Using Person Fit to Explore Misreporting in Surveys

• In Event: Rasch Models: Considerations and Novel Uses

Abstract

The purpose of this study is to develop a conceptual framework for identifying misreporting in surveys. Surveys are widely used in research in the social sciences. An important issue to consider in survey research is the identification of aberrant responses. Aberrant responses may reflect misreporting in surveys. Misreporting may occur for a variety of reasons ranging from lack of knowledge to deception. Person fit analyses based on Rasch measurement theory offer an approach for identifying misreporting in surveys. The framework is illustrated with a case study of a survey designed to identify food insecurity that is used by the USDA (N=200,959). After misfitting persons are identified, the next steps are diagnosing and categorizing the reasons for the unusual responses.

Authors

• Jing Li, University of Georgia
  Presenting Author

• George Engelhard, University of Georgia
  Presenting Author