Abstract
A study on response stability when a subject encounters personality inventory items is re-examined in terms of the latent trait model used. This re-examination shows, firstly that the study demonstrates an important type of validity for the application of latent trait models to such items, and secondly that the original recommendations for test construction based on the study require modification. A key aspect of the paper is an amplication of the role that sources of error must play in latent trait models.
Original language | English |
---|---|
Pages (from-to) | 253-262 |
Number of pages | 10 |
Journal | EDUCATIONAL AND PSYCHOLOGICAL MEASUREMENT |
Volume | 41 |
Issue number | 2 |
DOIs | |
Publication status | Published - 1 Jan 1981 |
Externally published | Yes |