Elusive evidence 2

October 5, 2007

In the "for and against" rankings a few weeks ago ("Debate reignited by new league tables", August 17), you omitted one of the most telling factors against, which is again relevant in the students satisfaction ratings ("Students: yes, we're satisfied", September 14).

Academics are teaching students the crucial need for a careful and correct statistical analysis of results to determine whether two groups of results are significantly different. Yet here we have another example of major conclusions being drawn from a list of averages with no statistical comment. No fewer than 49 institutions fall between 3.90 and 3.80. One might be forgiven for thinking that Staffordshire (34) is a much better place than Strathclyde (50), whereas their scores are 3.88 and 3.86 respectively.

I strongly suspect that, with the possible exception of the top few, there is no statistically significant difference between any of the institutions in the entire list.

Trevor Stone
Glasgow University

Register to continue

Why register?

  • Registration is free and only takes a moment
  • Once registered, you can read 3 articles a month
  • Sign up for our newsletter
Please Login or Register to read this article.


Featured jobs