Can personality tell you what to study?

What should I study? The form wants a decision by a date, and everyone asking about it wants a single word back. You have some subjects you’re better at, some you actually like, and a nagging sense that those are not the same list. Somewhere in the middle of that, somebody suggested a personality test.

It is a reasonable instinct — and the research does have something useful to say. Just not the thing people usually hope for, which is a result that names a degree.

The short answer

Two different measurements do two different jobs here, and it is worth keeping them apart.

Personality predicts how well you’ll perform. Weakly. Poropat’s meta-analysis of over 70,000 students found conscientiousness correlates with academic performance at ρ = .22 — about 5% of the variance. Openness manages .12. Extraversion is −.01, which is to say nothing at all.

Interests predict whether you’ll stay. Better. Across 92 studies and 1,858 correlations, the fit between a person’s interest profile and their field correlates with performance outcomes at .32 — double the .16 for interest scores on their own. And in a study of nearly 50,000 students across 25 US four-year institutions, interest–major fit and first-year grades independently predicted whether a student stayed in the degree they started.

So: if you want to know whether you can pass, look at conscientiousness and prior grades. If you want to know whether you’ll still be there in year three, look at fit. Neither explains enough variance to hand over the decision.

Does personality actually predict university performance?

One trait does, modestly, and it holds up better than you might expect.

Conscientiousness is the only Big Five trait with a substantial relationship to academic performance, and it is unusual in that it doesn’t fade as you go up the levels: Poropat found ρ = .28 at primary, .21 at secondary and .23 at tertiary. Every other trait’s correlation drops sharply with academic level — the things that help a fourteen-year-old get good marks stop mattering by university. Self-discipline doesn’t.

The most striking number in that paper is about incremental value. Controlling for secondary-school GPA, the partial correlation between conscientiousness and university performance was .17 — while the equivalent figure for intelligence was .14. Once you know someone’s prior grades, how conscientious they are tells you slightly more about their degree performance than how clever they are.

That result gets over-quoted, so here is the restraint: ρ = .22 is around 5% of the variance in academic performance. Openness at .12 is under 1.5%. Emotional stability and extraversion are statistically indistinguishable from zero. Poropat also notes the correlations were corrected for scale reliability but not for range restriction, and that most of the data is self-report personality against self-report or institutional grades.

A measure that explains 5% of the variance in your marks cannot responsibly pick your degree. It can tell you something about how you will need to work.

Do people in different subjects actually have different personalities?

Yes, at the level of group averages — and this is the finding most likely to be misused.

A systematic review by Vedel (2016) pooled 12 studies covering 13,389 students. Eleven of the twelve found significant Big Five differences between academic majors, and openness showed the largest and most consistent differences, regularly reaching large effect sizes. Arts and humanities, psychology and political science students scored highest; health sciences and business/commerce lowest.

Now the part that matters. These are group mean differences with enormous overlap between the distributions. A large average difference in openness between humanities and business students still means a great many low-openness people thrive in humanities, and plenty of high-openness people are perfectly happy in commerce. Knowing the average tells you very little about any individual, including you.

The review also cannot tell you which direction the causation runs. It is cross-sectional: it cannot separate selection (open people choose arts degrees) from socialisation (studying arts raises measured openness). Both are plausible and both are probably happening.

Is “follow your interests” actually supported, or is it just nice advice?

Partly supported, and genuinely contested — which is more useful to know than a clean answer.

The optimistic case is the congruence research above: fit predicts outcomes at .32, better than interest scores alone, and fit predicts persistence in nearly 50,000 students. There is also a striking result in Sackett and colleagues’ 2022 re-analysis of selection research — when they corrected what they argue were systematically overstated range-restriction adjustments across the whole field, interests was one of the very few predictors whose validity went up, from .10 to .24. That now places interests above overall conscientiousness (.19).

The sceptical case is real. Van Iddekinge and colleagues (2011) meta-analysed 74 studies and found single interest scales predicted job performance at only .14, training performance at .26, and actual turnover at −.15. Their estimates are markedly lower than the congruence figures, and the 2017 congruence paper exists partly as a response to them.

The honest synthesis, which both camps would roughly accept: interests predict persistence and training outcomes better than they predict how well you do the work. Fit tells you whether you’ll stick, not whether you’ll be brilliant. For choosing a degree — where dropping out is the expensive failure mode — persistence is arguably the more relevant outcome anyway.

So how should I actually use this?

Use it to narrow and to plan, not to decide.

• Let interests set the shortlist. They are the better predictor of whether you’ll still want this in two years. If a field bores you now, a degree in it is three years of the same feeling.
• Let conscientiousness set your expectations of the work, not the subject. A low self-discipline score is not a reason to avoid a demanding degree. It is a reason to build structure into it from week one, because the trait is stable and the deadlines are real.
• Ignore the group averages. “Psychology students score high on openness” is a fact about a distribution, not a fact about you.
• Check the thing neither test measures. Entry requirements, what the actual day-to-day of the degree involves, whether the assessment style suits you, cost, and what the people one year ahead of you say. None of that is in a personality profile.

And the caveat the whole post rests on: every relationship described here explains a small share of the variance. The research is good enough to inform a decision and nowhere near good enough to make one. Anyone selling you a test that names your degree is overselling what the field can do.

Where the research is genuinely useful

For one specific question: why does this keep going wrong for me? If you have switched courses, or you did well at school and are struggling now, the findings above give you something concrete to look at — whether the problem is fit (you are in the wrong field and interest never arrived) or execution (the field is right and the structure isn’t there). Those look similar from the inside and they need opposite responses.

That is the real use of a profile at this stage. Not choosing for you — giving you a more accurate description of the person doing the choosing.