|About the Book|
The Transtheoretical Model (TTM) is an integrative model of intentional behavior change that has achieved widespread attention from researchers and practitioners regarding a variety of health-related behaviors. Measures that assess concepts from theMoreThe Transtheoretical Model (TTM) is an integrative model of intentional behavior change that has achieved widespread attention from researchers and practitioners regarding a variety of health-related behaviors. Measures that assess concepts from the TTM applied to alcohol use, such as the University of Rhode Island Change Assessment (URICA), Alcohol Abstinence Self-Efficacy (AASE), and Processes of Change Questionnaire (PCQ) have been used in various settings and populations, yet little research to date has directly examined the structural equivalence of these measures across gender or ethnic subgroups using factor analytic strategies. This study tested the factorial invariance of these measures across gender and three different ethnic subgroups (White, Black, Hispanic) using data from Project MATCH. The entire sample consisted of 1,726 participants, distributed as: Male (n = 1,307- 76%)- Female (n = 419- 24%)- White (n = 1380- 80%)- Black (n = 168- 10%)- Hispanic (n = 141- 8%). Results demonstrated factor loading and threshold invariance across both gender and ethnicity for the AASE measures, with specific items that comprise the Social/Positive and Negative Affect constructs being non-invariant. Factor loading and threshold invariance across gender was achieved using a condensed version of the PCQ, with specific items from the Experiential Processes construct being non-invariant, but factor loadings and thresholds were non-invariant across the ethnic subgroups. The URICA was considerably more problematic in terms of testing for invariance, as a close-fitting baseline model could not be achieved. However, known differences across setting type (e.g., outpatient vs. aftercare), as well as other sample size issues may have contributed to the lack of a close-fitting baseline model. These results suggest that in general, comparison of the AASE scales across gender and ethnicity are appropriate given the partial invariance of the scales. Eliminating several items from the PCQ would be needed to reliably compare process scale scores across demographic subgroups. Based on the results presented here for the URICA, comparison of stage scores across demographic subgroups is not recommended for the Project MATCH sample due to the setting type differences. However, this measure may still be a useful tool for assessing an individuals readiness to change.