Construct and score comparability are central to interpreting test results to make valid and meaningful inferences. Many international large-scale surveys are tested worldwide in order to gain a cross-cultural perspective. Cross-cultural comparison requires that the measuring instrument and the construct being measured should be invariant across culture groups. The Program for International Student Assessment (PISA) is one of the largest cross-cultural study program up to date and provides a good database for cross-cultural study. The present study intends to examine the measurement invariance of the measures of motivation and engagement between the United States and China using PISA 2012, and to study how these constructs relate to the mathematic achievement. All the students (N = 6,703) from the United States and China who participated in the PISA 2012 and took the student questionnaire about motivation and engagement were included in the study. The 25 items testing intrinsic and instrumental motivation, math self-efficacy, and math engagement/behaviors were studied. A structural equation modeling (SEM) based measurement invariance approach was utilized. The results indicated that the measures of interest had a partial scalar invariance. The factorial structures, factor loadings, and most item intercepts were equivalent between the United States and China. These findings indicated that the measures were comparable between the two countries except for a few items. In addition, the math self-efficacy and math behaviors were found to significantly relate to the math achievement, but significant relations between intrinsic motivation and math achievement were found only for US students, and significant relations between instrumental motivation and math achievement were found only for Chinese students.

Keywords:

Cross-Cultural Research, Measurement Invariance, Motivation, Self-Efficacy, Engagement