close× Call Us
Title Voting intentions versus participation in voting
Post date 12/06/2018
C1 Background and Explanation of Rationale When scholars analyze voting behavior at the individual level, they usually have to rely on self-reports of past and intended voting. However, it is widely known that these responses suffer from recall-bias: that respondents over-report their past votes, and overstate the strength of their voting intentions. But by how much? We use a unique dataset to answer this question. We conducted a large-scale (n=2,500) survey in Armenia prior to a crucial parliamentary election. This together with unique policy of obligatory publishing of the names of all voters, allows us to answer these fundamental questions.
C2 What are the hypotheses to be tested?

1. Do people over-report voting intentions and past voting behavior. If yes, is there a difference between the misreporting of voting intentions and past behavior?
2. Is “do not know” option in the question about voting intentions is an excuse to not the say “known “no”? We hypothesize that the share of nonvoters among those who answered “do not know” will be as high as the share of nonvoter who answers “not going to vote”.

C3 How will these hypotheses be tested? * We can answer Question 1 by analyzing the distribution of the misreporting variables, and test the second part of question 1 using simple proportion tests of, and Fisher exact tests, with treatment being the year of elections. The hypothesis in question 2 is also directly tested through Fisher tests of participation_2018 with treatment variable being the answer “no” or “do not know” in voting_intentions survey question.
C4 Country Armenia
C5 Scale (# of Units) 2500
C6 Was a power analysis conducted prior to data collection? No
C7 Has this research received Insitutional Review Board (IRB) or ethics committee approval? Yes
C8 IRB Number IRBWZB101018.1
C9 Date of IRB Approval 10.10.2018
C10 Will the intervention be implemented by the researcher or a third party? Researchers
C11 Did any of the research team receive remuneration from the implementing agency for taking part in this research? No
C12 If relevant, is there an advance agreement with the implementation group that all results can be published? Yes
C13 JEL Classification(s) not provided by authors