Psychometric evaluation of the power of food scale in a diverse college sample: Measurement invariance across gender, ethnicity, and weight status

Abstract

The Power of Food Scale (PFS) is an instrument designed to examine individual differences in the drive to eat for pleasure (as opposed to in response to physiological hunger) and the effect of living in an obesogenic environment. Previous research supports the validity and reliability of the PFS, however, it had yet to be validated in an ethnically diverse college sample. The purpose of the current study was to test the factor structure and measurement invariance of the PFS across gender, ethnicity, and weight status. A sample of 432 college students completed the PFS (males=113, females = 319; non-Hispanic white=181, Hispanic=251; non-overweight=302, overweight=130). Confirmatory factor analysis was used to test a second-order, 3-factor (food available, food present, food tasted) structure of the PFS in each group separately (males, females, Hispanic, non-Hispanic white, non-overweight, and overweight) and tests of measurement invariance were conducted to test the equivalency of the measure across gender, ethnicity, and weight status. Results supported the measure’s original factor structure (second-order, 3-factor model) and indicated that the measure is equivalent across each of these groups, respectively. Although the small, unbalanced groups may impact the stability of the findings, the results nonetheless suggest that the PFS is a psychometrically valid measure in a diverse college sample, and that mean comparisons on this measure are meaningful across gender, ethnicity, and weight status. Given the measurement invariance of the PFS, there is support for use of the PFS among diverse college students in future work.