Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Date of Award
Master of Arts (MA)
Department of Graduate Psychology
The current paper reviewed two widely used approaches to assessing construct stability over two time points (rank-order and mean-level consistency), as well as common misconceptions about what each indicates. In addition, the application of longitudinal mean and covariance structures (LMACS) analysis as a modern approach to assessing construct stability was explained and demonstrated by assessing the stability of psychological entitlement over 1.5 years measured via the Psychological Entitlement Scale (PES). Confirmatory factor analysis supported a one-factor solution for the PES at both time points, and reliability of scores was adequate (ω = .88 and .89). Full configural and metric invariance and partial scalar invariance were established for the PES. Rank-order consistency of factor scores was moderate (r = .61) and the latent mean difference in psychological entitlement across time was not statistically significant. Results provided construct validity evidence for the PES regarding measurement invariance and also indicated that psychological entitlement tended to be stable on average but not at the individual level over 1.5 years. Discussion of the effects of differential item functioning (DIF) of scalar non-invariant items on mean difference testing and use of the PES with observed scores, the change in psychological entitlement at the individual level, and the advantages of LMACS analysis as a unified approach to assessing construct stability is also provided.
Bashkov, Bozhidar Mihaylov, "Using longitudinal mean and covariance structures (LMACS) analysis to assess construct stability over two time points: An example with psychological entitlement" (2012). Masters Theses. 144.