a PSC Research Project [ARCHIVE DISPLAY]
Investigator: Roger Tourangeau
This study will advance the knowledge about successful and efficient strategies for questionnaire development. It proposes a series of studies to investigate the use of latent class analysis (LCA) to assess questionnaire items. The LCA models can be used when two or more measures are available for the same latent class variable. The models can often be applied to extant data, since some surveys include multiple measures for key constructs and others routinely collect reinterview data. Surveys, and with them questionnaires, are used across many disciplines, some of which don't have access to large resources for questionnaire development and questionnaire testing. Finding a method that can draw knowledge about question performance from existing data (such as reinterview data) will be particularly useful.
|Funding:||National Science Foundation (SES 0549916)|
Funding Period: 03/01/2006 to 02/28/2010
This PSC Archive record is displayed for historical reference.