Project

Back to overview

Do MPs have accurate perceptions of public preferences?

English title Do MPs have accurate perceptions of public preferences?
Applicant Varone Frédéric
Number 172559
Funding scheme Project funding (Div. I-III)
Research institution Département de science politique et relations internationales Université de Genève
Institution of higher education University of Geneva - GE
Main discipline Political science
Start/End 01.11.2017 - 31.07.2021
Approved amount 601'608.00
Show all

Keywords (6)

Switzerland; Parliament ; Information; Representation; Preferences; Policy

Lay Summary (French)

Lead
Le bon fonctionnement d'une démocratie représentative présuppose que les élites élus par suffrage populaire évaluent avec justesse les préférences des citoyens. Ce projet étudie dans quelles conditions les députés perçoivent correctement les positions de leurs électeurs vis-à-vis de sujets politiques concrets.
Lay summary

 

Objectifs et méthodes

Deux questions de recherche sont traitées: dans quelle mesure les députés évaluent-ils de manière adéquate les préférences de la population en générale, et de leur électorat en particulier, sur des enjeux de politique publique? Pourquoi certains élus sont-ils capables de faire de meilleures estimations que d'autres?

L'hypothèse centrale est que les perceptions, plus ou moins justes, des députes dépendent de l'information dont ils disposent, de leurs propres intérêts partisans et de l'interaction entre information et intérêts.

Pour tester cette hypothèse, le projet confronte les réponses que donnent les élus et un échantillon représentatif de la population suisse à des questions (tout à fait identiques) sur des sujets d'actualité, comme par exemple: faut-il accueillir plus de réfugiés politiques, mieux réguler le secteur bancaire, augmenter l'âge de la retraite, renforcer la surveillance par l'Etat des échanges sur Internet, etc.? La confrontation directe des réponses que donnent les élus, d'une part, et de celles des citoyens, d'autre part, permet de mesurer le degré d'exactitude des perceptions qu'ont les élus de l'opinion publique et, ainsi, de leur aptitude à représenter les intérêts de leurs électeurs.

Contexte social

Ce projet est réalisé dans une période qui se caractérise par la montée du populisme dans plusieurs démocraties occidentales. Il livrera donc ces informations précieuses sur la capacité effective des élites à percevoir correctement les préférences des citoyens. Ces résultats sont importants pour discuter de la qualité de la démocratie représentative.

Direct link to Lay Summary Last update: 17.07.2017

Responsible applicant and co-applicants

Employees

Abstract

Representative democracy depends on the ability of elected members of Parliament (MPs) to accurately perceive voters’ preferences. The project addresses this topic and formulates two questions: (1) to what extent do MPs have correct perceptions regarding the preferences of the population they represent (i.e. both the population at large and their party’s electorate)? (2) Why do some MPs hold more accurate perceptions than others? The aim of the study is to develop and test a novel theory of MPs’ perceptions that explains under which circumstances MPs’ difficult perceptual task is facilitated or obfuscated. The core of the theoretical framework is that MPs’ perceptions are a function of information at their disposal one the one hand, their own preferences on the other, and of the interaction between information and preferences. To assess empirically whether MPs hold accurate perceptions about the public’s preferences, the project draws on novel data from Switzerland (and Belgium as well, since Professor Stefaan Walgrave will conduct a similar project at the University of Antwerp). The methodological design consists of an online survey of a representative sample of the Swiss population, a face-to-face MPs survey, and the coupling of both. The surveys propose 15 to 20 specific policy issue statements (e.g. “Early release of convicted criminals should be made impossible”) grouped into five issue categories with varying levels of visibility, opinion distribution and national/regional competence, etc. (e.g. asylum seekers, austerity, crime, education, and environment). Respondents will be asked to (strongly) disagree/agree with each statement and whether the statement refers to a policy they consider being important. The exact same questions are used for MPs. In addition, for each statement, MPs are asked about their perception of the (general and party electorate) population’s distribution over different issue positions and salience levels. Furthermore, for each statement separately, the survey examines the sources of MPs’ information about the public’s preferences. MPs are asked ‘how they know’ about public opinion regarding that issue (e.g. via opinion polls) and which type of information they trust when it comes to assessing the voters’ preferences on this specific issue. With survey questions it is hard to assess to which information about the public’s preferences elites are exposed and how they process and adopt this information. Therefore, a number of survey-embedded experiments are implemented. The sample of MPs respondents is randomly split up in several experimental groups that are exposed, in an unobtrusive fashion (e.g. letting them read a newspaper article and ask them to evaluate it on several other dimensions) to different pieces of information or hints regarding popular preferences. Finally, a field experiment is added. The research team releases specific (and true) information about public opinion (based on the results of the population survey mentioned above) in the media (or directly to MPs) in the middle of the interview period. This should affect MPs’ ability to guess population preference regarding the issue at stake: the project can assess this effect by comparing the perceptions of MPs, who are interviewed before and after the information release. In sum, this project innovates by highlighting the informational determinants of individual MPs’ preference perceptions and by investigating how system-level variables (e.g. presence of direct democracy in Switzerland) impact guesses' accuracy. It is highly relevant from a theoretical, empirical and normative point view about democratic representation.
-