A paper published last month in Physical Review Physics Education Research suggests that all differences in student performance between demographic groups in introductory calculus-based physics classes can be explained by differences in pre-college preparation.
Introductory calculus-based college physics provides an important scientific foundation for students majoring in engineering, physics and computer science – all disciplines in which underrepresentation of women and students of color is severe.
The authors, who are located at Cornell University, the University of Colorado and Stanford University, conducted the study at three universities that were not named in the paper. However, one of the universities was described as a highly selective university on the east coast of the US, a second was a highly selective university on the west coast, and the third was a large public university “in the middle of the country”.
The study was conducted using a statistical analysis of course performance and preparation data for more than 5,000 students who took a first-semester calculus-based physics class between 2012 and 2018.
According to the authors, the difference in course performance between women and men – which was significant at the institutions participating in the study – can be entirely explained to be the result of a difference in pre-college physics preparation. In this study, pre-college physics preparation was measured by taking one of two widely-used tests of the conceptual understanding of Newton’s Laws of Motion at the beginning of the college physics course. On the average, women scored significantly lower than men on these pre-tests.
Performance gaps suffered by members of underrepresented minority groups and first generation college students could be entirely explained by invoking a combination of the pre-test of Newton’s Laws understanding and ACT/SAT scores.
The authors motivated their study by noting:
Of particular concern to many institutions today are the average gaps in performance often seen between different demographic groups, such as course grades, exam scores, and passing rates. Underperformance of demographically underrepresented students in physics 1 can have considerable negative influence on their prospect of pursuing STEM fields, thereby preventing the increase of their representation in those fields. The factors that give rise to such gaps and how we can best design learning environments to address them are important unanswered questions. It is important to identify and remove factors that might produce such gaps, but there is also a danger associated with focusing on such gaps. There are negative consequences to labeling gaps as demographic gaps when the gaps are not arising from demographic status per se, but from the factors correlated with it. This mislabeling can result in bias and negative expectations for the labeled demographic group by both instructors and students.
The authors, who included Nobel Laureate and science education advocate Carl Wieman, argued that university physics faculty should accept the challenge of solving the problems caused by gaps in the preparation of students for college physics:
This work shows that creating instruction that enhances the success of every student across the full range of incoming preparations is also the solution to eliminating gaps in the performance across demographic groups. Future work will determine how to best do this, but we can offer some potential suggestions. Better matching the introductory course to the range of background preparations of the student population would likely ensure that many more students, particularly those who had the misfortune to attend K–12 schools that provided weaker education in STEM in general and physics in particular, would achieve better outcomes.