University Executive Board today agreed to a proposal that has come from the Admissions and Outreach Sub-Committee, that we should review the university's position with regard to the use of contextual data in admissions.  This is a controversial issue in which whatever universities do they can't win.  Either they are damned by government ministers and others for not admitting 'enough' students from non-traditional backgrounds (because 'normal' admissions criteria are adhered to with no recognition of the backgrounds of applicants).  Or they are damned by the Daily Mail contingent for apparently turning down middle class (and middle England) applicants in support of a social engineering model to allow applicants who don't 'deserve' a place to get in by a back door route.

We all talk about the wish to admit to university those who have the ability and potential to benefit from higher education.  The problem is that we have highly visible measures of ability in the shape of the results for the public examinations that candidates have taken; but we do not have any similar measures of potential.  It is clearly untenable to assume that ability as shown in past achievements is a perfect indicator for future potential.

Some have proposed that universities should use contextual information (on applicants' home background, school characteristics, neighbourhood surroundings and so on) to inform decisions on candidate offers.  But what this does is to lower the achievement, or proven ability, threshold without evaluating potential in any way.  As a geographer I am also very wary of what we call the 'ecological fallacy' - the belief that the characteristics of an area or wider population apply to all the individual elements within it.  Thus we might erroneously assume that all applicants from a particular school in a poor and deprived neighbourhood with high unemployment rates and low levels of higher education experience among parents would all share the characteristics of the neighbourhood: but the local vicar's son and daughter almost certainly don't share those characteristics, but come from a household with a very different set of life course attributes.

My own preference at this point would be for us to use certain types of contextual information to identify applicants who should then be considered in greater depth - for example through an interview - to seek to identify their potential.  But I look forward to the outcome of the review that has today been sanctioned.  And I will also be interested to see what other universities decide to do in this minefield of decision-making on applications for heavily over-subscribed places at good universities.  Others may want to add comments on this issue from their own perspectives.