Cookies?
Library Header Image
LSE Research Online LSE Library Services

Selection under uncertainty: affirmative action at shortlisting stage

Bovens, Luc (2016) Selection under uncertainty: affirmative action at shortlisting stage. Mind, 125 (498). pp. 421-437. ISSN 0026-4423

[img]
Preview
PDF - Accepted Version
Download (327kB) | Preview

Identification Number: 10.1093/mind/fzv157

Abstract

Choice often proceeds in two stages: We construct a shortlist on the basis of limited and uncertain information about the options and then reduce this uncertainty by examining the shortlist in greater detail. The goal is to do well when making a final choice from the option set. I argue that we cannot realise this goal by constructing a ranking over the options at shortlisting stage which determines of each option whether it is more or less worthy of being included in a shortlist. This is relevant to the 2010 UK Equality Act. The Act requires that shortlists be constructed on grounds of candidate rankings and affirmative action is only permissible for equally qualified candidates. This is misguided: Shortlisting candidates with lower expected qualifications but higher variance may raise the chance of finding an exceptionally strong candidate. If it does, then shortlisting such candidates would make eminent business sense and there is nothing unfair about it. This observation opens up room for including more underrepresented candidates with protected characteristics, as they are more likely to display greater variance in the selector’s credence functions at shortlisting stage.

Item Type: Article
Official URL: http://www.oxfordjournals.org/our_journals/mind
Additional Information: © 2016 The Author
Subjects: B Philosophy. Psychology. Religion > BF Psychology
Sets: Departments > Philosophy, Logic and Scientific Method
Date Deposited: 29 Jul 2015 10:52
Last Modified: 21 Apr 2018 23:00
URI: http://eprints.lse.ac.uk/id/eprint/62841

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year

View more statistics