Linking internationalisation and research productivity

The research productivity of Polish academics is strongly linked to international collaboration. The average productivity of Polish academics involved in international collaboration – ‘internationalists’ – is consistently higher than the rate of Polish ‘locals’ in all academic fields.

The impact of international collaboration on average productivity is much higher in Poland than in other European countries studied, with important policy implications.

The relationship between international cooperation and research productivity has been widely discussed, with the general assumption that collaborative activities in research increase research productivity.

But as Sooho Lee and Barry Bozeman pointed out in a 2005 study: “The benefits of collaboration are more often assumed than investigated… Do those who collaborate more tend to have more publications?”

Very much so, as will be demonstrated in the Polish case.

We analysed two specific aspects of internationalisation in research: first, the correlation between international academic cooperation in research and academic productivity and, second, the correlation between international academic cooperation in research and the co-authorship of publications with international colleagues.

The study

The data used in the study was drawn from 11 European countries involved in the ‘Changing Academic Profession’, or CAP, and ‘Academic Profession in Europe: Responses to societal challenges’, or EUROAC, projects.

The countries were: Austria, Finland, Germany, Ireland, Italy, the Netherlands, Norway, Poland, Portugal, Switzerland and the United Kingdom. The data was subsequently cleaned, weighted and integrated into a single European data set by a team at the University of Kassel.

The total number of returned surveys was 17,211 and included 1,000 to 1,700 surveys in most European countries and 3,700 surveys in Poland.

For our analysis, we used a subsample of 9,536 European academics who were employed full-time in universities (as defined by national research teams) only.

Research productivity correlation

The first research question was how strongly international collaboration in research correlates with above-average research productivity and whether the relationships hold across all academic disciplines in Poland.

Responses to the question, “How many of the following scholarly contributions have you completed in the past three years?” with the number of “articles published in an academic book or journal”, were analysed.

The analysis was conducted with reference to two separate groups of academics, termed ‘internationalists’ and ‘locals’. One group consisted of the academics indicating their involvement in international research collaboration, and the other group included those who indicated their lack of involvement.

The analysis of the Polish subsample shows that Polish academics employed full-time in the university sector are less internationalised in all academic fields than their European counterparts, but cross-disciplinary differences in internationalisation are much higher than in other countries.

Only academics in physics and mathematics collaborate with foreign colleagues to an almost equal degree (on average about three-fourths of the subsample). In life sciences and medical sciences, the proportion is about 55%, and in humanities and social sciences about 48%.

The two most internationalised clusters of fields are the same in Europe and in Poland: ‘physics and mathematics’, and ‘life sciences and medical sciences’.

On average, Polish academics across all academic fields involved in international collaboration publish more articles than those not involved.

In particular, in engineering they publish on average more than four times more (332%) articles, in physics and mathematics three times more (217%), and in life sciences and medical sciences almost 50% more than internationally non-collaborating colleagues.

The difference between average publication rates for ‘internationalists’ and for ‘locals’ is much higher in the case of Polish academics than in other European countries studied.

Consequently, international collaboration has a more powerful impact on productivity in countries that are only just entering European and global research communities.

Co-authorship

The second aspect of internationalisation studied is the difference in the share of internationally co-authored publications between the subsample of ‘internationalists’ and the subsample of ‘locals’.

The analysis of the Polish subsample shows an almost identical cross-disciplinary pattern for international article co-authorship correlating with international collaboration as in the case of the 10 European countries.

Across all five clusters of academic fields, the difference in percentages of internationally co-authored publications between ‘internationalists’ and ‘locals’ is statistically significant.

Amazingly, Polish ‘internationalists’ are more internationalised – that is, have a higher proportion of internationally co-authored publications – than European ‘internationalists’ in all academic fields except the humanities and social sciences, where they are slightly below the European average.

There are also no big differences between Polish and European averages for ‘locals’ except that Polish ‘locals’ in physics and mathematics have on average twice as high a proportion of internationally co-authored publications as their European colleagues.

Thus the European pattern not only holds in Poland, but is even stronger: while the multiplication factor between ‘internationals’ and ‘locals’ for European academics is on average between four and 7.5, the same factor for Polish academics is between four in physics and mathematics and 13 in life sciences and medical sciences.

Main findings

Our study shows that the research productivity of Polish academics – following European patterns – is strongly correlated with international research collaboration.

The average research productivity rate of Polish academics involved in international collaboration (‘internationalists’) is consistently higher than the rate of Polish ‘locals’ in all academic fields – by 60% to 140%.

Polish academics are less internationalised in terms of research than the European average, but the productivity rate of Polish ‘internationalists’ is on average much higher than that of Polish ‘locals’.

The impact of international collaboration on average productivity rates across all academic fields is much higher in Poland than in the European countries studied.

International publication co-authorship is also powerfully correlated with international research collaboration: the average international co-authorship rate is between five and 7.5 times higher for Polish ‘internationalists’ than for Polish ‘locals’, depending on the academic field.

The European pattern of a higher proportion of internationally co-authored publications for academics collaborating internationally in research compared with those not collaborating internationally holds strongly in Poland.

While the multiplication factor between ‘internationals’ and ‘locals’ for European academics is on average between four and 7.5, the same factor for Polish academics is considerably higher, from seven to 13.

In the context of the most recent Polish reforms (2009-12), which highlighted the role of international publications, the results of the present study imply a powerful policy conclusion: more international cooperation is the best way to have more internationally visible national research output.

And in the specific case of co-authoring articles with foreign colleagues, the policy lesson is even simpler: no international collaboration, no international co-authorship.

Polish academics involved in international collaboration differ much less from their European colleagues involved in collaboration in terms of patterns of research productivity than commonly assumed – the problem is the low research productivity of those not involved in international collaboration and a very high percentage of consistent non-publishers in the university sector, which is 43% of full-time academics.

Policy implications

Recent reforms resort strongly to new internationalising mechanisms: internationalisation matters heavily in institutional research assessment exercises – termed ‘parametrisation’ – which are closely linked to an institutional funding stream.

Internationalisation also matters as a prerequisite for getting access to competitive individual research grants distributed by the newly created National Research Council, as well as in new requirements for academic promotions.

In all the three areas, research internationalisation as analysed above is more important than ever. It is too early to link the correlations found to prior national strategies and ongoing reforms, but internationalisation boosts the competitiveness of Polish higher education.

What is clear is that the ever-larger proportion of knowledge is being produced in selected top university departments and most productive units of the Polish Academy of Sciences – there is stable concentration of knowledge production (and research funding) in the most internationalised institutions.

* Marek Kwiek is professor and director of the Centre for Public Policy Studies, UNESCO Chair in Institutional Research and Higher Education Policy and a chair at the University of Poznan in Poland. Email: kwiekm@amu.edu.pl. This article was first published in the current edition of Higher Education in Russia and Beyond.

COMMENTS

There is correlation, this seems clear. What I am missing is analysis on causality. Are the internationally oriented researchers more productive because they are "better" to begin with...?

Markus Laitinen on the University World News Facebook page


The direction of the causality arrow goes both ways, certainly. There are numerous conditions to become involved in international collaboration in research. These are resources available to both parties and their attractiveness to each other. Non-attractive partners will not be invited to collaborate - so, naturally, more productive academics tend to be more often invited than less productive academics. The more they collaborate, the more they produce, and get more rewarded, as the cumulative advantage theory holds. So, indeed, internationals produce more research - but also they become internationals because they are already highly performing academics... The paper was intended to show that more cooperation is correlated with more research production, and the causality goes in both ways! Poland needs both highly performing academics and internationals, and most often these two features come together.

Marek Kwiek on the University World News Facebook page


Thank you, makes sense! I would not think that the "non-internationals" could just snap their fingers and decide to become international, although attitude and willingness will likely be factors.

Markus Laitinen on the University World News Facebook page