Measuring Political Participation. Testing Social Desirability Bias in a Web-Survey Experiment

• How reliable are the measures of political participation used in the major social science surveys? We know that there is considerable variation in estimated levels of political participation in different surveys, but we do not know whether this is due to item constructions, survey mode or the level of non-responses. In this paper, we investigate whether different item constructions affect the estimated levels of reported political participation. We use an embedded web-survey question wording experiment to isolate the effect of item construction. We test three different question wordings with introductions (a) expressing social desirability, (b) normalizing inactivity, and (c) a control group without any introduction to the question in combination with two different sets of response options (with and without face saving options) in a full factorial design, including six treatments in total. In addition, we evaluate which citizens who tend to overreport their level of political participation when different item constructions are applied. Our results show that applying different question introductions do not have any effect on the reported levels of participation. However, including face saving response alternatives considerably decrease reported levels of participation. Moreover, we find no evidence that particular groups tend to overreport their political activity more than others. Rather, respondents who lie about their level of participation can be found among all groups of citizens.

Subject headings

SAMHÄLLSVETENSKAP  -- Statsvetenskap -- Statsvetenskap (hsv//swe)

SOCIAL SCIENCES  -- Political Science -- Political Science (hsv//eng)

Publication and Content Type

vet (subject category)

kon (subject category)