Multidimensional measurement of situational interest in physical education: Application of bifactor exploratory structural equation modeling

Multidimensional measurement is a common theme in motivation research because many constructs are conceptualized as having an overarching general factor (e.g., situational interest) and specific dimensions (e.g., attention demand, challenge, exploration intention, instant enjoyment, novelty). This review addresses current issues associated with the multidimensional measurement of situational interest in elementary physical education (PE) and illustrates the application and benefits of bifactor exploratory structural equation modeling (ESEM). I perform secondary analysis on a large, previously published data set used to provide validation support for the Situational Interest Scale for Elementary PE. Findings clearly demonstrate the advantages of capturing the multidimensional nature of situational interest using bifactor ESEM. Specifically, a more accurate measurement model of situational interest is reproduced using bifactor ESEM compared with other techniques such as first-order and second-order confirmatory factor analysis. There is empirical support for an overall general factor of situational interest when using the Situational Interest Scale for Elementary PE, however, examining the five dimensions of situational interest as unique factors after accounting for the general factor does not appear warranted.