A failure to "know thyself" is hurting UK universities, which must emulate their US counterparts and embrace institutional research, a conference heard this week.
Bernard Longden, professor of higher education policy at Liverpool Hope University, said universities were shooting themselves in the foot by ignoring "known knowns" about themselves.
Speaking to Times Higher Education before the conference "Exploring the Hinterlands: Mapping an Agenda for Institutional Research" at Southampton Solent University, he said: "Academics like evidence, but when you get to (university) management perhaps things are different. I get the feeling that some people want to hear only the evidence that supports the view they hold, and certainly don't want evidence to the contrary."
Institutional research, which is well established in the US, aims to improve strategic decision-making in areas such as student recruitment and retention, but it also includes pedagogic and marketing research.
Professor Longden said: "During the Iraq War, Donald Rumsfeld, the US Secretary of Defence, spoke of 'known knowns, known unknowns and unknown unknowns'.
"That creased me up, but in institutional research we are surrounded by 'known knowns' - there are lots of data, but often we don't know what to do with them."
Some vice-chancellors and managers are forward-thinking, Professor Longden said, but the sector as a whole has to improve.
"I've been working for ten years now on student retention and non-completion, but we're not making big progress nationally. Institutions are being told they ought to be doing certain things to improve the quality of the student experience, it's a known known, but they're not doing much with it."
He added: "Institutional research is sort of seen as quasi-scientific social science, and the attitude is 'we don't have a lot of faith in that'."
In the US, where it has been a fixture for 30 years, things are different, Professor Longden said.
"If you go to any US institution, it will have an office - a significant office - of institutional research. That's the place you go to find out how many students you've got, the retention rate, the scholarship level, how much money you pay your staff and so on. It is the centre around which all of the data fit, but we do not have anything like that.
"If Ford Motors had cars coming off the production line that weren't working, it would say, 'We'd better understand what's going on in our factories.' If you don't 'know thyself', how can you possibly progress? The Americans are good at it not only in industry but in higher education, too. We need to get better at it."