I have my own thoughts on this, motivated by the work of Hanke Korpershoek, the author of a Dutch study into the relationship between a student’s personality and their subject choice at school. She wrote a really fascinating paper, and showed that introverted, socially awkward students stick with science subjects, while extroverted, highly socialised students drop maths and science subjects at the first opportunity. (@JoMarchant did a piece for New Scientist on it here).
The study related only to Dutch schoolchildren, but the astonishing thing is that this hasn’t really been studied elsewhere: we don’t know if it’s a global phenomenon. I think someone should rectify this asap. There is good reason to believe that significant, revolutionary scientific breakthroughs are generally achieved by personality types that could be classed as adventurers and risk-takers. So having most of them discard science at the first opportunity is the brain drain we should really be worried about.
What motivates the dropout? That’s not clear, either.
It’s well-established that extraversion is related to social and enterprising interests. Korpershoek’s studies have found that many non-STEM students in higher education had negative preconceptions about STEM subjects – that they wouldn’t fit their self-perception, that these subjects are too difficult, or that they would make them too narrowly-focussed.
But, she points out, these studies show correlation, not causation. So it’s difficult to say whether introverted students tend to explicitly choose STEM subjects over, say, politics (though it wouldn’t be a big surprise), while extroverts do the opposite.
My own view is that the generally-accepted but thoroughly misleading Brand Science depiction of scientists as earnest, cold, logical, ultra-rational thinkers doesn’t do a lot to entice the more ebullient students into the subject.
For my money, any scientists moaning about students giving up on their subject ought to do an audit of how much time they spend publicly talking about how careful and rational scientists are and how people should trust them more, then compare it to the time they spend communicating their passion for their research or displaying an enthusiasm for the beauty and awe of nature and the cosmos.
In public at least, I think Sagan and Feynman (and these days Attenborough and Cox) had (or have) the balance right, and very few who have come into close contact with them and their work can bear to walk away from science.