Diversity FAQ’s No 14: Are psychometric tests the best way to verify ‘candidate-fit’ for the job?


 There has been a great deal of continued debate on the actual appropriateness of using psychometric testing as a key selection tool in the recruitment process. Concerns for example have been raised over the adverse impact they have with particular regard to candidates from ‘non-traditional backgrounds’ such that we ask the following questions:

  • How fair, transparent and inclusive are psychometrics at the point of making actual selection decisions?
  • Are the candidates selected, selected fairly and is due regard particularly paid to the diverse experiences, skills and backgrounds of all ‘candidate types’ at the initial point of designing psychometric tests?

An added concern also exists in terms of the fact that even the most ardent supporter of psychometrics accepts that they characteristically possess adverse impact on certain groups and are therefore not as capable, as had been previously thought, of predicting future candidate success or suitability post being recruited into particular positions. Indeed, there is the added thinking that psychometrics, if not designed and used appropriately can contribute to the syndrome known as ‘recruiting in ones image’ – leading to monocultures remaining rife at particular organisational levels and importantly, indirectly fuelling certain intellectual and social stereotypes related to particular groups of people, i.e. women, ethnic minorities, candidates with disability, etc.

In terms of organisational best practice that specifically meets the needs of today’s typical ‘global’ candidate then, it should be a fundamental prerequisite that all psychometric types (verbal, numerical reasoning, etc) be thoroughly diversity proofed or equality audited prior to being put to use to ensure they are in fact fit-for-purpose. Indeed, test publishers should consider the necessity of proactively working more collaboratively with diversity professionals in the design of tests initially.

In addition, all tests, as standard practice, should be used amongst a host of other assessment types (interviews, individual and group exercises, presentations, etc) to genuinely ensure fairness in candidate selection. This will go a long way to eradicating adverse impact – and is to be seen as a priority for promoting effective talent management and development in our individual organisations.

Suffice it to mention that the thoughts expressed above actually sit at the very heart of the corporate global agenda – alluded to in part in Lord Davies’ report last month on Gender inclusivity in the FTSE 100, but importantly, is actually all about promoting best practice and efficiencies in our recruitment, selection and development processes – regardless of particular ‘diversity based’ initiatives.

NB: Look out for a forthcoming article where we explore these issues a bit further entitled: ‘Are test publishers doing the ‘diversity agenda’ a disservice??’

4 Comments

Filed under Uncategorized

4 responses to “Diversity FAQ’s No 14: Are psychometric tests the best way to verify ‘candidate-fit’ for the job?

  1. GMC

    Hi Jude,

    This does make me wonder if certain psychometrics (which carry quantified adverse impact) could be moderated by the intelligent use of different norm tables for scoring the performance, to then eradicate any adverse impact.
    This would I think lead to some interesting questions over the consistency of the test – a solution might be to utilise raw scores with a baseline cut off point for all candidates to determine capability to do the role in an absolute term, then the normative percentile scores can be used to discriminate quality of candidate without prejudice.

    I am however going to pull you up for only mentioning adverse impact on non-traditional backgrounds, surely our concern is inconsistent impact across groups and a favorable impact on non-traditional backgrounds would be an equally important a concern to ensure that we do not swing into positive discrimination. This preaches to the converted I know but it’s the pedant in me that had to point it out.

  2. Christina Strupinska

    I would be interested in knowing how the “adverse impact” manifests itself; that doesn’t seem to be addressed.

    • Hi Christina,

      Gregor’s point is absolutely spot on: Adverse impact in psychometrics really begins when at the point of desiging the tool – which should involve consultation and engagement with a range of people or ‘groups’ to understand their backgrounds, experiences, skills, etc – particular experiences relating to particualar groups are not as widely consulted on (or taken into consideration) and as such has a negative impact in the overall design and ‘flavor’ of the assessment from the outset. The consequence of this ‘reduced engagement’ means that the final solution – the test itself – is likely to be biased in favor of the people with whom the greater consultation has taken place.

      The resultant consequence of the above is that there is a lesser likelihood of success for candidates from those groups whose experiences backgrounds, contexts, etc – may not have been taken into account in designing the tests themselves.

      Unfortunately, research shows that this appears to be the case particularly with women, ‘minority groups’, people with disabilities and ‘the disadvantaged’ – which is the specific point Gregor has made in previous comment.

      With thanks
      Jude-Martin

  3. GMC

    Christina, You’ll find numerous studies on the subject it roots down to the fact that different cultures have different reference points and ways of demonstrating intelligence and often psychometrics are presented in their form of questions etc with a very western perspective.

    A good article to start with is this one at the BPS website which actually dates back a few years but is still relevant I think.

    http://www.thepsychologist.org.uk/archive/archive_home.cfm/volumeID_20-editionID_148-ArticleID_1199-getfile_getPDF

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s