Careers & University / Big Questions / Getting in
From the research interviews we hold · updated 09 Nov 2026
School grades do predict university success, but the strength of that link depends heavily on which school you went to and what background you come from. The research here mostly agrees that grades matter, while showing that the same grades can mean different things for different students.
Word for word, from interviews with the researchers. Each opens the moment they say it.
The relationship between high school GPA and college graduation is consistent across high schools, with students with higher grades being more likely to graduate college than students with lower grades at the same high school.Elaine Allensworth, University of Chicago — High School GPAs and ACT Scores as Predictors of College Completion · 5:30
Students with the same high school GPA or the same ACT score are more likely to graduate college if they attended some high schools than others.Elaine Allensworth, University of Chicago — High School GPAs and ACT Scores as Predictors of College Completion · 4:41
Children from poorer areas who go to university perform better at university than people with the same prior attainment from more advantaged areas.David Willetts, Resolution Foundation — The British education system is tilted against social mobility · 2:21
Poorer students are more likely to drop out of university, both because the dropout rate is higher at lower tariff institutions and because poorer students are more likely to drop out than rich students within the same institution, with pressures and lack of academic preparation being possible reasons.Anna Vignoles, University of Cambridge — Widening Participation in Higher Education: A Quantitative Analysis · 4:56
Each additional tutorial a student attended raised their grade by about six percentage points.Carlos Cortinhas, University of Exeter — Nudging Students Back to Class: Behavioural Insights on Attendance and Performance in Higher Education · 10:12Paper
The clearest support for grades as a predictor comes from Elaine Allensworth's work [2]. Across many high schools, students with higher GPAs were consistently more likely to finish college than students with lower GPAs from the same school. That pattern held whether the school was strong or weak, which suggests grades carry real information about a student's habits and readiness. But her second finding [3] complicates the picture: two students with identical GPAs or identical ACT scores had different odds of graduating depending on which high school they attended. A B average at one school may represent more, or less, than a B average at another. Grades predict, but they are not a clean, universal signal.
Other findings push back on the idea that grades alone tell the story. David Willetts [22] found that students from poorer areas actually did better at university than students with the same prior attainment from wealthier areas. In other words, equal grades did not mean equal outcomes, and the disadvantaged students outperformed what their grades predicted. Anna Vignoles [33] adds a caution: poorer students are more likely to drop out, partly because they attend lower-tariff institutions and partly because of pressures and weaker academic preparation within the same institution. So the same grade can sit on top of very different levels of support.
Some studies suggest that what you do at university matters alongside what you arrived with. Carlos Cortinhas [35] found that each additional tutorial a student attended raised their grade by about six percentage points, meaning day-to-day engagement shifted outcomes beyond entry grades. And Chathurika Kannangara [31] argues that success and retention come from a combination of factors that can be spotted early, not from a single number.
Why do researchers disagree? They study different countries, different groups, and different measures. Some look at GPA, others at test scores, dropout, or degree class. Some compare students within one school, others across regions or income levels. A finding that grades predict graduation [2] and a finding that poorer students beat their predicted grades [22] can both be true, because they ask slightly different questions. The honest answer is that grades are a useful but imperfect signal, and their meaning depends on context.