A chart of unemployment by major can make a four-year decision feel urgent. The enrollment figures underneath that anxiety are easier to check, and they say something different from “leave.” In spring 2026, undergraduate computer science enrollment fell while undergraduate enrollment overall rose. That is a reason to read your own coursework with a cooler head. It is a weak reason to file a change-of-major form on a Tuesday night.
The National Student Clearinghouse described the spring 2026 term on 11 August 2026. At four-year schools, undergraduate computer science enrollment was down 8.4 percent. At two-year colleges it was down 11.2 percent. Overall undergraduate enrollment rose 1.3 percent. Second-fall persistence was 85.0 percent in computer science and 93.1 percent in engineering. Persistence here means students who started in the field and were still enrolled the following fall. Computer science kept fewer of its starters than engineering did. Both shares are still large majorities.
Those are headcounts and persistence rates. They are not the unemployment rate for computer science graduates, and they are not a forecast of your offer letter. A widely shared post can list a major-level unemployment rate in the same week. Unless you have opened the table and can say the age group and the year, leave that rate out of the decision. This post stays with the Clearinghouse counts.
What a decline in enrollment does not decide
An 8.4 percent drop means fewer undergraduates were enrolled in computer science than a year earlier. It does not mean the students who stayed are doomed, and it does not mean the students who left were wise. People leave a major because of grades, money, interest, and the mood of the moment. The Clearinghouse number adds those motives together. You cannot recover your own motive from it.
The persistence gap is more useful for a course plan. Engineering kept 93.1 percent of students into a second fall. Computer science kept 85.0 percent. Eight points is a real difference in who is still in the seat. It is also compatible with a great many computer science students continuing. If your reason for switching is “everyone is getting out,” the persistence figure says most people in the major were still there.
@aakashgupta recalled, on 6 October 2026, that new computer science majors fell by about half after the dot-com crash, and that those graduates later met a different technology market. Treat the “half” as his account of that earlier period. The Clearinghouse did not publish that figure, and the post is not a prediction. The useful part of the anecdote is the lag. The students who enrolled into a slump did not graduate into the same slump. A major chosen to dodge this year’s headline can graduate into a different one. Do not ignore a major you hate because of that lag. Separate “I do not want this work” from “a chart scared me.”
Questions that belong in the decision
Sit with four questions that a national headcount cannot answer.
Can you do the next required course, based on the last one you finished? A switch that dodges a course you are ready to pass is a different act from a switch that gets you out of a sequence you are failing for the third time.
Do you want the work the major’s jobs actually contain? Read two postings. If the tasks repel you, the enrollment chart is irrelevant and your distaste is the evidence.
What would you major in instead, and what does that field ask in the first job? A move from one crowded major to another crowded major, done in a weekend, repeats the headline problem with new nouns.
What artifact will you have either way? A course that produces a program, a dataset, a design, or a client memo survives a major change. A course that produces only a grade does not travel as well. The Clearinghouse cannot see your artifact. You can.
If the answers point away from the major, switch with a written plan for the new one. If they point toward staying, stay, and spend the hour you would have spent on forums picking the next course that produces proof. The 8.4 percent decline will still be true. It will not have picked your Tuesday.
Talk to the advisor who can see your transcript before you talk to a stranger’s chart. An enrollment drop is a fact about institutions. Your remaining requirements are a fact about you. Both can be true in the same week. Only one of them signs the form.
How to read the next headline
When a post lists majors beside unemployment rates, ask for the table, the ages, and the year. A rate in the high single digits still means most people in that row are working. A rate that moved a lot from a previous year is a change, and the base still matters. Do not let a list of majors become a ranking of your worth. The Clearinghouse post is about who enrolled. Labor-market tables are about who is employed. Mixing them in one sentence is how a course plan gets thrown out for a reason neither source stated.
Write down the figure you actually checked. “Clearinghouse, spring 2026, four-year computer science enrollment down 8.4 percent, overall undergraduate enrollment up 1.3 percent.” If a newer release replaces it, replace the sentence. Do not let an old screenshot keep steering the major.
How Flightpath helps
Flightpath does not tell you to leave a major. The portal plan by year starts from the year, major, and goals you give it. When you pin a role, the gap list shows skills the posting asks for against facts you have confirmed. A course you choose can close one of those gaps if it produces work you can confirm. The weekly check-in is a small loop for that choice. It is not a change-of-major button.
You still register for the class yourself. The product can hold the gap and the posting. It cannot see whether you like the work. Eligibility labels stay on jobs, including work authorization, and a role you cannot take is never ranked above one you can take. The enrollment percentages stay on the Clearinghouse page. The plan stays on the courses you are willing to finish.