Grade inflation and college graduation

The number of students graduating from colleges rose from 1990 to 2010. There are several possible explanations for this. Perhaps the incoming students were better prepared than in the past. Perhaps colleges have improved teaching so that, on average, students learn more than they did in the past. Or perhaps the rise in graduate numbers mostly reflects grade inflation.

The last explanation is like J.Effrey T. Denning, Eric R. Eide, Kevin J. Mumford, Richard W. Patterson, and Merrill Warnick in their research paper “Why have college graduation rates increased?” (American Journal of Economics: Applied Economics, July 2022, 14:3, p. 1-29, subscription required).

This figure shows the increase in college graduates (within six years of enrollment), with the black dotted line showing the overall average and the colored lines showing different groups of schools.

Can better prepared incoming students explain this trend? Probably no. As the authors note, the proportion of students attending college has risen over time, which likely means that it is now being done by less qualified students who have not attended college in the past. Good for them! But this means that a theory based on the idea that middle-income students are better prepared is unlikely. The average math and reading test scores of 17-year-olds on the nationally representative National Assessment of Educational Progress (NAEP, sometimes referred to as the “national report card”) also did not rise during this interval. There does not appear to be a shift towards majors with higher graduation rates over time. The partial evidence that exists suggests that college students spend fewer hours studying than in the past, but more hours of paid work outside the classroom. As the authors write:

We discuss current trends that may affect college graduation, such as college salary supplement, enrollment, student preparation, study time, college employment, price, government support for higher education, and college attendance. Trends in these
variables will predict a decline in college graduates almost uniformly. … [L]Longitudinal datasets at the student level contain information about high school backgrounds, academic preparation, college enrollment, and graduation outcomes. We find that student characteristics, institutional resources, and school attendance do little to explain the change in graduation rates. … In other words, similarly trained students from later cohorts of the same zip code, same gender and race, with the same initial courses, the same major, and in the same institution have a higher first-year GPA than the earlier cohorts.

But along with the number of college graduates, college grades are also rising in different types of colleges (with the exception of commercial schools), suggesting the possibility of inflated grades.

Faculty, departments, and institutions may have incentives to inflate grades or raise GPAs. [grade point averages] for reasons unrelated to student achievement. Teachers who give students higher marks receive higher teacher grades (Krautmann and Sander, 1999; Langbein, 2008). Higher grade departments enroll more students (Butcher, McEwan, and Weerapana, 2014). In addition, colleges have strategic incentives to provide less informative grades (Boleslavsky and Cotton 2015), and institutional efforts to contain grade inflation may fail to make transcripts more informative and instead result in lower welfare (Bar, Kadiyali and Zussman 2012). The increase in grades explains, in a statistical sense, most of the changes in the number of graduates in our decomposition exercise.

How can one find evidence for the tricky question of whether higher grades and graduation rates are simply a reflection of students learning more, or is it inflated grades? The authors have detailed data on students from an unnamed “public liberal arts college” whose grade and graduation trends are similar to those of a larger sample. At this school, they have access to the results of the final exams, as well as the grades given for the classes. In some courses, the same (or very similar) GCSEs were taken over time: as far as the authors can tell, GCSEs do not get harder over time. They write: “Students with the same scores on the same final exam received higher marks in subsequent years. Our finding that grades improve over time, even when considering student characteristics and performance on the same comprehensive GCSE, suggests that public liberal arts college degree standards are softening over time.”

Why did the average score go up? The authors emphasize that the public and political emphasis is on higher graduation rates and that “the increase in graduation rates is concentrated in public schools, whose funding is more tied to graduation. However, we find that grades have risen in all types of schools except high schools.
profitable colleges. In addition, per student spending has gone down, indicating that colleges have not increased spending to help students graduate.” This hypothesis is probably part of the truth. I also suspect that since college costs have skyrocketed, higher education makes more commercial sense—that is, you’ve paid your bills and can expect to earn a degree.

As the authors note, a higher percentage of college graduates based on inflated grades is not necessarily a bad thing for society as a whole. After all, a given college graduation standard is not a natural fact, like the boiling point of water, but rather a judgment. Maybe it would be a little easier to graduate from college? But when the standard is relaxed, it also affects how a college student with average or below-average grades will be treated when trying to enter the job market.

For some of my previous posts on college and university grade boosts, see: