After hearing about it on TWIS I spent some time reading Other-rated personality and academic performance: Evidence and implications by Poropat. This paper is a meta analysis of various works and (TLDR): indicates that ‘intelligence’ is not as good an indicator of academic performance as is ‘conscientiousness’ (my loose interpretation of this is: ‘willingness to work hard’) and ‘openness’ (my loose interpretation of this is: ‘curiosity and interest in a subject’).

My ears really perked up when this was mentioned on the TWIS podcast as it is something I have always believed myself. This is possibly something to do with my interaction with research students. I also believe it is linked to my own educational trajectory that really benefited from my high school physics teacher who managed to show me that hard work paid off. I blogged about that here where this photo (showing a report card claiming that I did not work hard enough) is posted:

The paper

Overview of factors that potentially have an impact on academic performance.

The paper starts off by giving an overview of the ‘general intelligence factor’ denoted by g which has apparently been closely associated to academic performance. The contrast to this (again discussed in the paper) is the Five Factor Model for personality which maps individual personality to the following 5 dimensions:

  1. Agreeableness
  2. Conscientiousness
  3. Emotional Stability
  4. Extraversion
  5. Openness

One of the difficulties I had with understanding this paper was with the psychological vocabulary that I was not familiar with. Those 5 dimensions are not always easy to define but a loose (and almost certainly) incorrect interpretation of Conscientiousness is an individual’s ‘work-hard-ability’. My interpretation of Openness is an individual’s ‘want-to-learn-stuff-ability’ but the paper goes in to a pretty good discussion of each of those so I would recommend taking a look.

The paper gives a nice description and review of each dimension. I am mainly going to concentrate on what the paper says about Openness and Conscientiousness but there was one particular thing said about Extraversion that I thought was worth noting:

“Although more extraverted students may get greater attention leading to higher performance at primary level, the reduced strength of teacher student relationships at higher levels of education appear to eliminate this effect.”

Whilst this effect no longer being present at higher levels of education is perhaps a positive I feel that it does not show a ‘levelling of the playing field’ but rather that the student specific education has a lesser emphasis at higher levels of education (that is a beast of a long and terribly written sentence, it is late and I am too tired to fix it so instead added this parenthesis that makes it even longer; if you have read this far: I congratulate you). I will not dwell on this as I am not sure it is the main point of the paper nor that there is a quick solution (I teach 150+ students in my first year class…) but I thought it was interesting.

Self vs non-self evaluation

One of the important things when trying to correlate academic performance and personality is obviously getting the correct measurement for the Five Factor Model. An in depth overview of self versus non self evaluation is then given in the paper and Poropat describes how various studies have shown that non-self evaluation is a better predictor of academic performance than self evaluation (note that at this point no comparison is given to the general intelligence factor - that comes later). I think this kind of points to the idea that ‘teachers and peers know you better than you know yourself’ (or at least in terms of the Five Factor Model). It is particularly relevant to the work of Poropat’s paper as he then collects studies that look at the correlation of the Five Factor Model with academic performance: in particular only non-self evaluation studies are considered.

Using the Five Factor Model

One interesting aspect of the paper is that it emphasises that certain pedagogic approaches would be better suited to certain personalities:

“For example, discovery learning approaches help students who are higher in openness to learn while students lower on openness are aided by programmed instruction…”

I need to think about this in relation to the fact that there is other research that shows that students achieve better academic performance in an active learning environment (such as discovery learning which is apparently another term for inquiry based learning).

Hard work and curiosity are better predictors of academic performance than intellect

This is one of the nicer takeaways of the paper, by analysing 16 reports of studies that linked the five factor model to academic performance; Poropat shows that the correlation of Conscientiousness is stronger than the correlation of Openness. This is in turn stronger than previously reported correlations of the general intelligence factor.

I am sure that there will be studies and findings that report different things but I know that I’ll be using this meta analysis as a basis for further pedagogic work (in particular this paper will be very helpful for my current undergraduate research student and I: we are looking at student personality in a flipped classroom).

I know that I have always had a major preference to work with students that are (according to my personal evaluation) high on the Conscientiousness scale.

TLDR: Summary

I realise that I really have rambled in the above (the notes in my notebook are far messier still) so here are 3 bullet points I would take away from this paper:

  • There exists a 5 dimensional measure of personality;
  • Non-self evaluated versions of this measure are more accurate with regards to academic performance;
  • There is evidence that hard work is a better predictor than intellect for academic performance.

If there was one point I wish all students would believe it would be that last one. I often feel that some students believe that they have to settle for a certain level of achievement (‘because that is just how smart they are’) and this is something I have never personally been satisfied with. This is mentioned in particular in Poropat’s paper as there is evidence that personality can be modified more easily than general intelligence factor (even in older students).

It also seems evident as students of higher intellect would perhaps have been used to getting by without effort. Once things became ‘hard’ then perhaps those students who are used to working hard could indeed achieve success…

### My personal experience

To finish off this blog I thought I would throw in an xkcd style graph of my own personal ‘academic journey’ (to fully understand it, the blog post about my Physics teacher I linked to earlier is probably of interest):

(Matplotlib code here if it is of interest)

As a general summary I can certainly say that I firmly believe that any/all of my academic achievements have had very little to do with my ‘intellectual ability’ as opposed to my work ethic. Here’s a quote of Larry Bird that I really fell in love with the first time I heard it:

I’ve got a theory that if you give 100% all of the time, somehow things will work out in the end.