What the data actually shows
The clearest finding comes from Roese and Summerville's 2005 review of what people regret most. Pooling regrets across many studies, they found that education was the single most commonly reported domain of regret in American life — ahead of career, romance, and finance. Their explanation is the key to the whole topic: people regret most in the areas where they perceive the most opportunity for action. Education feels like a domain wide open to second chances (you can always go back to school, change fields, retrain), so it generates the most lingering 'what if.'
Self-reported graduate surveys point the same way, though the exact figures vary a lot by survey, wording, and year, so they are best treated as approximate. Across various polls of college graduates, it is common to see somewhere around a third saying they would choose a different major if they could start over, with regret concentrated in fields whose graduates feel the degree did not pay off or did not match the work they ended up doing. These are rough, self-reported numbers — useful for showing that degree regret is widespread, not for any precise share.
It is worth separating the kinds of regret. Some people regret the field or major (wishing they had studied something more practical or more interesting); some regret the institution or the debt rather than the education itself; and a smaller group regret going at all. These are different questions with different answers, and a survey headline that 'most graduates regret their degree' usually collapses them into one.
Why this feels different from how it actually is
Education feels uniquely regrettable because counterfactual thinking has so much to grab onto. You can vividly imagine the alternative life — the other major, the other school, the career you might have had — and the imagined version is unconstrained by the messy realities that the actual path involved. Comparing a real, complicated outcome to a clean, idealised counterfactual reliably makes the road taken look worse.
The irreversibility deepens it. Big educational choices are made young, with limited information, and they shape years of life afterwards, so any dissatisfaction gets attributed back to that one fork in the road. We tend to forget how much was genuinely unknowable at 18 or 22, and judge the past decision with information we only have now — a classic hindsight effect.
And the value of education is often invisible to the person holding it. A degree can shape how you think, who you met, the doors that quietly opened, and the options you never had to take because something steadier came along — none of which show up as an obvious 'return.' Because the costs (time, money, debt) are concrete and the benefits are diffuse, the ledger can feel negative even when it is not.
What the research says to do about it
The most useful move the regret research suggests is to reframe the comparison honestly. Regret is generated by an idealised counterfactual, so deliberately filling in the alternative path with its real likely difficulties — not just its upsides — tends to shrink the felt gap. The other degree, the other field, would have had its own frustrations, dead ends, and trade-offs that the daydream leaves out.
Where the dissatisfaction is genuinely about the present rather than the past, the research on starting over and mid-life change is encouraging: skills, fields, and direction are far more changeable in adulthood than the 'I picked wrong at 18' framing assumes. Acting on the part you can still influence — building toward the work you actually want now — tends to do more for wellbeing than relitigating the original choice.
Roese and Summerville's own implication is worth holding onto: the domains we regret most are the ones we believe are still open. That belief is a resource. If education feels like a regret precisely because it feels fixable, that is a reason to treat the regret as information about what you want next, not as a verdict on a closed past.
What the research says does not help
Ruminating on the counterfactual degree rarely resolves anything, because the imagined alternative is unfalsifiable and always flattering. Replaying 'if only I had studied X' tends to entrench the regret rather than settle it, since there is no real outcome to test the fantasy against.
Treating a single survey statistic as proof you made a mistake does not help and is not warranted. Self-reported regret figures are noisy, sensitive to how the question is asked, and inflated by hindsight; the fact that 'a third of graduates' would choose differently tells you the feeling is common, not that the original choice was wrong for you.
Assuming the only fix is a dramatic do-over — quitting to retrain from scratch, taking on new debt to 'correct' the first degree — is also not well supported. Sometimes that is right, but the research on change in adulthood suggests smaller, additive moves (new skills, a pivot within a field, a change of role) more often improve things than a wholesale restart.
Real numbers in context
In Roese and Summerville's 2005 analysis, education was the most frequently reported regret domain in American life — the top of the list, ahead of career, romance, parenting, and finance. The mechanism they identified was opportunity: people regret most where they see the most room to have acted differently, and education feels wide open to second chances.
Graduate survey data on degree or major regret is best read as approximate and self-reported. It is common to see roughly a third of college graduates saying they would choose a different major in hindsight, with the figure swinging widely by field, survey, and year. Read these as evidence that degree regret is widespread and normal — not as precise measurements, and not as proof that the original choice was a mistake.