DIGSSCORE Seminar: How to Measure Second-Order Beliefs

Ingrid Kvåle Faleide, PhD candidate at the Department of Government, University of Bergen will present for us today. Her presentation is titled "How to Measure Second-Order Beliefs".

The event is in a hybrid format, you are welcome to join us in the Corner room at DIGSSCORE. Bring your own lunch, we serve coffee, tea and water. Zoom link for digital attendance.

Abstract:

Public opinion research is a cornerstone of political science. Within this field, growing attention is now directed toward the less-explored area of perceptions of public opinion. This emerging field is underdeveloped both theoretically and methodologically, despite the crucial role that beliefs about public opinion play in shaping political attitudes, outcomes, representation, and public debate. Political scientists increasingly treat these second-order beliefs as both independent and dependent variables. However, shifting the focus from measuring individuals' own opinions to what they believe others think presents unique methodological challenges. This paper reports findings from a series of experimental studies that explore the optimal approaches for measuring second-order beliefs in surveys. Through randomized split-sample designs, the study compares different response scales, examines potential spillover effects between first- and second-order beliefs, and evaluates the influence of question placement within a survey questionnaire. The findings demonstrate that the nature of the perceptual task significantly shapes the conclusions, making the choices during the design phase crucial to the interpretation of results. However, concerning the issue of heightened projection bias when recording both first- and second-order beliefs within the same individuals in the same survey, the results are more reassuring. The effects are minimal, suggesting that researchers need not be overly concerned about its impact on their findings. The paper contributes to a better understanding of these methodological challenges and offers practical guidance for improving the study of second-order beliefs in survey research.