Personality questionnaires are part of assessment processes frequently because they are – amongst other measures – good predictors of future job performance. One of the key assumptions here is that personality is consistent across situations and stable across time. However, some time ago we learned that humans tend to underestimate the extent to which their personality will change in the future. Thus, to what extent is personality really stable across the life span?
This is what Petar Milojev and Chris G. Sibley from the University of Auckland in New Zealand were interested in. They had almost 4,000 New Zealanders between 20 and 80 years of age complete a short questionnaire assessing six personality traits, twice with a test-retest interval of two years. The traits were extraversion, neuroticism, conscientiousness, openness, agreeableness, and honesty-humility (all of them part of the HEXACO model of personality structure). Overall they found personality to be rather stable across the two-year interval. However, stability varied with age: it increased with increasing age until the 40s or 50s and then decreased again. Interestingly, the only trait that did not decrease in stability after the 50s was agreeableness: individuals became less and less agreeable across their entire life span. Generally there were differences between the traits, with extraversion being the most stable one. Moreover, extraversion and neuroticism were most stable in the late 30s, whereas the other traits were most stable in the 40s or 50s.
Thus, personality is rather stable across the lifespan, but not entirely. It is most stable at mid-age and less stable in young and old age. Agreeableness is the only one out of six traits that constantly decreases with increasing age.
The original article was published in the Journal on Research in Personality. There is an outline of it on the BPS Research Digest Blog.
The findings have to be interpreted with some care. The study was conducted in New Zealand and what is true for NZ is not necessarily true for other countries in the world. This is particularly important because the authors themselves say that the differences between the traits in the stability peaks are due to different environmental and social influences. These may well differ across cultures. Moreover, they only looked at a span of two years of within person variability. Thus, the effects might be due to different generations studied. The picture might be a different one when following up on study participants across their entire lifespan.
However, given these limitations, one might still draw two conclusions: First of all, we can see that personality is not entirely stable. From a previous post we know that we change more in the future than we think and that this sometimes leads us to decisions we make in the present that we regret in the future. Second, personality is relatively stable and therefore it seems appropriate to use personality questionnaires in selection and recruitment.