Are Test Publishers doing the ‘diversity agenda’ a disservice?

Let’s begin with a clarification of terms: ‘Diversity agenda’? What does that mean? This article is not about pointing the ‘castigating finger’ at Test Publishers. Nor is it about attempting to juxtapose the ‘diversity agenda’ over-against the ‘agenda’ of Test Publishers as though they were in fact distinct agenda’s. If diversity is fundamentally about the effective management of all people, regardless of background, then the ‘diversity agenda’ alluded to in our topic above, is in fact at the very heart of Test Publishers agenda – whether this is understood or not.

Test Publishers have a lot to be acknowledged for: the application of the principles of psychology to the workplace, amongst other positive benefits, signalled a positive and fundamental organisational shift in assisting ensure that talent selected for recruitment and/or development purposes, were in fact selected based purely on their skills and abilities – and not on what I often call one’s ‘equality symptoms’ (gender, disability, ethnicity, etc). In this sense, psychometric testing commenced the process for actually beginning to ‘objectively’ eliminate bias from selection processes, thus overcoming the syndrome known as ‘recruiting in one’s image’.

I was reading an article in yesterday’s Evening Standard dealing with the question of female representation at executive level in the finance sector. The articles main’s point was to state that “despite all…efforts to show a modern and inclusive face, finance houses had not advanced”. But surely this is an allegation that can be levelled across a variety of sectors? The facts speak for themselves: Total female board level representation in the FTSE 100 is 12’5%, women from ethnic minority backgrounds is 0.4% – indeed a study suggests that it will take approx 60 years to have the same amount of female directors as men in the FTSE 100. And this is just to do with gender. Is recruiting in one’s image still an issue then??

The overarching question we are addressing here is in fact about improving efficiencies – which is actually at the heart of what the ‘diversity agenda’ is all about. Improving efficiencies in organisations through the effective use of people is at core of why psychometrics were introduced into the workplace – to enthrone fair selection and hence talent utilisation. But there remain questions that need to be addressed regarding the use of psychometrics and the approach adopted by Test Publishers in designing them that does not appear to assist eradicate the worrying picture described in the preceding paragraph.

Some of these questions are:

  • How widely do Test Publishers consult and involve a range of people and/or ‘groups’ during the initial design phase of psychometrics to ensure the reliability of data variables? Indeed, what lessons learnt from involving a range of people during the consulting phase subsequently inform the ‘communication’ and ‘style’ used for the actual psychometric design?
  • What proactive steps are being taken to ensure that language used in the actual tests are thoroughly inclusive and do not conjure up workplace stereotypes that adversely impact candidate performance?
  • Where evidence suggests ethnic group ‘differences’ in certain ability tests exist, what practical actions are Test Publishers taking to reduce or eliminate these? Indeed, what publicly accessible advice is provided to test administrators to ensure transparency in selection decisions?
  • As a follow-on from the above: where it is known that particular tests are know to have adverse impact on particular groups of people, is advise provided to organisations/test administrators suggesting that tests be used as part of a suite of assessment procedures to ensure fair, inclusive and transparent selection decisions? There are some organisations, for example, that solely use verbal reasoning tests as an initial ‘selecting-out’ recruitment mechanism.
  • What correlation exists, if any, between statistical measures and the actual wide experiences of individuals used in designing tests?
  • How are Test Publishers ensuring that test administrators themselves fully understand and are indeed informed of how to administer tests that bear in mind the candidate variety typical of the 21st century workplace? For example, what steps are Test Publishers taking to ensure tests are administered in such a way as to promote a Positive Mental Attitude (PMA) and/or eliminate ‘stereotype threat’ through such interventions as ensuring positive role modelling in assessment centres, the availability of timed exercises as a reasonable adjustments caveat, advise on managing disability in assessment, employing task descriptors that do not conjure up stereotypes linked with negative social identities – all of which are shown to positively enhance the psychological state needed for ensuring true demonstration of potential?
  • How do Test Publishers demonstrate due diligence to the requirements of the Equality Act 2010, and associated legislative and diversity best practice workplace proceedings and how are these communicated to organisations and test administrators?

These questions, themselves question the very presuppositions behind the supposed efficacy and effectiveness of psychometric tests and indeed the methods deployed in their design. The questions are all about providing the needed impetus for going underneath the skin of organisational processes to investigate arising development needs that facilitate efficiencies in our selection processes, thus ensuring fairness to all ‘talent types’. Indeed, it is a worthy note to mention that going underneath the skin to drive improvements is not a task limited to Test Publishers alone, but one that delves into the various parts of the employment life-cycle and is at the heart of the innovation needed for facilitating true inclusion and talent diversification.

With Lord Davies’ report on Gender Inclusivity in the FTSE 100 which drew on parallels across various countries in Europe, with talent management sitting on the top of the agenda of CEO’s as a means of stimulating the economic growth of UK Plc, and with the ‘face’ of talent drastically changing in the 21st century global marketplace, HR professionals, and those with people management responsibilities need to lead the way in going underneath the skin of organisations’ to investigate new and innovative ways of removing ‘structural bias’ from candidate selection mechanisms – of which Test Publishers must play a fundamental part. Indeed, honest answers to the questions posed above lie in a suggested solutions review – to be driven and led by the Test Publishers themselves.

So there is a call to action being made here. For acting on identified and required steps sits at the core of providing an effective and efficient service. Doing nothing about it, is what ‘disservice’ is. Indeed, not acting at all, otherwise called ‘in-action’, is what I term as the new form of indirect discrimination: not doing what we know we ought to as professionals.

The ‘ought’ suggests a deep moral imperative – as well as a fundamental baseline whereby the question our topic poses, can be initially individually (as well as collectively reflected upon)…before providing a genuine personalised answer.


Filed under assessment, diversity

3 responses to “Are Test Publishers doing the ‘diversity agenda’ a disservice?

  1. Tests begin from (hopefully) a diffent perspective then many policies in the sense that they often begin with a construct the developer wants to measure, from there they develop a method, and finally think about the legal and social issues. It is the segregation of steps 2 and 3 which are the problem in my view, in that the test content is set and prediction established before diversity issues are considered fully. What I think is required is an expectation that steps 2 and 3 will take place in tandem so that item content showing differential group performance can be explored and if necessary removed. I know some will argue that validity is everything but if you cannot sell a test because it isn’t legally compliant or use it fairly a perfect correlation with the desired outcome isn’t very useful. Of course some will argue that if it predicts what we want then psychologists shouldn’t have to be concerned with group differences and socialietal differencesbut that hard nosed business empiricism is a short term view in the global marketplace.
    In test manuals and publisher marketing materials mention of the test users statutory obligations is rare and little data or advice is given to enable these statutory obligations to be met. It appears that some publishers abdicate that work entirely to the client who is must be clear is ill equipped. Publishers are those with the data and are best placed to at least partially complete these assessments on behalf of clients. If publishers were to seriously consider how to support test users to fulfil their legal and social obligations and to evidence the impact of their efforts, they may be able to open up new markets and new opportunities to sell fairer, better structured tests.

  2. Jus a few lines, will return later in discussion I value highly. “How to test the untestable?”, it is the title of a publication by Lies Sercu on intercultural communicative competence assessment. Also studies of Fontini, Carla Deardorff .. rise the issue. Whereas scoring and rating (deductive learning, classroom) belongs to the analytical field, elecited behaviour is part of the interpretative domain (inductive learning, real world). “How effectively and appropriately can an individual behave in an intercultural context (with ­ or without- ability in the target language)?
    Notions of ‘effectiveness’ and ‘appropriateness’ suggest two views of the issue. Whereas effectiveness is often a judgment from one’s own perspective, appropriateness is based on judgments made from the host’s perspective” cites Lut Baten in discussions within the CEFcult community (EU funded program to assess oral language production for intercultural professional purposes).

Leave a Reply

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

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

Google photo

You are commenting using your Google 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 )

Connecting to %s