Skip to content
Best College Rankings

Choosing

Is college worth it? What the outcome data shows

The short answer

For most students a bachelor's degree still produces a substantial lifetime earnings premium, but the variance by major, institution and — above all — completion is enormous. The students who lose money on college are overwhelmingly those who borrow and do not finish. Choosing a school you will graduate from matters more than choosing the highest-ranked school that admits you.

Three things to remember

  1. 1Completion is the dominant variable. Debt without a degree is the worst outcome by a wide margin.
  2. 2Major explains more earnings variation than institution does, for most students.
  3. 3Selectivity produces a real earnings gain for low-income and first-generation students, and little gain for everyone else.

The premium is real but unevenly distributed

Bachelor's degree holders continue to earn substantially more over a lifetime than high school graduates, and the gap has not closed. But averages conceal the spread. Median earnings ten years after entry range from around $48,000 at some institutions in this dataset to around $124,000 at others, and within a single university the gap between an engineering graduate and a humanities graduate can be larger still.

The most important finding in the research is about completion, not selectivity. Students who borrow for college and leave without a degree carry the debt without the earnings premium, and they are the group for whom college genuinely does not pay. That risk is highest at institutions with low graduation rates — which is why a 68% six-year graduation rate should weigh more heavily in your decision than a twenty-place ranking difference.

Major matters more than institution, with one exception

Studies comparing students admitted to selective colleges against similar students who attended less selective ones generally find small earnings differences. What does drive earnings is field of study. Colorado School of Mines produces median earnings around $110,000 not because it is a more prestigious institution than most of the Ivy League, but because nearly all of its graduates are engineers.

The exception is important. For low-income and first-generation students, attending a more selective institution does produce a measurable earnings gain. The likely mechanisms are aid generosity, higher graduation rates and access to professional networks that are not otherwise available. If you are a first-generation student admitted to a selective school that meets full need, the case for attending is stronger for you than it would be for a wealthier classmate.

How to decide in practice

A reasonable borrowing heuristic: total student debt at graduation should stay below your expected first-year salary. If you are heading into a field with $55,000 starting salaries, $55,000 of debt is manageable and $110,000 is not. Median federal debt at graduation is published for every institution, and it ranges widely — about $6,000 at Berea, about $27,000 at Drexel and Spelman.

Then weigh three things in this order: will you graduate, what will you study, and what will it cost. Only after those should institutional prestige enter the calculation. This ordering is the opposite of how most families approach the decision, and it is the single highest-value correction available.

The honest alternatives deserve mention too. Skilled trades, two-year technical degrees, apprenticeships and community college transfer pathways all produce good outcomes for many students, and a community college transfer to a state flagship yields the same degree at a fraction of the cost. The question is not whether college is worth it in general, but whether this college, for this student, at this price, is.

Frequently asked questions

Usually, if you finish. A reasonable rule is to keep total debt at graduation below your expected first-year salary. The group for whom college genuinely does not pay is students who borrow and do not complete a degree — they carry the debt without the earnings premium. This is why a school's graduation rate should weigh more heavily in your decision than its ranking.

For earnings, generally yes. Research comparing students admitted to selective colleges with similar students who attended less selective ones finds small differences for most students, while the gap between engineering and some humanities majors is large at every institution. The exception is low-income and first-generation students, for whom attending a more selective school does produce a measurable gain.

Eliminating expensive-looking schools based on sticker price without running the net price calculator. Stanford's published cost is about $87,000 but its average net price is roughly $17,500, and for families under $100,000 of income it charges no tuition, room or board. Families routinely never apply to schools that would have been their cheapest option.
8 min readUpdated September 1, 2026

Read next

  • Sticker price vs. net price: the only cost number that matters

    Sticker price is the published cost of attendance; net price is what a family in your income band actually pays after grants and scholarships. The gap is often enormous and varies wildly by school — Stanford's sticker price is about $87,000 but its average net price is roughly $17,500, less than the out-of-state cost at many public universities. Never eliminate a college from your list based on sticker price.

  • How college rankings actually work

    A college ranking is a weighted average of measurable inputs chosen by an editor. Change the weights and the order changes — which is why the same school can rank #76 on one list and #36 on another. No ranking measures whether a college is good for you; they measure what their authors decided to count.

  • How to compare financial aid offers side by side

    Aid letters are deliberately hard to compare because every college formats them differently and most bundle loans in with grants. Reduce each offer to one number: total cost of attendance minus grants and scholarships only. Multiply by four — or five for co-op programs — and compare those totals. Then check whether merit scholarships renew.