Math Homework Purpose Scale: Measurement Invariance and Latent Mean Differences across Gender

The present study evaluated the psychometric properties of the Math Homework Purpose Scale (MHPS). After randomly splitting the sample (N = 854) into two subsamples (n = 427 and n = 427), we carried out exploratory factor analysis (EFA) on subsample 1 and confirmatory factor analysis (CFA) on subsample 2. The Kaiser-Meyer Olkin value of .88 indicated that EFA was appropriate for subsample 1. EFA results showed that a three-factor solution explained 72.2% of the variance. All ten of the MHPS items loaded quite substantially on three factors labeled Academic, Self-Regulatory, and Approval-Seeking. CFA results further showed that the MHPS was composed of three factors: Academic, Self-Regulatory, and Approval-Seeking. From this adequate level of measurement invariance, we further examined the latent mean difference across gender for the entire sample. Our findings showed no statistically significant mean differences in Academic, Self-Regulatory, and Approval-Seeking across gender. In addition, the HMPS was found to have adequate alpha reliability coefficients and omega coefficients. Concerning the validity evidence of the MHPS, Academic, Self-Regulatory, and Approval-Seeking purposes were positively correlated with homework effort, homework completion, and math achievement. The present study provides solid evidence that the MHPS is a reliable and valid measurement of math homework purposes.