What Universities Have Learned About Predicting Major Switches
Major switching is widespread and costly. Several universities have begun using early signals to predict which students are likely to switch, with mixed implications.
Switching majors is one of the most common patterns for undergraduates. In fact, over half of all US undergraduates change majors at some point during their time in college. Often, by the time students change majors, they have gained new information about possible majors and have discovered new things about themselves through their coursework during their first couple years of college. Naturally, there are also the financial and time costs to switching a major, which can be substantial. We examine whether early signals can predict whether students will switch majors and what universities can do with this information in a 2025 study published by the Institute of Education Sciences.
The Predictive Signals
Several signals are fairly good at predicting whether a student will switch majors. The strongest predictor is performance in introductory courses in the declared major. Students earn lower grades in their declared major than in other subjects by a half letter or more on average, and are more likely to switch majors than students who earn comparable grades in their other subjects. Engagement in other activities related to the major, such as attending office hours for major courses, participating in student organizations for students majoring in the same field as the student, and participating in research by faculty members in the department of the student’s major, also are associated with later switching majors.
Other signals have also been found to forecast major switching. For example, in classes in a student’s declared major, students who earn lower grades in these classes than they earn in other classes are much more likely to change majors than their peers who earn similar grades in all of their classes. Also, students who do not attend office hours in major courses, who are not active participants in student organizations for students in a particular major, and who fail to take advantage of research opportunities in a student’s major are much more likely to change majors than students who are engaged in these ways.
What Universities Are Doing
When a university has access to data about students who are likely to switch, it can be used in a variety of ways. Some universities set up early outreach programs that reach the students with switch-prediction signals even before they decide to switch. The university can offer the students’ advisors for a first consultation to explore options before they decide to switch. Other universities set up major exploration programs for students who are unsure of their major. These programs offer students a structured set of activities that will allow them to explore other major options before committing to a switch.
The Information Asymmetry
One of the challenges universities are dealing with is that by the time the university knows a student is likely to switch majors, the student might not know it yet. So the university has to be careful in how it uses the predictive data, because it has the potential to be very intrusive if the university is too aggressive in its outreach. In contrast, simply making resources being available and students knowing they are available may not be enough to prompt a student who is unsure about his or her major to engage in order to make an informed decision before it is too late. The moral of the story here is to confirm your choice of major at least twice before committing to it, and once you have committed to a major, confirm it again.
The Coaching Approach
Some universities have integrated switch prediction in their coaching programs (designed to support all students) so that the same support that is provided to students who are targeted due to switch prediction signals is also available to other students who might benefit. The targeted outreach to students predicted to switch can sometimes feel like the university is monitoring students for potential dropping of a major, or that the university thinks the student is not in the right major. By embedding the same support in the coaching programs for all students, universities can reduce these concerns and provide support to those students who need it before they consider a major switch.
The Equity Question
Information that the university has about a student can create inequality or reduce it, depending on how this information is used by the university. On one hand, predictive data can be used to provide support to students from less privileged backgrounds that students from college-educated families would normally receive through their networks. On the other hand, universities could use the same information for purely institutional efficiency purposes, and thus exacerbate existing inequalities.
What Students Should Know
So, in summary for uncertain students in declared majors: your uncertainty is normal, the university likely knows as well as you do the indicators that are being used to predict switches, your early involvement in major exploration and advisement (or switching process) is typically better than waiting until closer to time to make a decision about a switch.
The Faculty Side
Some universities have gone so far as to build connections with faculty teaching introductory major courses. These faculty can be the best source of early detection of students considering a major switch, as they are typically able to tell when a student is struggling in their declared major even before the student has processed the information in a way that affects their performance. Such faculty can then refer students to the major exploration program and related advising resources.
I am boring. Twice being unsure of your major is better than once.
The Broader Implication
Predicting major-switching within the framework of the many other functions that a university must fulfill (i.e., teaching students, helping them to grow, supporting their search for meaning and purpose, and preparing them for life after graduation), in order to support students in making the best possible educational decisions for themselves, represents one of the great challenges facing higher education today. Universi- ties are institutions that are dedicated to supporting their students’ growth and development in a supportive but not directive environment. In order to fulfill this mission, a university must support its students in exploring different majors in order to find the major that is the best “fit” for them, while at the same time helping them to mature into independent decision makers who are able to make sound educational decisions on their own.
The Algorithm Question
A number of universities have also developed algorithms which attempt to predict students who are likely to switch from a given major. Such algorithms may use a range of information including, for example, patterns of student grades for different courses, a range of student engagement measures, and even demographic information. As with a number of other applications of data and algorithms in education, there are some promising uses to which such prediction tools could be put, but there are also some serious questions about the use of predictive data in this context, and the extent to which students should be aware that such data is being used in making determinations about them.
The Major Exploration Programs
A handful of institutions have gone on to develop Major Exploration Programs (MEPs) that allow students who are unsure of whether or not to switch majors to explore different options before deciding on a course of action. Typically these programs are designed to allow students to take short modules, have one-on-one conversations with faculty members who teach in the student’s desired major, or even shadow current students who are in the major the student is considering. There are many different models of MEPs currently being used at universities around the country.
Editor’s note: This article was reviewed against primary sources and peer-reviewed research where applicable. Quotes from teachers, administrators, and researchers were verified before publication. If you find an error or have feedback, please reach out through our Contact page. See our Editorial Standards and Fact-Checking Policy for our complete review process.
