Law students whose degrees are broken down into units and taught in semesters get lower grades than students studying under the old three-term academic cycle.
Southampton Institute's law faculty has compared the performance of students operating in semesterised and non-semesterised teaching environments and found a steady decline in pass rates in the former.
David Bailey, head of academic operations, believes concerns about the impact of semesterisation on weaker students are widespread and growing and the old three-term year will return to dominate university teaching. Semesterisation has gone hand in hand with modularisation - where courses are broken into examinable units to fit in with the semester timetable.
"Weaker students just don't have time now to absorb information and our preliminary findings suggest their end degree results are better in non-semesterised environments," said Mr Bailey.
Susan Barber, senior lecturer in law, said the research started out as an attempt to find out what made students fail. The results, which are likely to be of interest to a significant number of institutions, she said, are described as "very serious" for law.
"What we don't know is whether comparable findings would emerge from different disciplines," said Dr Barber. "There just isn't enough research into the strengths and weaknesses of semesterisation, although many places are now reviewing their provision as a result of the current debate."
Her research compared the exam results of different cohorts of students. Cohort A was not semesterised until it reached year three, cohort B was not semesterised until year two and cohort C was semesterised throughout the three years, as were cohorts D and E. Year one performances of the B cohort (1993 intake non-semesterised) was therefore compared with cohorts C, D and E (1994/95/96 intakes semesterised).
Dr Barber said the results indicated that following semesterisation there was a steady decline in the pass rate at year one with the C, D and E cohorts unable to match the performance of cohort B. The same pattern was repeated a number of times.
The difference in the pass rates of the two cohorts of students studying under semesterised and non-semesterised systems was more than 17 per cent according to the research.
But when the two cohorts progress to year 3 and are both taught and assessed in the same semesterised system the difference in the pass rate drops to 10.6 per cent and after resits it is down to 5.13 per cent.
"There seems to be a common thread of failure and we are confident of a causal link, although other factors may also be playing a part," Dr Barber said.
Alan Smithers, professor of education at Liverpool University, said universities had got into a muddle over semesterisation and there was now a clear case for stepping back and analysing the impact of such a major shift.
"The rationale for semesterisation came about when universities envisaged a significant expansion in student numbers - something like 50 per cent - and realised they would need to teach all year to accommodate them," he said. "But of course the Treasury became alarmed about the cost implications of this and the numbers were capped."
What is left behind, he said, is a rather untidy system - some have moved over fully, others partially and some not at all - so institutions are out of step with each other and the old rhythm of the university year is disrupted.
"The old rhythm may have been archaic since it was organised around religious festivals, but it did allow students to pace themselves for one set of exams at the end of the year," Professor Smithers said.
"Now they have to break for Christmas and keep the revision going and then pick up the pace again after the holiday."
Changes to the nature of assessment brought about by semesterisation could also play a part in changing student performance, Professor Smithers said.