This paper is a contribution to the ‘know your data’ approach to the issue of measuring corruption. It presents findings of mixed-methods research focused on the way perceptions of corruption are formed, and how this process impacts our efforts to measure corruption. The conclusions of the first, quantitative part of the research, are explored in more depth in the second, qualitative study. The first part of the paper includes the results of a survey experiment on how the formulation of questionnaire questions influences the outcomes of surveys. In three parallel surveys conducted in May and June 2019 in the Czech Republic, respondents were asked to estimate how many politicians and civil servants are involved in corruption. The first (pilot) survey was conducted on students; the second was a survey on a representative sample of the public and the third on businesspeople. Within each survey, respondents were randomly divided into four groups, and each group was presented with differently formulated questions (on bribery, corruption, corruption with definition and corruption with basic definition including examples). The second part of the paper analyses findings from a focus group study (multiple focus groups on the same topic, both with students and non-students), in which participants discuss how they formulate their perceptions of corruption and how the various formulations of questionnaire questions impact their responses. The research deepens our understanding of what indicators measuring perceptions of corruptions actually reflect and enables their better interpretation.
Dr. Aneta Pinkova