Using Rasch Analysis to Assess Students’ Learning Progression in Stability and Change across Middle School Grades

This study developed a measurement instrument to assess students’ learning progressions in the crosscutting concept of Stability and Change across middle school grades (from Grades 7 to 9). Based on existing concept development models, frameworks, and research on student learning progressions, this study’s learning progression framework comprises four primary levels (i.e., Identifying, Understanding, Analyzing, and Designing) with three sub-levels (i.e., Static, Dynamic, and Cyclic), from basic to the most sophisticated. During the field test, three versions of the test comprising a total of 24 constructed-response items were administered to 136 seventh graders, 139 eighth graders, and 67 ninth graders. A partial credit Rasch model analysis was employed to inform instrument development and evaluation. Specifically, this study used step calibrations and item measures anchoring to express student performance across three grades on the same linear scale. Results provided evidence of reliability, content validity, construct validity, and predictive validity of measures of the instrument, suggesting the measurement instrument meets the quality benchmarks. The results illustrated that higher-grade students were more proficient than lower-grade students in Identifying, Understanding, Analyzing, and Designing regarding Stability and Change. None of the seventh graders and less than 5% of eighth graders were proficient at Cyclic level in Understanding, Analyzing, and Designing, whereas between 3% and 16.4% of ninth graders were proficient at level 3 in Understanding, Analyzing, and Designing.