How many students change their college major?

About 30% of students who declared a major changed it at least once within three years of enrolling, according to a National Center for Education Statistics Data Point drawing on the 2012/14 Beginning Postsecondary Students Longitudinal Study. The rate is higher for bachelor's students at 33% and lower for associate's students at 28%. Around 1 in 10 changed more than once. Those figures are the useful context for the decision: switching is common enough that it is a normal part of the process rather than a failure, and common enough that keeping your first choice cheap to reverse matters more than getting it right immediately.

Which college majors do students leave most often?

Mathematics, by a wide margin. About 52% of students whose original declared major was mathematics had switched within three years, followed by natural sciences at 40%, education at 37% and humanities at 36%. The lowest switching rates are in applied fields: other applied subjects at 22% and health care fields at 26%. Grouped together, 35% of students who started in a STEM field changed, against 29% of those who started outside STEM. The pattern is worth noticing, because the majors with the highest switching rates are frequently the ones described as the safe, rigorous choice.

When do I have to declare my major?

For most four-year US programmes, the end of sophomore year. Both ACT and the Princeton Review give that as the standard deadline, which means you typically have two years of enrolment before the choice becomes binding, and general education requirements usually fill a good portion of the first of those years. The practical consequence is that entering undecided is a supported option rather than a fallback, and the first year is better spent generating evidence than defending a choice you made at 17.

Does my major decide my career?

For a small number of fields it substantially does, and for most it does not. Some paths run through a licensing or accreditation chain that starts with a specific accredited undergraduate programme, and entering those late is expensive in time and money. Most other fields hire on demonstrated evidence rather than the subject printed on the degree. That split is the single most useful thing to establish about your own shortlist, because only the gated ones justify treating this as a high-stakes decision at 18.

Should I choose a major based on salary?

Salary belongs in the decision, but it is a poor primary filter at 18 because it is the input you can change later most easily and the one you can research at any point. Interest in the day-to-day work is harder to acquire and harder to fake across four years of coursework. A more useful version of the money question is narrower: does this major keep me out of any field I care about, and can I reach the pay I need through it without a second qualification? That is answerable now. Peak lifetime earnings are not.

How to Choose a College Major When a Third of Students Switch

Q: What should I do if I have no idea what to major in?

Treat it as a shortage of evidence about yourself rather than a shortage of options, because that is almost always what it is. Among people who come to our career quiz with no work experience yet, the two most-cited struggles are having no idea which direction to pick and not being able to tell what they are good at. Neither is solved by reading another list of majors. What moves it is taking the introductory course of a subject you suspect, keeping prerequisites open for anything gated, and paying attention to which work you return to when nobody has assigned it.

The honest method is shorter than most articles admit. Work out which of the majors you are considering are gated behind a licensing or accreditation chain and which are not, take the introductory course of the one you suspect before you commit to the sequence, and choose the option that keeps the most doors open per dollar spent. Then stop treating the decision as final, because for most students it is not.

About 30% of students who declared a major changed it at least once within three years of enrolling, according to a National Center for Education Statistics Data Point built on the 2012/14 Beginning Postsecondary Students Longitudinal Study, a nationally representative sample of roughly 25,000 first-time students. For bachelor's students specifically the figure is 33%. About 1 in 10 changed more than once.

Nearly every guide to this question, including the ones currently ranking above this one, tells you to consider your interests, your strengths and your career goals. That advice is not wrong. It is just an instruction to interview yourself using evidence you do not have yet, which is why you can read three versions of it and still be exactly where you started.

A third of students switch, and the switching is concentrated

33%of bachelor's degree students who declared a major had changed it within 3 years of enrolling (NCES Data Point 2018-434, BPS:12/14, about 25,000 students)

Two things follow from that number, and they point in opposite directions.

The first is reassurance. A third of your future classmates will change their mind, so a choice made now with imperfect information is the normal case rather than a mistake you are about to make. Colleges are built around this: general education requirements fill much of the first year, and both ACT and the Princeton Review put the standard declaration deadline at the end of sophomore year.

The second is a warning. Switching is only cheap if you switch early and if the thing you switch into does not have a two-year prerequisite chain you skipped. A student who moves from history to sociology in their third semester loses very little. A student who decides in their third year that they want the nursing route loses considerably more, because the sequence starts earlier and is not built to be joined late.

So the goal is not to avoid switching. It is to keep the cost of switching low while you gather the evidence that would have made the first choice better.

The majors students leave most are not the ones you would guess

The same NCES analysis breaks switching down by the field a student originally declared. The spread is much wider than the headline average suggests.

Original declared major	Changed major within 3 years
Mathematics	52%
Natural sciences	40%
Education	37%
Humanities	36%
Engineering and engineering technology	32%
General studies and other	32%
Social sciences	31%
Business	31%
Computer and information sciences	28%
Health care fields	26%
Other applied fields	22%

Source: NCES Data Point 2018-434, figure 2. STEM fields overall sat at 35% against 29% for non-STEM.

Read the top and the bottom of that column together. The fields with the highest switching rates are abstract and are frequently the ones described as the rigorous, respectable, keeps-your-options-open choice. The fields with the lowest are applied, and a student picking them can usually picture the job at the end of them.

That is the mechanism worth taking from the table. An applied major is chosen against a concrete image of the work. An abstract major is chosen against an image of yourself, and self-image at 18 is the least tested thing you own. Mathematics does not lose half its declared majors because the subject is bad. It loses them because "I was good at maths in school" and "I want to spend three years doing university mathematics" are different statements, and the gap between them only becomes visible in the second semester.

Before you declare, find the syllabus for a third-year course in that major and read it. Introductory courses are designed to be broadly appealing and tell you almost nothing. The upper-level reading list is what the degree actually is, and your reaction to it is the cheapest signal available.

The three inputs everyone recommends are all self-report

Every mainstream guide to this question runs the same three: what you enjoy, what you are good at, and what career you want. Each one asks you to look inward and report back, and each one depends on evidence that eleven years of school were not designed to produce.

Classrooms assess a specific and unusual kind of performance: fixed syllabus, external deadlines, solitary work, someone else deciding what counts. You now know a great deal about how you operate when the task is set for you. You know almost nothing about what you do when it is not, and the second condition is the one that describes most working life.

That gap shows up in what people tell us. Among quiz respondents with no work experience yet, the two most-cited struggles are having no idea which direction to pick and not being able to tell what they are good at. Between them they account for the majority of that group. The third most common is having too many interests to commit to one. None of those is an options problem, and none of them is fixed by another list of majors.

Undecided at 18 describes the size of your sample, not the strength of your character. A first year spent producing new evidence beats a summer spent interrogating the evidence you already have.

Two kinds of major: gated and ungated

This is the distinction that does the most work, and almost nobody writing about major selection makes it.

Some career paths run through a chain that begins with a specific accredited undergraduate programme: clinical health professions, architecture, several engineering licensure tracks, teaching certification in most states, and anything with pre-professional prerequisites attached. Enter these late and you pay in extra semesters, extra tuition, or a second qualification.

Most other paths do not work that way. Employers in the large ungated middle of the economy hire on demonstrated evidence: what you have built, worked on, or can show. The subject on the degree functions as background rather than as a key.

The action here is specific and takes an afternoon. For every field you are curious about, find the relevant licensing or professional body and read their entry requirements directly, not the college's marketing page. You will end up with two lists.

Gated fields you care about. These deserve genuine attention now, because the prerequisite sequence starts early. If any appear on your list, keep their first-year prerequisites open even if you do not declare, since that is usually two or three courses rather than a commitment.
Ungated fields you care about. These do not need to be resolved this year at all. Choose the major you will engage with, and build the evidence separately.

If your entire shortlist is ungated, the decision carries much less weight than the pressure around it suggests, and you can pick on interest with a clear conscience.

What you did when nobody was assigning it

If self-report is the weak input, the fix is to use evidence you produced without meaning to.

Long before anyone was grading or paying you, you decided how to fill hours that nobody was supervising. Nothing was riding on those decisions, which is precisely what makes them readable now. The child who reorganised the shelf, the one who argued with the rules of the game, the one who took apart something that already worked: each was expressing a preference no exam recorded and no questionnaire can recover, because by 18 most people have written over it with whatever they were praised for.

There is a second signal worth trusting, which is which work absorbs you completely once you are in it. It is a poor guide to job titles and a good guide to conditions: whether you want depth or variety, structure or open problems, people in the room or not.

Among people who arrive at our quiz already questioning their direction, the share naming boredom as their core struggle climbs with every level of education, from a small minority of those with high school only to roughly one in five of those holding a master's degree or higher. Full figures and bases are in the Career Change Report 2026.

Boredom rises with every degreeShare naming boredom as their core struggle, by education.

Boredom rises with every degree. % naming boredom as their core struggle.
Answer	Share
High school	6.1%
Bachelor's degree	15.5%
Master's degree or higher	19.5%

Source: MyPassion.ai Career Change Report 2026 (n = 3,460 respondents, updated 2026-08-01)Download image

That group is self-selected, so it describes people who came looking for a change rather than graduates in general. Even so, the direction is worth a 19-year-old's attention: further credentials did not resolve the question for these people, and in several cases they arrived at it later with more sunk cost attached. The degree is not the thing that protects you from ending up bored. Choosing work you would engage with unpaid is closer to it.

This is the variable our career quiz for students is built to surface. It runs 26 branching questions in under 10 minutes, asks directly when you completely lose track of time and what you would do if money were not a constraint, and maps the pattern onto one of 20 archetypes and then onto career directions rather than a single job title. It will not tell you which major to declare, because that depends on your finances and your family. What it gives you is a set of directions with evidence behind them, so the major you choose is pointed somewhere you have a reason to go.

What to do in your first year if you are undecided

Entering undecided is a supported route, not a holding pattern, and it is only wasted if you spend it passively.

Take the gateway course, not the survey course. Every department has one course where students discover whether they want the major. It is rarely the introductory one. Ask an advisor which course in the sequence loses people, and take that.

Protect the gated prerequisites. If anything on your shortlist is licensing-gated, its first-year requirements are usually two or three courses. Taking them keeps the option alive at low cost and they normally count for something else if you drop the plan.

Talk to fourth-years, not admissions. Someone three years into the major will tell you what the middle two years are like, which is the part nobody advertises and the part that decides whether people finish.

Write down what you notice, weekly and briefly. Which classes you looked forward to, which reading you did beyond what was assigned, which assignment you kept tinkering with after it was good enough. By spring you will have a term of behavioural evidence about yourself, which is more than you had when the question first came up.

If two options feel equally good

Rank them on the questions that predict how the next three years go, not on projected salary, which you can research at any point and change later more easily than you can manufacture interest.

Which required course list would you take voluntarily? Which department's upper-level material can you read for twenty minutes without your attention sliding? Which choice keeps the other reachable as a minor or a later specialisation, because one of these pairs is usually asymmetric in a way that decides it for you?

If they survive all three, they are genuinely close, and closeness is useful information: pick either and stop paying the cost of deliberating. The switching data says a third of students revise this decision anyway, and revising a decision you engaged with is far cheaper than never making one.

Picking, without pretending to be certain

What is in front of you is not a career decision. It is a decision about which subject you will spend two years producing evidence inside, taken against a deadline most programmes set at the end of sophomore year.

So establish which of your options are gated and treat only those as urgent. Read an upper-level syllabus before you commit to a sequence. Take the course that separates people rather than the one that recruits them. Then let the first year do the work that no amount of deliberating at 17 can do.

If you want a starting read on which directions fit the patterns you already have, the archetype quiz for teens is free and runs in under ten minutes. If the wider decision is still open, what to do after high school covers the routes either side of a degree, and childhood interests and the career connection explains why unsupervised behaviour reads more honestly than a self-assessment. For instruments built for this age specifically, YouScience measures aptitude rather than asking you to report it, and the free O*NET Interest Profiler maps interests onto occupations. If money is the binding constraint on the whole plan, highest-paying careers without a degree has the routes that skip it, and best career aptitude test compares what the main assessments can and cannot tell you.