Factor Analysis with Ordered Categorical Indicators and Measurement of Self-Efficacy in Physical Activity Contexts: A Substantive-Methodological Synergy

ABSTRACT

Factor analysis of ordered categorical indicators in kinesiology is pervasive. However, some assumptions made regarding the indicators within methods commonly used in this research may increase analytic errors. Methods that relax these assumptions have been available for decades, but uptake has been slow. Therefore, the methodological focus of this tutorial was factor analysis of ordered categorical indicators. Self-efficacy has been one of the most important psychosocial constructs in physical activity research for decades. Unfortunately, sub-optimal practices in constructing self-efficacy scales in physical activity contexts are frequently observed and decrease validity for scores produced from responses to these instruments. Guidelines for constructing self-efficacy scales have been available for decades, but adherence remains inconsistent. Therefore, the substantive focus of this tutorial was measurement of self-efficacy in physical activity contexts. The synergy within this tutorial was factor analysis of simulated ordered categorical responses to self-efficacy indicators. Annotated files from Mplus and R are provided.

KEYWORDS:

• structural equation modeling
• ordinal variables
• binary variables
• self-efficacy theory
• physical activity promotion

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/1091367X.2023.2186789.