Who do you think would do better at school or in the workplace: someone who is smart, but lazy, or someone who is not naturally brilliant but will keep working at a problem until they get it right? Intuitively, we know the hard workers give the naturally gifted a good run for their money. But, this hasn’t always been reflected in policy or how we instruct our teachers.
Consider the extensive focus on NAPLAN in Australia. These sets of literacy and numeracy tests make up a large focus of the My School website, were used extensively in arguing for the Gonski reforms, and feature heavily in the school-based targets for the Closing the Gap policy agenda.
There is nothing wrong with using literacy and numeracy to target and evaluate policy or to assess our students. But, are we missing something equally important? An emerging body of research summarised in a new Brookings Institute paper suggests we are.
This research is contributing to a broader set of skills, often called “non-cognitive ability” which refers to attitudes and behaviours that lead to success but can’t be captured by cognitive tests. Non-cognitive skills include perseverance, hard work, motivation, self-control, self-efficacy, interpersonal skills and communication skills.
An early set of research on non-cognitive skills analysed students in the US who completed the General Education Development program. This certifies that students have equivalent cognitive ability to a high school graduate. It sounds like a good investment, but it turns out that those who complete the General Education Development have poorer outcomes than high school graduates and, what’s worse, do not have better outcomes than “high school dropouts”. Authors of the study concluded that General Education Development students lack the non-cognitive skills of their school-completing peers.
Non-cognitive ability also comes up when looking at the long-term benefits of investment in early childhood. A number of influential studies showed that while initial increases in cognitive ability (like IQ) dissipated, the effects of high quality preschool on non-cognitive ability like motivation and behaviour held. What’s more these improvements predicted a range of positive long-term outcomes.
How do you measure non-cognitive skills?
The complexity and breadth of the term “non-cognitive skills” has impaired its use in research and policy. But, when we focus on specific measures, results are quite strong. In the study mentioned earlier, authors show a significant relationship between drive and prudence and higher educational attainment.
There has been much less research on non-cognitive skills in Australia. What has been done has tended to show that non-cognitive skills have significant effects on employment outcomes. Until recently though, there has been little comparative data available to see how non-cognitive skills vary within Australia, or across comparable countries. The most recent version of PISA, however, fills some of this gap.
In 2012, students in 66 countries or regions (including Australia) were asked two sets of questions related to their perceived control over their own outcomes, as well as their perseverance. We compared these two measures of non-cognitive skills with test scores across six other major English speaking, developed countries.
Australia did relatively well in terms of test scores, but relatively poorly on the two measures of non-cognitive skills. Significantly lower than Canada, the UK (including Scotland) and the US in terms of perceived control, and significantly lower than Canada, Ireland and the US in terms of perseverance. To the extent that these 15 year olds will be competing in a global labour market in 5, 10 and 50 years time, our measured non-cognitive skill is a serious cause for concern.
The things that predict non-cognitive skills in Australia are equally troubling. We found preschool attendance, parental education and parental employment were all found to be significantly associated with the two measures of non-cognitive skills.
What is perhaps most troubling is that females in Australia were found to have substantially lower levels of self-perceived perseverance. This was also true when we undertook similar analysis for almost all other English-speaking countries. The important exception to this finding was the US, where there was no significant difference between males and females in self-perceived perseverance.
The policy response
Richard Reeves from the Brookings Institute argues that in terms of policy responses to the literature on non-cognitive skills:
step one is to harness education and put character development firmly on the school agenda; step two is to invest in parenting, especially in the very early years.
Schools should be assessed on more than what they do for literacy and numeracy. Policies should be evaluated for their effect on a broader range of skills. Ultimately, what we measure matters.