Measuring and Reducing Inconsistency Among Questionnaire Items Through Imputation: An Application to the NSOPF
Ault, K. L., Fahimi, M., & Heuer, R. E. (2005, August). Measuring and Reducing Inconsistency Among Questionnaire Items Through Imputation: An Application to the NSOPF. Presented at Joint Statistical Meetings, Minneapolis, MN.
For complex surveys the task of imputing a large number of variables is a major undertaking, since the resulting data must satisfy various consistency checks that are often intertwined. For instance, values of certain variables might have to add up to those of others, while in other cases certain variables might restrict the values other variables can take on. When an observed/imputed of an item is deemed inconsistent, oftentimes logically imputation is used to resolve (edit) the observed inconsistency. Alternatively, such values are set to missing and then imputed to avoid the preceding laborious process. This work provides an overview of a methodology for measuring that can be used for imputation of a large number of items and discusses a technique for measuring and reducing the number of inconsistent cases through imputation. Missing data are imputed using a weighted sequential hot-deck methodology, while ensuring that all values are consistent with respect to all known skip patterns and logical constraints. The research is based on the 2004 National Study of Postsecondary Faculty (NSOPF:04) survey data.