Data not up to the rigours of ranking

October 20, 2000

High dropout rates should not be used to judge universities and colleges because the data on which this performance indicator is based are insufficient.

This was the view presented by economist Robin Naylor of the University of Warwick to a meeting at the Royal Statistical Society this week.

"There is a lot of statistical uncertainty in the figures. If institutions are to be ranked, one needs to be confident of ranking in a statistically significant way. We can be reasonably confident that the top few are different from the bottom few, but those in the middle are not statistically different," said Dr Naylor.

He and his colleagues have analysed the cohort of 100,000 students who graduated from the old universities in 1993 and demonstrated that men from regions with high unemployment, for example, are more likely to drop out of university.

The meeting was also addressed by David Draper, professor of mathematical sciences at the University of Bath.

At present, the funding council publishes data on dropouts for each institution alongside a benchmark figure showing the percentage of students who could be expected to dropout given the institution's entry requirements and subject mix.

It also breaks them down into young and mature students and looks at whether students come from an under-represented neighbourhood and whether they have any previous higher education qualifications.

In general, institutions that have high entry requirements and offer subjects such as medicine and law tend to have lower dropout rates.

Professor Draper told the meeting that other factors also influence the dropout rate, albeit less significantly.

Men are more likely to drop out than women, according to Professor Draper's analysis of 1996-97 data from the Higher Education Statistics Agency. The male dropout rate was 11.5 per cent and the female dropout rate was 8.3 per cent.

Students who had previously attended a state school were more likely to discontinue their studies than those from an independent school.

Social class was also relevant - those from skilled manual, semi-skilled and unskilled backgrounds were more prone to drop out.

Figures published by the funding councils two weeks ago showed that on average 9 per cent of all entrants fail to continue with their studies at the end of the first year.

Please login or register to read this article.

Register to continue

Get a month's unlimited access to THE content online. Just register and complete your career summary.

Registration is free and only takes a moment. Once registered you can read a total of 3 articles each month, plus:

  • Sign up for the editor's highlights
  • Receive World University Rankings news first
  • Get job alerts, shortlist jobs and save job searches
  • Participate in reader discussions and post comments

Have your say

Log in or register to post comments


Featured jobs

Office Manager, UCD School of Computer Science

University College Dublin (ucd)

Research Associate

Kings College London

StoryFutures Childrens Immersive Audience Researcher

Royal Holloway, University Of London