Careers & University / Big Questions / Choosing a subject
From the research interviews we hold · updated 09 Nov 2026
The research here mostly agrees that your school subjects don't lock you out of a future: early specialisation can narrow movement between disciplines, but students who seem unlikely to succeed often gain the most from taking a chance. The bigger limits usually come from information, confidence and family circumstances rather than the subject list itself.
Word for word, from interviews with the researchers. Each opens the moment they say it.
Early specialisation, while important for depth, has often prevented students from moving across from their home discipline into other disciplines and strengthening their knowledge of the home discipline by encountering the other discipline.Dilly Fung, University College London — A Connected Curriculum for Higher Education · 8:42
Students with a low likelihood of attending selective colleges experience the largest returns to attendance, even larger than those for more advantaged students with a higher likelihood.Jennie E. Brand, University of California Los Angeles — Overcoming the Odds · 0:34Paper
Students unlikely to complete college can still receive substantial rewards, a phenomenon termed 'negative selection'.Jennie E. Brand, University of California Los Angeles — Overcoming the Odds · 1:01Paper
Peer effects or sibling effects influence college choice decisions, partly through the transmission of otherwise hard-to-perceive and hard-to-obtain information about the college and college major experience.Joshua Goodman, Boston University — O brother, where start thou? · 5:36
Decisions about higher education involve not only the individuals but also their families and communities.Sue Kilpatrick, University of Tasmania — Informing Aspirations in Rural Regions · 10:25Paper
Categorizing students by learning styles or intelligence risks fostering a fixed view of their abilities, which constrains future learning and willingness to take risks, whereas viewing learning as resulting from effort makes students more open to future learning.Professor Stephen Dinham, University of Melbourne — Applying educational research to practice · 0:00
The question is really two questions: do your subjects close doors, and do they decide how well you do? The evidence here splits along those lines.
On the 'yes, subjects limit you' side, [11] is the clearest. Dilly Fung argues that specialising early, while it builds depth, has often stopped students moving from their home discipline into another one, and stopped them strengthening their own subject by meeting a different one. That is a real cost of choosing a narrow A-level or equivalent mix: the longer you stay in one lane, the harder a sideways move becomes.
But several findings push back hard. [13] and [15] both come from Jennie Brand's work on selective colleges. Students who look least likely to attend or finish a selective college get the largest returns from going, and even students unlikely to complete college can gain a lot. In plain terms, the people whose backgrounds or grades suggest 'this isn't for me' often benefit most from trying. That cuts against the idea that your school record or subject choices should stop you applying. [27] adds a related point: treating ability as fixed, including sorting students by 'learning styles', makes people less willing to take risks and keep learning, whereas seeing learning as effort keeps doors open.
[18] is mixed. Peer and sibling effects shape college and major choices, largely by passing on information that is hard to get otherwise. That can widen or narrow your options depending on who is around you, so it is not the subjects themselves doing the limiting.
[21] is also mixed. Decisions about higher education involve families and communities, not just individuals. So the practical limits on a 17-year-old are often money, transport, family expectations and who you know, not the timetable you picked at 14.
Why do researchers disagree? They study different things. Fung looks at university curriculum structures across disciplines; Brand measures earnings and completion for particular college-goers in the US; Kilpatrick looks at families and communities, largely in Australia. Different countries, different outcome measures, different age groups. None of them is measuring 'A-level choices' directly, so treat the findings as strong hints rather than a verdict on your exact situation. The honest summary: subjects matter somewhat, especially for highly specialised degrees, but confidence, information and circumstance matter more than most people assume.