While my recent post on personality testing attracted some unexpected attention (from an organiser of the course), it did also generate some interesting discussion. Some I have spoken to have found more value in personality tests like Myers-Briggs and HBDI than I have, while others shared my irritation.
The reason I am posting again on this subject is because an old friend of mine, who is a practising psychiatrist had emailed me with his thoughts on the topic, but preferred to remain anonymous and not post on the forum himself. He was, however, happy for me to share them, so here they are:
- A lot of these tools fall down in the assumptions that personality is stable across contexts and over time, neither of which appears to be the case. Long live the reformed home devil/work angel!
- Define these terms – temperament, character, personality, identity. Are we talking venn diagrammes or different languages here? What is the relationship between personality and cognitive style? How honest are people with themselves, let alone their employers, or perhaps more importantly, how good are people at judging themselves? Do you think the corporations peddling their mindreading wares have answered all these questions sufficiently, because if not, how can you be sure what their tool is measuring?
- One of your commenters proffers a positive test/retest experience of Myers Briggs, but from what I remember reading these tests have pretty poor reliability compared to the more rigorous tools used in clinical research such as the Minnesota Multiphasic Personality Inventory (MMPI).
- The same learned commenter picked up on the dispute between type theories proffered by the mad Swiss and taken as gospel by MB et al (even if in somewhat dodgy translation), and the now ascendant trait models. The current banker is the Five Factor Model, with spectra of introversion/extraversion, openness/closedness, conscientiousness/the lack thereof, agreeableness/disagreeableness and neuroticism/emotional stability. In of themselves, they do have face validity and have been extensively replicated, but how far they take us in further understanding personality is unclear. There is quite a bit of research trying to link specific traits to everything from genes to neurotransmitter systems, with interesting if inconclusive results. What are the chances of successfully finding a key if you’re not sure of the shape of the lock? I’d be interested to know your take on the factor analysis techniques used to derive these traits.
- As for the Sixteen Types in the MB, I fear this is moving into transformative hermeneutics of quantum gravity territory. To be fair, the typology of abnormal psychology I was taught at Shrink School is little better, with its borderlines and its narcissists, but it has lingered this long as it has everyday clinical utility and nothing better has come along to replace it. Until now. It seems that it may, in the next version of the Bible, DSM-V, be replaced or complemented by a four factor trait model which you won’t at all be surprised to find fits very neatly with black bile, yellow bile, phlegm and blood. And who says those Greek geeks weren’t onto something!
Possibly Related Posts (automatically generated):
- I have a love/hate relationship with psychometric testing (19 October 2009)
- I Hate Personality Tests (20 August 2008)
- Reduce, Re-use, Recycle (23 September 2009)
- Coffee day 1 (2 December 2010)