T: 01423 563644   E: info@bloojam.co.uk     Follow us:


Bloojam Blog - Recruitment, People Development & HR Bloojam

Return to Blog page

Normative or Ipsative. Which Personality Questionnaire should I use?

Thursday, December 8th, 2016

Normative personality tools allow people free rein to answer each question in the questionnaire along a scale, such as ‘Strongly agree’ to ‘Strongly disagree’.  Their answers are then compared to a large comparison or “norm” group to give them a position on a personality dimension or scale that is relative to that comparison group.

Ipsative formats (also known as “forced choice”) force people to choose between a set of words, statements or group of behaviours by, for example, asking them to select which one from the group is “most like me” and which one is “least like me”.

Put simply, the main advantage of the normative format is that you can compare candidates to each other, because they are each being measured against the same scale.  So, for example, you can see how “Sociable” one candidate is compared to another.  The potential issue is that this style of questionnaire is more open to “faking” by the candidate, as they could in theory rate themselves “highly” on everything (albeit it shouldn’t be too obvious what the best answer is each time, and good questionnaires will have a “faking” measure so that you can challenge any candidate with a high “faking” score).

So how big a problem is faking in reality? There is limited research in this area but we would always recommend that the questionnaire output is “validated” during a subsequent feedback session and interview – so that you can explore the results and probe deeper with your candidate into any areas of concern for the role.

Nonetheless some employers may feel uncomfortable using a normative version in a competitive (selection) scenario, especially when the ideal role characteristics can be easily guessed (e.g. in sales).

Ipsative formats aim to overcome any issues of “faking” because people don’t have free rein to answer each question; they are almost impossible to manipulate and the options the test takers have to choose between can all look favourable.  Some researchers have argued though that ipsative formats still suffer from “socially desirable” responding but the issue is just more hidden.

Another issue with ipsative questionnaires is that you can’t truly compare candidates to one another because you are only looking at an individual’s behavioural preference relative to their other preferences.  For example, their own strongest preference may be “Sociability” but this doesn’t allow you to say how “Sociable” they are compared to the next candidate. Many experts therefore suggest that ipsative tools are more appropriate for development interventions than for selection purposes.

Some newer tests to the marketplace have starting to use a combination of both formats– ‘nipsative’ – which essentially means you can compare candidates whilst having some confidence that the questionnaire was harder to fake.  At Bloojam we are increasingly using these types of questionnaires for both selection and development, and there are some really powerful examples of these available.

So far so good, but here is a more advanced geek issue that test users should consider when selecting their tools:


The Ipsative format when used to assess a small range of personality categories can lead to inaccurate results

As described above, Ipsative (or forced choice) format means that the questionnaire presents some options that you have to choose from – e.g. this behaviour is “most like me” and this behaviour is “least like me”.  The questionnaire then assigns these scores to a number of personality categories or dimensions. The technical issue with this is that essentially there is a finite “bucket” of scores for divvying up. If someone has a high score in one area it means another area must score lower. This can result in inaccurate or overly exaggerated profiles.

For instance, consider these over-simplified examples from an imaginary 4- category questionnaire:

–       Someone is actually very high in all 4 categories, but there aren’t enough scores to go round to show this: they have to be scored lower in 1 or 2 categories, giving them some “false negatives”.

–       Someone is actually low in all 4 categories but their scores have to be assigned somewhere, resulting in some “false positives” – potentially showing strengths where there aren’t any.

Essentially, the 4 categories are interdependent, in that your score on one affects your score on another.

Many questionnaires that use an ipsative format tend to assign scores across 16-32 (or so) traits, and research suggests that the issues are minimal beyond 12 categories. So the effects are much reduced and you won’t get the same false positives or negatives, perhaps just a somewhat “spiky” profile.

BUT, as described above, a big word of caution is needed when there are only a small number of personality categories being measured by an ipsative format (such as certain DISC-based tools that only provide scores against 4 categories).  Here the issues of interdependence between the scales really come into play.

To be fair to many test publishers, some of the questionnaires that fall into this category (MBTI, Disc Profile) don’t recommend/allow their use for selection purposes at all, and limit their application to individual development, team building and conflict resolution.

For the true geek, search the British Psychological Society’s test reviews for one DISC-based questionnaire – Thomas’ PPA, which highlights these (and other) issues in a far more technical and eloquent way.