| Session Title: Design and Construction of Measures |
| Demonstration Session 109 to be held in Panzacola Section H2 on Wednesday, Nov 11, 4:30 PM to 6:00 PM |
| Sponsored by the Quantitative Methods: Theory and Design TIG |
| Presenter(s): |
| Lee Sechrest, University of Arizona, sechrest@u.arizona.edu |
| Katherine McKnight, Pearson Corporation, kathy.mcknight@gmail.com |
| Mei-kuang Chen, University of Arizona, kuang@u.arizona.edu |
| Abstract: Measures used for evaluation should be deliberately and carefully conceptualized, designed, and constructed in order to provide the best possible information. That is true even if, eventually, measures are selected from among those already available. Three classes of measures can be identified: common factor measures, emergent variable measures, and quasi-latent variable measures. Each of these classes of measures entails particular assumptions about their structure and require distinctly different approaches to their construction and interpretation. The three classes of measures are defined and differentiated, and the implications of their definitions for reliability and validity are outlined. Procedures for the construction and assessment of each of the three types of measures are presented and illustrated with simulated and real data. Their use in evaluation and their proper interpretations are discussed. |