Quantifying the Influence of Partial Factorial Invariance on Reliability: A Proposed Effect Size

• In Event: 65.080 - Roundtable Session 24
  In Roundtable Session: 65.080-18 - Reliability: Methods and Application

Abstract

Reliability is a central concept in educational and psychological measurement. Accurate estimates allow for a determination if the level of consistency is appropriate for decisions at the person or group level. Measurement invariance also is a central concept in measurement. In the absence of invariance, score validity is compromised. It is not clear how a lack of invariance influences reliability estimates. This study introduces an effect size measure to quantify the effect of noninvariance on reliability estimates. Results indicate that the effect size does successfully quantify the lack of group parameter invariance, providing an easily interpretable index that researchers can use to determine the extent to which the lack of invariance has impacted their estimation of scale reliability.

Authors

• Brian F. French, Washington State University

• William Holmes Finch, Ball State University