Recall Errors in Surveys

This paper considers measurement errors from a different perspective. Concentrating on errors that are caused by the insufficient ability of individuals to acquire, process and recall information, we devise a structural model for the response behavior of individuals in surveys. Our aims are twofold: Using nonseparable models, we first explore the conse-quences of such a modeling approach for key econometric questions like the identification of marginal effects and economic restrictions. We establish that under general conditions recall measurement errors are likely to exhibit nonstandard behavior, in particular be non-classical and differential, and we provide means to deal with this situation. Moreover, we obtain surprising findings indicating that conventional wisdom about measurement errors may be misleading in many economic applications. For instance, under certain conditions left hand side and right hand side errors will be equally problematic. The second aim of this paper is to provide a formal framework for understanding the actual response behavior of respondents in surveys. Specifically, we provide formal arguments and explanations for the frequently encountered substantial effects of the design of a survey. Also, we suggest a list of issues that should be considered when eliciting a survey to avoid adverse effects. Finally, we apply the main concepts put forward in this paper to real world data and to a controlled experiment, and find evidence that underscores the importance of focusing on individual response behavior, as well as taking the design of a survey seriously