Hi Nicola, welcome to the communities. While you wait for some answers from the community you may wish to do a quick search for historic threads as this subject does crop up quite regularly. Also, I just wanted to check that the individual(s) with responsibility for running these tests are at least Level A qualified in psychometric testing? :-)

In reply to Clare:

Hi Nicola and Welcome!

Think your employers need to be clear about exactly what they're wanting to ascertain with all this - e.g. Is it personality profiling or reasoning ability or problem solving ability or whatever or in what combination?

Major providers such as Savile and Holdsworth or Thomas International tend to charge major fees, but will advise on which profiling material might fit the bill for the kind(s) of job role you have in mind. Seem to remember too, that at least one of these providers will use the BPS accreditation on your behalf and mark and interpret and feed back the results to applicants directly, thus freeing yourselves of this responsibility.

But, for entry-level junior / trainee roles, personally I'd be questioning the overall cost - effectiveness of paying for psychometric assessment compared with using other usually readily-available - and 'free' - soures of evidence for selection decisions.

Hi Nicola... you've posted this to the 'Ireland' group, so just checking you are based in the Republic?

In reply to David:

Hi David,

Thank you for your helpful reply. The purpose of introducing this step would be to generate a personality profile of the candidate. I believe Thomas International offer BPS accreditation but as you addressed, there are other elements to be considered such as overall cost-effectiveness.

Would you have any recommendation of free sources of assessments you think might be suited to offer a sample or allow us to pilot the introduction of assessments to our recruitment process?

In reply to Steve Bridger:

Hi Steve,
That's correct, we are based in Cork, Ireland.

Hi Nicola, I am based in Kerry now and was in the UK originally. I trained with SHL (Savil and Holdsworth) Level A & B as John stated below, SHL who are now CEB have a full range of testing available. Please check this out www.cebglobal.com/.../practice-tests If you need any further help let me know. I would suggest you need to understand what you want to achieve and then think about the design on the test. Values, Skills, Behaviors or a mixture. Sean.

You should start by profiling existing employees to determine which profiles correspond to high performance in these positions otherwise your test results on candidates will be invalid for selection purposes and may, in some cases, infringe regulations on askling questions unrelated to the ability of a candidate to perform the work required (especially psychometric profiling where many questions are about social situations).

In reply to DAVID GRIPPER:

Hi David

Afraid I cannot agree with the requirements you outline. Re being invalid for selection purposes, the material from the major test suppliers will invariably have been validated by the supplier, in trials elsewhere. It can of course be possible that these trial populations etc aren’t in fact the same as yours in significant ways, therefore the profiling results won’t be valid predictors of being a success in your particular workplace. BUT validating such material in your particular workplace is at best potentially very difficult and hugely expensive if you’re going to try to make it a credible ‘scientific ‘ trial and is at worst case impossible if for example the numbers involved in your workplace trial aren’t statistically significant anyhow. It’s an enormous potential can of worms and most employers don’t bother and rely on the expert guidance of the psychometrics suppliers.

Nothing wrong with doing that IMHO, so long as the above potential validity problems are borne in mind. After all, it’s just a selection tool to be used as a guide alongside other indicators. And, so long as it isn’t amounting to unlawful discrimination, the employer can use whatever selection criteria they like - for example some have been known to use graphology or astrology or firmness of handshake or tidiness of appearance.

Secondly, by ‘infringe regulations’ assume you mean be unlawfully discriminatory. Again, the suppliers are usually at great pains to remove the possibilities of cultural etc bias from their material. It can still happen in some circumstances eg that the end user makes selection decisions on the basis of these materials to unlawfully discriminatory effect, such as applying material needing good reading and writing skills successfully to complete to people who might be dyslexic.

All this is of course why supply of proper psychometric selection material is restricted to users who have been suitably trained to BPS standards which ought to lead to their understanding of the limitations and potential pitfalls involved in their use.

In reply to David:

Hi David,
Probably the most important comment that you make is that 'the employer can use whatever selection criteria they like' and, unfortunately, many do. I see one of the roles of the CIPD as bringing a more professional approach to recruiting.

I agree that the results of the tests have been validated in terms of personality but was making the point that they are not predictive of performance unless you have data from top performers in your company or business sector to use as comparisons. As you point out, this is expensive for small companies and sample sizes may lead to unreliable data but specifying personality criteria for recruits without it may be an even bigger waste of time and money.

My comments on regulations relate to legislation in some countries and best practice in many that prohibits personal questions during recruitment that are unrelated to a candidate's ability to perform the work required of them. The nature of the questions and test conclusions can also lead to issues related to privacy. Furthermore, some tests can identify physical and mental disabilities and are, as you pointed out, consequently discriminatory and illegal in some countries.

In the case in point, I suspect that the intention is to use a zero cost on-line test which may not have the safeguards and reliability that the professional providers you recommend ensure. I have difficulty in seeing this use as contributing to a more professional or predictive recruitment process for this company.

In reply to DAVID GRIPPER:

Thanks for explaining, David!

In reply to DAVID GRIPPER:

Hi David

I don't know what tests you use, but part of the briefing to candidates taking the OPQ (I did my training with Saville and Holdsworth, now CEB) is that the person should think of themselves as they are at work, not as they are in a social setting.

Many organisations do not have the numbers of staff or the resources to carry out the type of study you have recommended to Nicola. That's why the results are compared to norm groups. The tester is able to select which norm group is the best fit.

I'd suggest you take a look here: www.cipd.co.uk/.../selection-factsheet

While the CIPD's view is that psychometric tests *can* be predictive of performance, this view is qualified.

Psychometric tests such as the OPQ do ask personal questions. The answers to questions in the format of which of 4 adjectives is most like you and which is least, are used collectively to produce a profile. However, since the days when tests started being carried out online rather than pencil and paper, no one gets the answers to individual question so there is no privacy issue in the form that you suggest. Anyone qualified to administer or interpret tests should be explaining to applicants before they take the test how the results will be used and who will have access to them.

As I have only practiced in this country, I don't know which countries you are referring to when you say "some tests can identify physical and mental disabilities and are, as you pointed out, consequently discriminatory and illegal in some countries". I was taught that psychometric tests such as the OPQ are *not* suitable for diagnosing mental or physical disabilities. I very much doubt UK companies are using tests designed for that purpose in their selection process.

Part of the training to interpret psychometric testing involves a heavy emphasis on using the test result as just one element in the assessment process to create a more rounded assessment.

When I did my training, it was also stressed that the best way to use the results was to discuss them at interview with the applicant. This gives the person to opportunity to express their own views about the result e.g. to explain that a major event outside work was influencing the answers that produced the result.

Finally, I don't think we can assume that Nicola's organisation doesn't care about safeguards or reliability. If Nicola didn't want a reputable test, she wouldn't have come to this forum for recommendations.

In reply to Elizabeth Divver:

Indeed, Elizabeth!
"the best way to use the results was to discuss them at interview with the applicant. This gives the person to opportunity to express their own views about the result"
I had the tremendous luck to have been trained by Peter Saville, Roger Holdsworth and Gill Nyfield in the late 70s, somewhere in the wilds of Buckinghamshire.
The very clear message that I took away from their training was that the results of the tests should be seen as a "window through which the candidate is invited to look at themself", and in no way a prescriptive tool.
By getting them to "look at themselves from the outside", it made it a lot easier to get candidates to discuss the degree to which the test results aligned with how they saw themselves in a working context.
If they disagreed with the description, that also opened avenues of discussion where they could be invited to give practical examples as to why they disagreed.
I found these follow-up exchanges more useful than the raw test results when recruiting.

Thank you everyone for your insights, recommendations and contributions to this post. To offer an update, we investigated a number of options regarding the introduction of psychometric testing during recruitment and decided not to pursue it as of yet. As a number of people have mentioned, the use of psychometric tests are not in themselves enough of an indicator of a candidates suitability for a role and should not be used in isolation, but as part of a robust selection process. We dug a bit deeper with our leadership team to understand and appreciate what they see as the positives and negatives when it comes to recruitment.

From this, we reviewed and revamped our interview process, created a more comprehensive interview guide (incl. a refresh of our competency based scoring system) for our Team Leads and focused on providing additional training for our Interviewers. This was done through collaboration and utilizing the feedback from our team and proved to be very successful during our recent hiring phase.

We have been extremely happy with our recruitment process in the past but I think this exercise really allowed us to get fresh eyes on our process and share/work collaboratively with our leadership team.

Wishing everyone all the best,

Nicola

In reply to Nicola:

My own experience is that personality profiling tends to be useful in confirming / triangulating other indicators but not much more than that. And, whilst personality traits are usually fairly stable, they can be somewhat fluid as well.  Some people for example are prone to (as TS Eliot put it) "To prepare a face to meet the faces that you meet"

See far more prosaically eg
www.ncbi.nlm.nih.gov/.../

In reply to David:

Hi David

Very true.

If people are not fed up with me going on about my training, something else that was covered in quite a lot of depth was which traits tend to be stable over time and which tend appear to be affected by age. I say "appear to be" because it could also be a cohort affect. One example that I remember is that people tend not to move much throughout a lifetime on the optimism - pessimism scale. If you are an optimist, you will still be an optimist after you've lost your job and your house has burned down.

Some tests include questions intended to indicate whether the person is to answer according to what they think is the desired answer rather than just allowing themselves to react honestly. However, we were told that this could also be influenced by the candidates' view of themselves - some people really believe they are marvelous and others are self-deprecating. Tests ask people about themselves and the result will only be as good as their self-knowledge. However, when you get practiced at interpreting results, you can get a feeling that the profile just doesn't seem to hang together. I would then test that intuition by discussing the profile with the candidate.

I don't do much testing these days but in a previous job I did a lot. My personal experience is that a valid and reliable test can be an extremely useful tool, if you know how to use it.