Panel conditioning arises if respondents are influenced by participation in previous surveys, such that their answers differ significantly from the answers of individuals who are interviewed for the first time. Having two panels—a trained one and a completely fresh one—created a unique opportunity for analysing panel conditioning effects. To determine which type of question is sensitive to panel conditioning, 981 trained respondents and 2809 fresh respondents answered nine questions with different question types. The results in this paper show that panel conditioning only arise in knowledge questions. Questions on attitudes, actual behaviour, or facts were not sensitive to panel conditioning. Panel conditioning in knowledge questions was restricted to less-known subjects (more difficult questions), suggesting a relation between panel conditioning and cognition.
|Place of Publication||Tilburg|
|Number of pages||8|
|Publication status||Published - 2008|
|Name||CentER Discussion Paper|
- panel conditioning
- measurement error
- panel surveys
Toepoel, V., Das, J. W. M., & van Soest, A. H. O. (2008). Relating Question Type to Panel Conditioning: A Comparison between Trained and Fresh Respondents. (CentER Discussion Paper; Vol. 2008-4). Econometrics.