[Paper Review] Implementing Competency-Based Grading Improves the Performance of Women and First Generation Students in Introductory Physics

M. S. Richard, Jennifer Delgado+2 more|arXiv (Cornell University)|Jun 21, 2022

Mathematics Education and Programs4 citations

TL;DR

This study implements a competency-based grading system (ALPaCA) in calculus-based introductory physics at the University of Kansas, allowing students multiple opportunities to demonstrate proficiency across content areas through repeated assessments. The system significantly improved performance—especially for women and first-generation students—by reducing stress and accommodating diverse learning paces, with women showing the largest gains in course performance and DFW rates decreasing across all groups.

ABSTRACT

We present a model for competency-based grading for calculus-based introductory physics that encourages students to obtain proficiency with all course content. By allowing students to continually improve their proficiency with skills and content throughout the semester, this formative grading system is designed to create a more flexible learning environment that better accommodates the varying schedules and needs of students. While all students show improvement in their performance following the implementation of this grading system, the largest gains were found for women and first generation students, both of whom often pose a retention risk in science and engineering degree programs.

Motivation & Objective

• To develop and implement a flexible, competency-based grading system in calculus-based introductory physics that supports diverse student learning pathways.
• To address performance disparities in STEM by reducing barriers related to time constraints, prior preparation, and stereotype threat.
• To evaluate whether multiple assessment opportunities improve student outcomes, particularly for underrepresented and at-risk groups such as women and first-generation students.
• To quantify the relative difficulty of different physics content areas to inform instructional adjustments and curriculum design.
• To explore the potential of competency-based grading as a foundation for a new course prerequisite model based on demonstrated proficiency rather than course completion.

Proposed method

• Design a formative grading system (ALPaCA) that requires students to demonstrate proficiency on more than one assessment to ensure retention and mastery.
• Allow multiple attempts across the semester for each content area, including weekly quizzes and a final exam, to accommodate varying learning paces.
• Calculate course grades using the geometric mean of the highest scores from at least two assessments per content area, including the final exam.
• Use institutional data to categorize students by gender, first-generation status, and underrepresented minority (URM) status for cohort-based performance analysis.
• Administer content-specific quizzes and exams focused on individual learning objectives, with a shift from longer exams to shorter, frequent assessments in later iterations.
• Retain group work in instruction but use individual assessments for grading to ensure individual accountability and accurate proficiency measurement.

Experimental results

Research questions

• RQ1 How does a competency-based grading system with multiple assessment opportunities affect overall student performance in calculus-based introductory physics?
• RQ2 Do women and first-generation students experience disproportionately larger performance gains under this grading model compared to their peers?
• RQ3 Which content areas in introductory physics (e.g., force-based mechanics, energy-based mechanics, momentum) are most challenging for students, and how do these difficulties vary across student cohorts?
• RQ4 To what extent does the ALPaCA grading system reduce the DFW (D, F, Withdrawal) rate, particularly among at-risk student populations?
• RQ5 Can proficiency in specific content areas of PHSX 211 predict performance in subsequent courses, and how might this inform a new prerequisite model based on demonstrated mastery?

Key findings

• The implementation of the ALPaCA competency-based grading system led to significant improvements in course performance across all student groups, with the largest gains observed among women and first-generation students.
• Women showed the most substantial relative improvement in course performance, suggesting the system effectively mitigated gender-based performance disparities common in physics courses.
• First-generation students also experienced the largest performance gains, indicating that the system helped reduce retention risks associated with lack of academic socialization and prior STEM exposure.
• The DFW rate decreased across all cohorts, with the most notable reductions observed among women and first-generation students, reflecting improved course persistence and success.
• Students struggled most with momentum-based quantitative problem solving, followed by force-based mechanics, while energy-based mechanics was perceived as less difficult, informing targeted instructional improvements.
• The system’s design—using the geometric mean of the two highest scores per content area—ensured that proficiency was demonstrated across multiple assessments, supporting long-term retention and reducing the impact of single high-stakes failures.