais, and Simon Labbé St-Vincent (Forthcoming.) Which Matters Most: Party Strategic Exit or Voter Strategic Voting? A Laboratory Experiment. Political Science Research and Methods

Posted in Blog | Comments closed

Come for the electoral systems, stay for the debate

capture-decran-2016-09-14-a-10-02-29

by Katherine V. R. Sullivan, Université de Montréal

There has been ongoing talk of a possible electoral system reform in Canada. But what are the various options and what would be their consequences for voters and parties? In order to offer insights on these questions, the Centre for the Study of Democratic Citizenship and the Research Chair in Electoral Studies at the Université de Montréal will be hosting a public forum on electoral reform on October 20th.

This public forum will begin with a presentation by Professor André Blais on the following four voting systems:

First Past the Post (FPTP)

Alternative Vote (AV)

Mixed-Member Proportional (MMP)

Small District Open List Proportional Representation (SOP)

Four political scientists will then be allowed 12 minutes to argue in favour of each of them. This will be followed by an open debate about the pros and cons of these systems before audience members are invited to ask questions – using a mobile platform called Pigeonhole Live – and then vote for their two preferred options.

The forum will take place on October 20, 2016, from 19h30 to 21h30, at the McGill New Residence Hall (3625 Av du Parc, Montréal) and you can register here http://electrefo.eventbrite.com. The forum will be streamed live by CPAC (https://www.cpac.ca)

Posted in Blog | Comments closed

I usually vote but I didn’t vote this time

By

Alexandre Morin-Chassé, Université de Montréal

Damien Bol, King’s College London

The goal of our research project is to improve the quality of post-election survey data on electoral turnout by reducing abstainers’ tendency to lie. Usually, the turnout reported in post-election surveys is much higher than in reality, and this is partly due to abstainers pretending that they have voted. Why do they lie? In every society, there exist social norms that are widely shared by the population. Also, in many countries, voting at elections is the norm and it is common to speak of voting as a duty to fulfil in order to be a good citizen. Because this norm exists and people are aware of it, some abstainers prefer to lie than to report a behaviour that is perceived as socially irresponsible.

Previous research has shown that some people lie even when they complete an online questionnaire, a context where there is no interaction with the interviewer and no risk of being judged. Another psychological mechanism is at play here: the desire to preserve self-esteem. Some of the abstainers share the social norm that voting is a duty. They lie because they prefer to avoid the discomfort they would feel when admitting that they derogate from a norm they endorse.

One way to reduce the abstainers’ tendency to lie is to frame the turnout question in a way that allows these abstainers to claim that they adhere to the norm even if they did not vote. Abstainers who accept the social norm can save face even if they report their true voting behaviour. To test the efficacy of this framing, it is possible to run a survey experiment during which half of respondents are randomly assigned to the classic “yes/no” turnout question, while the other half is presented with a new question. If the new question shows a reported turnout that is closer to the actual turnout, it is presumed that the new version succeeded in reducing lies and thus produces more accurate survey answers. Various studies conducted in the United States have tested which question best succeeds in reducing abstainers’ tendency to lie. Table 1 presents the “classic” and the “face-saving” turnout question wordings.

Table 1. Question wording

[Comment preamble]

In each election we find that a lot of people were not able to vote because they were not registered, they were sick, or they did not have time.

[Standard yes/no voting question]

 

Were you personally able to vote in this election?

1. Yes

2. No

9. Don’t know/Prefer not to answer

[Face-saving voting question]

 

Which of the following statements best describes you?

1. I did not vote in the election

2. I thought about voting this time but didn’t

3. I usually vote but didn’t this time

4. I am sure I voted in the election

9. Don’t know/prefer not to answer

Results from the United States suggest that including face-saving response items in surveys following national elections can reduce reported turnout by a range of 4 to 8 percentage points. However, we do not know whether this new version of the turnout question is also efficient in other countries and whether face-saving response items can also reduce abstainers’ tendency to lie in elections at other levels of government, such as local, regional, or European Elections. Our research aims to fill this gap by analysing 19 surveys experiments in post-election surveys conducted in Canada, France, Spain, Switzerland and Germany for the Making Electoral Democracy Work project.

Figure 1 presents results for each of these 19 survey experiments. The top and the left margins identify the country and the level of the election; the bars represent the difference in percentage points between the turnout measured in the group exposed to the yes/no question and the one measured in the group exposed to the face-saving version.

Figure 1. Treatment effects of the inclusion of face-saving response options on reported turnout

blogue

Note: The error spikes represent the 95% and 99% confidence intervals.

In 11 out of 19 surveys, the inclusion of face-saving response items significantly reduced the reported turnout; in 4 other surveys, the effect goes in the expected negative direction, even if it does not reach conventional levels of statistical significance. Finally, in 4 other surveys, the new question produced a turnout level that is virtually identical to the one produced by the traditional yes/no question. When we combine the 19 datasets into a single large one we find that the face-saving question reduces reported turnout by 7.6 percentage points (p<0.001, N=15,185).

Reference: Morin-Chassé, Alexandre, Damien Bol, Laura Stephenson and Simon Labbé St-Vincent. “How to survey about electoral turnout? The efficacy of the face-saving response items in nineteen different contexts.” Political Science Research Methods (accessible online ahead of print). http://dx.doi.org/10.1017/psrm.2016.31

Posted in Blog | Comments closed