Summary

The article discusses the use and implications of GPA as a metric for student achievement, examining variations in GPA across majors, age groups, and gender, and its correlation with high school GPA and SAT scores, while also addressing concerns of grade inflation and gender disparities in certain academic fields.

Abstract

The article titled "GPA: What Does This Mysterious Metric Teach Us?" delves into a study by Walstad and Bosshardt (2019) that scrutinizes the Grade Point Average (GPA) as a measure of student performance in higher education. Despite the limitations and noise in GPA data due to diverse courses, grading methods, and grade inflation, the authors argue that GPA still provides valuable insights into the relative academic standings of students. The study uses a nationally representative dataset of U.S. college graduates from the 2007-08 academic year, excluding dropouts, to analyze GPA trends. Findings reveal an overall undergraduate GPA of 3.24 on a 4-point scale, with Education and Health Care majors at the higher end and General Studies and Marketing at the lower end. Economics, the authors' primary focus, has an average GPA of 3.16. The article also explores how age correlates with GPA, with younger and older graduates generally having higher GPAs than those in the 22-25 age range. High school GPA and SAT scores are found to be good predictors of college GPA. Gender differences in course completion and grades are also observed, with females more likely to complete courses in certain subjects like Biology but less likely in Business, Calculus, Economics, and Engineering. The study suggests that interventions by organizations like the American Economic Association to increase diversity in fields like Economics are well-founded, given the gender disparities evident from the first course grades.

Opinions

The authors consider GPA a useful, albeit imperfect, tool for assessing student achievement, acknowledging its limitations but affirming its value in ranking students when academic norms are stable.
There is an implied concern about the heterogeneity of factors influencing GPA, such as the variability in course content, class sizes, difficulty levels, instructor grading practices, and student abilities and efforts.
The authors note the issue of grade inflation, which is not directly addressed in the study but is suggested by the overall high GPAs reported.
The data suggests a pattern where younger and older graduates tend to have higher GPAs, which may reflect differences in life circumstances, maturity, and motivation.
The authors seem to validate the predictive power of high school GPA and SAT scores for college academic performance.
The article highlights gender disparities in course enrollment and performance, particularly in STEM fields, supporting the need for initiatives to increase diversity in these areas.

GPA: What Does This Mysterious Metric Teach Us?

A Discussion of “Grades in Economics and Other Undergraduate Courses”

The Grade Point Average (GPA) is the metric we use to represent student achievement, but should we? It serves as a measure for who qualifies for honors, scholarship,s internships, and job offers. That is a lot riding on one number.

The reality is there is a lot of heterogeneity across students all wrapped up in that single number.

Students take different courses with different assessment methods.
Some courses are large and impersonal; some are small and personalized.
Courses differ in their level of difficulty.
Instructors differ in their grading practices.
Students differ in their ability and effort level.
And there is much discussion of grading standards evolving over time, that is, grade inflation.

Even given all of that, Walstad and Bosshardt (2019) assert that while GPA is a “noisy signal,” it does “contain valuable information on the relative ranking of students when academic norms are fairly stable.” (p. 266)

In this paper, the authors are accepting the GPA as a useful metric to examine how the average GPA varies across majors and if there are any gender differences across subjects, using a government data set that includes “a large, nationally representative sample of US college graduates in the 2007–08 academic year.” (p. 267)

One note, the data only includes college graduates. Any students who drop out before graduation are not in the data set. This would make all GPA averages reported higher than they really are because dropouts would be expected to pull the averages down. (p. 267)

All the GPAs are reported as a 4-point scale with an A as 4.0, B as 3.0, C as 2.0, and D as 1.0.

The overall undergrad GPA across all majors is 3.24, which is between a B and a B+ (3.3).
The highest GPA is in Education at 3.43 followed by Health Care at 3.38.
The lowest GPA is in General Studies at 3.10 with the next lowest being Marketing at 3.13.
Economics, the particular interest of the authors who teach economics, came in at 3.16.

They investigate some other ways to break down the data. The impact of age on GPA is not particularly surprising.

College graduates under the age of 22 had the highest GPA at 3.36.
Those college graduates aged 22–23 have a GPA that drops to 3.19.
Those college graduates aged 24–25 have the lowest at 3.09
Those college graduates who are over 25 increase significantly to 3.29.

This is not surprising to me because it seems reasonable that the pool of students who graduate before turning 22 would be skewed towards high achievers. While the pool of students who graduate at age 24–25 is going to be skewed to students who have struggled in college for one reason or another.

Those age 26 or older are adult learners who are generally more focused and self-motivated than their younger counterparts.

More interesting is the analysis that seems to bolster the validity of GPA being used as a metric of student achievement and the SAT test as a predictor of college performance.

Students who had a high school GPA of 3.5 or higher also had a higher than average college GPA at 3.37 (compared to the overall average of 3.24).
Students who score 540 or higher on the SAT verbal section have a 3.37 average GPA.
Students who score 540 or higher on the SAT math section have a 3.35 average GPA.

Next, the authors turn to discovering any gender differentials in the data. They look into completion of at least one class in a particular subject regardless of their major. (p. 269)

Females are 11–15% more likely to complete a course in Biology, Foreign Languages and Psychology.
And females are 11–16% less likely to complete a course in Business, Calculus, Economics, and Engineering.

Finally, they look at the grade earned in the first course in these different subjects with the idea that those who make A’s would be more likely to continue taking more classes in that area.

While females are less likely overall to take quantitative or analytical courses, with the exception of biology, those that do enroll in Business, Calculus, Economics and Engineering do make more A’s than the males in Business and Calculus.

That means females are not making as many of the highest grades in Economics and Engineering.

Interestingly, the economics field is on a soul-searching path to discover why it is less attractive to females to see if they can increase its diversity.

In 2019, the American Economic Association put together a 4 part plan to increase diversity in the economics field by addressing issues around research, collegiate behavior, students, and recruitment and promotion.

We see here in this study the authors are discovering there is a gender difference appearing in the first class so it seems the AEA is on the right path.

References:

Walstad, William and William Bosshardt (2019). “Grades in Economics and Other Undergraduate Courses.” AEA Papers and Proceedings, 109: 266–270.