The story behind our figures...

Ben Marks of YouthSight, which carried out the survey for THE, explains the reasoning and methodology that created this year’s university rankings

April 25, 2013

More than 12,000 full-time undergraduates took part in this year’s survey. All respondents were members of YouthSight’s student panel - who are recruited via the Universities and Colleges Admissions Service - and their views were gathered between October 2011 and July 2012.

The student experience survey is broken down into 21 attributes of universities, chosen by students as key indicators. Participants were asked to rate how their university performed in each of the areas using a seven-point scale. Each attribute was assigned a weight reflecting its importance within the overall student experience.

The same wording and weighting methodology have been used for the past four years, with the greatest weight applied to the attributes that correlated most closely to whether or not the respondent would recommend the university to a friend. Only universities achieving 50 or more ratings have been included in the final data set, and each university’s score was indexed on a scale from one to 100. A total of 102 institutions met the minimum sample threshold based on respondents from a total of 11,459 respondents.

The difference in scores of similarly ranked institutions will not be statistically significant. When results are based on a sample of 100, we have to accept some imprecision to arise from sampling variability. But that does not mean to say that these results are without meaning. In this context, the relatively high level of consistency in our data from year to year is immensely reassuring.

Although the 12,012 respondents in this year’s survey make up less than 1 per cent of the UK’s full-time undergraduate population, this sampling fraction is relatively high in comparison with a typical political opinion poll or large-scale government survey.

More importantly, the overall sample size is large enough to generate only a small sampling error. A sampling fraction of one in 100 is actually fairly high compared with other surveys. The average UK opinion poll, for example, with a sample size of around 1,500, has a sampling fraction of about one in 30,000.

YouthSight believe this makes its survey effective at highlighting where best practice occurs and where certain universities have room for improvement.

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

Operations Officer

Kings College London

Network Engineer

United Arab Emirates University

Communications Studio Technician

St Marys University, Twickenham