Using adjusted GPA and adjusted course difficulty measures to evaluate differential grading practices in college

Area of Science:

• Educational Measurement
• Psychometrics
• Higher Education Administration

Background:

• Traditional grade point averages may not fully capture student achievement or course complexity.
• A need exists for more nuanced analytical tools in higher education to evaluate academic performance and institutional effectiveness.

Purpose of the Study:

• To apply the Rasch rating scale model to a large dataset of college freshman grades.
• To derive and compare measures of student achievement and course difficulty using the Rasch model versus simple grade averages.
• To explore the utility of the Rasch model for addressing administrative concerns in higher education.

Main Methods:

• Utilized the Rasch rating scale model (Andrich, 1978) on grade data from four freshman cohorts at a large public university.
• Analyzed data representing approximately 34,000 students, 1,700 courses, and 119 departments.
• Derived departmental difficulty indices through secondary analysis of course difficulty measures.

Main Results:

• The Rasch rating scale model generated distinct measures for student achievement and course difficulty.
• Significant differences were observed between Rasch model measures and traditional simple grade averages for students, courses, and departments.
• Departmental difficulty indices were successfully derived from course difficulty measures.

Conclusions:

• The Rasch rating scale model provides a more sophisticated assessment of student achievement and course difficulty than simple grade averages.
• The identified differences suggest the Rasch model's potential as a valuable tool for college administrators.
• This approach can inform various administrative decisions related to academic assessment and resource allocation.