As universities around the world face pressure to meet employers¡¯ demands for more literate and numerate problem-solvers, a new study has found little evidence that higher education cultivates these attributes.

Research by Australian econometrics expert Ross Williams suggests that generic skills are mostly nurtured at school, and universities do little to enhance them.

Professor Williams reached this conclusion after comparing national-level scores from the Organisation for Economic Cooperation and Development¡¯s Programme for International Student Assessment (Pisa), which gauges 15-year-olds¡¯ abilities in areas including mathematics and reading, with corresponding results from the OECD¡¯s Programme for the International Assessment of Adult Competencies (PIAAC).

¡°There doesn¡¯t seem to be any development of generic skills at university per se,¡± said Professor Williams, a former University of Melbourne economics dean and honorary fellow with the Melbourne Institute of Applied Economic and Social Research. ¡°It emphasises the importance of schools in developing generic skills.¡±

³ÉÈËVRÊÓƵ

The study has been??in the?Journal of ³ÉÈËVRÊÓƵ Policy and Management. Professor Williams collected Pisa and PIAAC scores from 31 countries in Europe, North America, Asia and Oceania, undertaking exhaustive efforts to ensure the figures were comparable.

They included matching cohorts from the two assessments, modifying the Pisa data to allow for poor performers who were unlikely to have progressed into higher education, and adjusting PIAAC scores to take account of early school-leavers who had not participated in Pisa.

³ÉÈËVRÊÓƵ

PIAAC data were only collected for people aged between 20 and 24, to ensure that the findings reflected recent higher education policy.

Professor Williams stressed the ¡°pitfalls¡± of using aggregated national data and said that his study relied on assumptions. But robustness checks had corroborated the findings, which also echoed the results of earlier studies based on individual-level data.

Overall, schooling explained 70 per cent of adults¡¯ generic skills. Post-school learning, including employment and life experience as well as higher education, was responsible for the remaining 30 per cent.

The study rated Finland highest for adults¡¯ generic skills, followed by the Netherlands, Austria, Sweden and Belgium. Professor Williams said that these countries also had impressive scores at school level.

³ÉÈËVRÊÓƵ

Anglosphere countries performed relatively poorly at both stages. Professor Williams was hesitant to offer an explanation for this, stressing that he was an economist rather than an educationalist, but said ¡°maybe they don¡¯t drill their students as much. If you think of Finland and the Nordic countries, they go more for the basics.¡±

He said that the most surprising result was that there was not more ¡°convergence¡± in different countries¡¯ scores when adult skills were assessed. Most countries that performed poorly in Pisa did likewise in PIAAC.

An exception was Russia, where adults¡¯ numeracy scores were substantially better than those of school students. But this was the ¡°standout example¡±, he said.

The study found that in countries with significant higher education expenditure, universities contributed more to generic skills cultivation than their counterparts in nations that invested less. But somewhat surprisingly, it found the opposite in countries with the strongest research publication performance ¨C suggesting that when research is prioritised in university systems, acquisition of literacy and numeracy can suffer.

³ÉÈËVRÊÓƵ

Professor Williams stressed that the negative effect of research was ¡°very small¡± and said that the relatively poor generic skills in the US, UK, Canada and Australia ¨C all strong research performers ¨C may have been a factor.

john.ross@timeshighereducation.com