1 Research training and mobility of Australian scientists Richard Woolley University of Western Sydney r.woolley@uws.edu.au Tim Turpin University of Western Sydney t.turpin@uws.edu.au Abstract: This paper emerges from an ARC Linkage project looking at the research training and circulation of scientists from the Asia-Pacific region. The researchers surveyed more than 8,000 scientists based in the region plus approximately 2,000 of their co-authors from outside the region. This paper considers the role of post-graduate research training in the circulation of research scientists, presenting data about Australian survey respondents who did their PhD overseas. The first part of the analysis seeks to disentangle different factors within in the circulation of this group of respondents. Two sub-groups are described amongst Australian respondents who trained overseas, one which is characterized as representing an in-flow of skilled migration into Australia, and a second characterized as engaged in scientific mobility acquiring skills and qualifications in international locations through institutional channels. The second part of the analysis looks at the role of research training in structuring further scientific mobility, as indicated by the take up of post-doctoral positions by Australian respondents who trained overseas. The article concludes that whilst migration and scientific mobility are aspects of the circulation of science workers that are inextricably interwoven in individual decisions to move, they can be disentangled to some degree analytically to improve understandings of flows between different locations in the global context. Introduction The subject of this paper is the circulation of highly skilled science research workers. The paper addresses the issue of different patterns of movement that may be contained within the broad description of scientists mobility, from a perspective based in sociological studies of science.
2 The empirical data discussed in the paper derives from a survey of natural and physical scientists, which was part of an ARC Linkage partnership with UNESCO about the careers of scientists from the Asia-pacific region. 1 The sub-group of respondents to the survey who are the focus of the paper are those Australian nationals who undertook their research training overseas. The basic argument of the paper is that two types of flows can be discerned migration and scientific mobility and their interplay can be disentangled analytically to some extent, providing improved understanding of the factors involved in the circulation of skilled science personnel. This is not to argue that migration and mobility have entirely distinct temporal frames or determine contrasting lived realities, rather we would agree with those researchers who argue circulation is a better way of conceptualizing movements that are of varied temporalities, complex and fluid (Ackers 2004a, Iredale and Appleyard 2001). As migration researchers have noted, Australia has remained an attractive destination for skilled migrants due to a range of factors, many of which are not related to professional or career factors. Factors such a relatively relaxed lifestyle and a good environment for children are often as important in migration decisions as salary, promotion or other career development benefits (Khoo et al. 2006). This is an important context to understand in the Australian case, as when considered from the point of view of elite research scientists from North America, Europe and Japan, Australia will not usually be the most desirable destination considered purely in terms of a scientist s intellectual vocation. The paper thus uses data on research training to try and distinguish migration and mobility whilst recognizing that these patterns are likely to be formed by entanglements of these and other factors. Research training is one element that can be thought to structure the circulation of scientists. Research training has enabled highly developed countries, particularly the USA, to attract some of the most talented scientists and technologists from around the globe, seeking quality training and mentoring, the best infrastructure and potential financial rewards (Whitley 2000). Japan has also been benefiting from increasingly strong net migration of highly skilled S&T personnel since the early 1990s (NSB 2006 Vol. 1: 3-34), including for research training. However, the circulation of scientists is complex, influenced by a variety of dynamics. These include transformations in scientific activity itself and its organization, national priorities and policies (particularly education and immigration) (Whitley 2000), and issues of social and economic
3 development (APEC 2000). This is a far from an exhaustive list of the broader dynamics that shape this complexity, but which lie outside the scope of this short paper. These factors need to be balanced by attention to the training, intellectual mobility (Shinn and Benguigui 1997), and career trajectories of scientists. The geographic mobility of scientists can be understood both as a driver and a consequence of the transformation of aspects of scientific work (Mahroum 2000). This has implications for national governments like Australia s, which for better or worse, have adopted policy orthodoxies (Godin 2004) that value the production of a highly skilled science and technology workforce (human resources in science and technology or HRST) as an important element of national innovation policies and programs oriented toward economic growth and social development (David and Foray 2002; OECD 1997; Powell and Snellman 2004). Research into the production, retention and circulation of skilled S&T personnel and the structure of their careers has thus become increasingly common, as policy-makers try to assess their capacity to fill scientific and technological roles and global institutions examine ways to manage the brain drain/gain in the interests of development and fairness (InterAcademy Council 2004; OECD 2002; Ozden and Schiff 2005). There is a significant literature on the dynamics of science careers, markets and geographies of knowledge (Gaillard and Gaillard 1997; Johnson and Regets 1998; Meyer et al 2001). Recent work has also studied ways in which diaspora networks of researchers may continue to stay in touch, collaborate or seek to advance the interests of their home countries, even while living as long-term residents abroad, and how nations may best profit from these networks (Barré et al. 2003). However, there is relatively little empirical evidence on science careers and mobility. As Laudel describes, this is in a large part due to the fact that perspectives taken by international labour market and migration studies do not afford sufficient resolution to identify such small and functionally specific groups as characterize elites (2005:378). An emerging qualitative literature reporting on investigations into scientific mobility within the European Union (Ackers 2005) is correcting this situation, for example providing relatively fine-grained insights into the influence of partnering in personal relationships on forms of tied migration, and the resulting impacts on the career of women scientists (Ackers 2004b).
4 The migration currents of elite scientists have been highlighted to vary according to specializations (Laudel 2005). The thinking sciences (theoretical physics for example) and the experimental sciences (nuclear physics for example), are likely to have different relations to space and place, due to their ways of working and the techniques deployed and materials used. For example, many of the experimental sciences use large-scale research facilities or specialized laboratories in their work, one factor which has contributed to what has been described as the denationalization of the sciences (Crawford et al. 1993). Such movements may be far from permanent in nature, but it has been common for stays of two years to be referred to as migration in science mobility literature although this is commonly only the length of a single post-doctoral position (Laudel 2005). This is a weakness in much of the literature on the circulation of scientists. Such movements might be better understood as part of what Mahroum (2000: 367) defines as scientific mobility ; stays in different countries for no less than one year that proceed through channels of institutions that enjoy a high reputation for excellence and expertise. Such a description would seem a better way of describing undertaking a post-doc position overseas, particularly in the Australian context, where migration refers more frequently to longer-term decisions about settlement, adjustment and building more substantial ties to place over time (Castles and Davidson 2000; Guo and Ireland 2001). The complexity of the circulation of scientists thus remains somewhat obscure, even at the seeming fundamental level of trying to disentangle migration and relatively short-term scientific mobility. Survey respondents The data discussed here is drawn from a large survey of publishing scientists conducted in late 2005. All these scientists had published a paper in the Science Citation Index (SCI), in English in 2004. All author email addresses were collected from papers where at least one author was based in one of 23 locations in the Asia-Pacific region. These locations were identified for inclusion in the survey in collaboration with UNESCO and the survey used the SCI to obtain a sample of convenience. The survey was not an attempt to construct a representative sample of scientists in Asia-Pacific, which would likely be impractical. The survey was thus exploratory in nature, seeking to capture something of the training patterns, mobility and collaborations of scientists across this large and very diverse region.
5 A total of 10,132 useable responses were received. A raw response rate could be calculated on the basis of the number of responses divided by the number or messages delivered (n 85,000). However, this calculation would not take account of those who deleted the email upon receipt or otherwise did not read it and cannot be considered to have declined to participate. A total of 79% (8,013) of respondents were from the Asia-Pacific region. Respondents nominating India (20.7%), PR China (20.5%), Japan (16.4%). Australia (12.9%) and Korea (9.2%) comprised approximately four-fifths (79.7%) of all respondents from the region. Respondents were overwhelmingly working in public sector organizations: including universities/higher education (70.6%) and government organizations (19.4%). Only 4.3% of respondents were currently working in industry. The survey respondents who are focus of this paper are Australian nationals who did their postgraduate research training overseas (n=208). As could be expected due to the focus of the survey sampling frame on SCI authors publishing from institutional locations within the Asia-Pacific region, 80% (n=171) of the Australian respondents who trained overseas were currently working in Australia. A further 8% of Australian respondents were working elsewhere in the Asia-Pacific region, mainly in New Zealand and Singapore. Distinguishing flows within the overseas trained group Untangling factors involved in the mobility of scientists is complicated, particularly using survey data in which a complete narrative of movements and motivations cannot be established. Nevertheless, an attempt can be made to disentangle migration and what we will call, following Mahroum (2000), scientific mobility where the respondent appears to have proceeded through channels of institutions and departed Australia specifically for the purpose of undertaking their PhD. The way we do this is by controlling for migration to some extent by considering the location of the undergraduate degree of each respondent in the subgroup. The reasoning is that Australian nationality is more likely to be a product of migration for those respondents who did both their undergraduate and postgraduate degrees overseas. This is not to argue that migration itself may not have been motivated by scientific factors, partially or predominantly. This point cannot be discerned from these data, however, what we can say is that this sub-group who completed all their tertiary training overseas (n=116) are more likely to represent an in-flow of
6 skilled scientific researchers into Australia. (There may be a small number of cases where individuals departed from Australia for their entire university education.) As Table 1 shows, of those Australian respondents who completed all their tertiary education overseas, by far the largest proportion was from the UK. This is what would be intuitively expected given the historical colonial ties between the UK and Australia. Migration flows from the UK, although reduced from earlier peaks, still represent the largest single source for Australian immigration as a proportion of the total migration program. Similarly, migration flows between Australia and New Zealand are historically and currently strong (DIMA 2005). Table 1. Location of overseas degrees, Australian respondents with international undergraduate and postgraduate qualifications Degree Location of UG Undergraduate Postgraduate Location of PG degree N. % N % degree United Kingdom 44 38 54 47 United Kingdom Russian Fed. 11 10 14 12 United States United States 8 7 11 10 Russian Fed PR China 7 6 9 8 New Zealand India 7 6 5 4 Canada New Zealand 5 4 4 3 India Hong Kong 4 3 Sub-Total 86 74 97 84 Sub-Total Other 30 26 19 16 Other TOTAL 116 100 116 100 TOTAL Interestingly, given none of this sub-group did their undergraduate degree in Australia, the fact that a greater proportion of this sub-group completed a research degree (n=54) compared to undergraduate degree (n=44) in the UK shows that some Australian survey respondents had moved between a minimum of three countries. Ten Australian respondents had moved from a third country where they had done their undergraduate degree to the UK to do their PhD, and
7 similar movements can be seen in relation to the US and New Zealand. Eleven scientists wholly educated in the Russian Federation also feature in these data and it can be hypothesized that these reflect migration movements, possibly extending from the time of the disintegration of the Soviet Union. The uptake of further scientific opportunities in Australia would no doubt central to such movements; nonetheless this is the kind of circulation that can be characterized with some confidence as migratory. The alternative explanation for these data would require an even more complex movement of Australian nationals to the Russian Federation (perhaps due to family or cultural ties) to undertake their tertiary education. As no respondents in the sub-group were currently working in Russia, this alternative explanation seems far less likely than a straightforward migration in-flow into Australia. A second smaller sub-group (N=92) of those who did their PhD overseas, undertook their undergraduate degree in Australia and can be assumed to be either products of the Australian education system from the ground up or to have migrated to Australia prior to commencing university. This sub-group departed from Australia subsequent to their undergraduate degree apparently for the purpose of undertaking their PhD. Table 2 summarises the overseas research training locations of this sub-group. Table 2. Location of overseas research training, Australian respondents with Australian undergraduate degrees Country N. % United Kingdom 36 39 United States 23 25 New Zealand 8 9 Canada 7 8 Azerbaijan 3 3 Bahrain 3 3 80 87 Other 12 13 92 100. 0 The largest sub-groups of the Australian respondents who had trained overseas did so in the UK, followed by the USA, which is not surprising given the language and cultural similarities and the
8 status of these two countries as highly developed locations for scientific research. In the case of New Zealand the picture is complicated somewhat by the closeness of the two countries and the ease and frequency of movements between the two. The smaller numbers of Australian researchers who undertook research training in Azerbaijan and Bahrain are interesting. Given the divergence of language and culture between Australia and these countries, it is highly likely that those researchers who trained in these settings had pre-existing linguistic skills and possibly cultural ties to these countries. In all likelihood a combination of several factors would be needed to describe these flows fully. In comparing the two sub-groups, the UK was been the dominant site of circulation for Australian respondents, whether due to migration movements or to scientific mobility or entanglements of both these factors. Other flows can also be observed. In Table 1, Australian respondents who did undergraduate and post-graduate degrees in the Russian Federation were prominent. However, in Table 2, not a single Australian respondent who did their undergraduate degree in Australia subsequently undertook research training in the Russian Federation. This would tend to confirm the earlier hypothesis of an in-flow into Australia that can be characterized as skilled migration. In contrast, the number and proportion of Australian nationals who did their PhD in the US are both larger amongst those who had completed their undergraduate degree in Australia. This would suggest that the circulation of respondents between Australia and the US at the level of research training is primarily related to scientific mobility (Mahroum 2000) and less entangled in migratory flows. Mobility after research training A key question in understanding the importance of research training in structuring scientists mobility is the step that follows the completion of the research thesis. In particular it is interesting to consider whether research training gives access to post-doctoral research positions, and if so, where these opportunities are located. The majority of respondents in both sub-groups were successful in obtaining a post-doc research position. A total of 66.1% of the sub-group of Australian respondents who did both undergraduate and postgraduate degrees outside Australia
9 had obtained a post-doc position after completing their training, compared with 53.3% of those respondents who did their undergraduate degree in Australia. Table 3 compares the locations of the post-docs taken up by the two sub-groups. Table 3. Australian respondents with international PhDs, location of post-doc positions Location of post-doc Undergrad degree awarded overseas N % Location of post-doc Undergrad degree awarded in Australia N % Australia 30 42 Australia 20 43 USA 16 23 USA 13 28 UK 10 14 UK 6 13 Canada 5 7 France 2 4 61 86 41 89 Other 10 14 Other 5 11 TOTAL 71 100 46 100 A very similar proportion of the post-docs taken up by respondents from the two sub-groups were in Australia. Very similar and significant proportions of those who obtained post-docs outside Australia did so in North America and Europe. The difference in the types of flows that underlie this circulation may well have been masked in an aggregate analysis that did not seek to untangle migration currents and scientific mobility. What is clear is that post-doc positions provide an important entry point for research scientists into Australia. In the case of the sub-group who did not do their undergraduate degree in Australia, this may well coincide with a decision to move to Australia for the first time. Whilst post-docs thus appear to represent an important institutional factor encouraging inflows of science talent, they also appear to be an important source of opportunities for those returning from an overseas sojourn doing their PhD. However, more than half of the second sub-group who left Australia to undertake research training did not return immediately upon its completion, but remained overseas for at least one post-doc position. Of those respondents who left Australia and undertook a PhD and then a subsequent overseas post-doc position, 68% had returned and were
10 working in Australia at the time of the survey. This highlights the fact that scientific mobility can involve multiple steps that eventually lead back to the country of origin. Conclusions There are multiple aspects to the complex question of the mobility of elite workers. We have argued that analytically disentangling migration currents and scientific mobility to some extent can be an important step in gaining a better understanding of the circulation of highly skilled science researchers. Of course, these and other elements cannot be fully addressed without detailed qualitative studies, that situate such movement within the biographies and careers of individuals and the institutional and broader socio-economic contexts they confront in pursuing science research. However, the survey data analysed has been used to point out different patterns of movement within the category of the circulation of scientists within the context of the particular history of Australia as a country of large-scale and diverse migration. The specificity of the results in terms of the mobilities identified and their direction may not be common to other countries, however such analyses are likely to benefit from also taking particular national contexts into account. The movements across borders of scientists were analysed using data on education, research training and post-doctoral positions. It showed that Australian scientists have gravitated toward the major English language science research centres, the UK and the USA, in undertaking research training abroad. Analysis highlighted the fact that more than half of those respondents who had traveled off-shore for research training after doing their first degree in Australia, took up at least one post-doctoral position overseas before returning to work in Australia. International scientific mobility can thus be viewed as often involving multiple steps. Post-doctoral positions also appear as very important institutional entry points for scientists moving to Australia. This was the case both for those overseas-trained scientists we characterized as likely to be implicated within in-flows of skilled migration, and for those whose time overseas was characterized as predominantly a case of scientific mobility whereby Australian researchers followed institutional channels to acquire scientific skills and knowledge through formal research training in global settings.
11 Footnotes 1 ARC Linkage Project LP0455124. References Ackers, L. (2004a) Moving People and Knowledge: The Mobility of Scientists within the European Union, ECW paper EWC/2004/01, http://www.liv.ac.uk/ewc/ Ackers, L. (2004b) Managing relationships in peripatetic careers: Scientific mobility in the European union, Women s Studies International Forum, 27: 189-201. Ackers, L. (2005) Scientific Migration Within the EU: Introduction to the special issue, Innovation, 18, 3: 275-76. APEC Economic Committee (2000) Toward Knowledge-based Economies in APEC, APEC Economic Secretariat, Singapore. Barré, R., Hernandez, V., Meyer, J-B. and D. Vinck (eds) 2003 Scientific Diasporas, IRD Editions, Paris. Castles, S. and A. Davidson (2000) Citizenship and Migration: Globalization and the Politics of Belonging, MacMillan, Basingstoke. Crawford, E., Shinn, T. and S. Sorlin (eds) (1993) Denationalizing Science, Dordrecht, Kluwer. David, P. A. and D. Foray (2002) An introduction to the economy of the knowledge society, International Social Science Journal, 54, 171: 9-23. DIMA (Department of Immigration and Multicultural Affairs) (2005) Population Flows: 2003-04 Edition, CoA, Canberra. Gaillard, J. and A. Gaillard (1997) The International Mobility of Brains: Exodus or Circulation?, Science, Technology and Society, 2(2): 195-228. Godin, B. (2004) The New Economy: what the concept owes to the OECD, Research Policy, 33:679-690. InterAcademy Council (2004) Inventing a better future: A strategy for building worldwide capacities in science and technology, Royal Netherlands Academy of Arts & Sciences, The Netherlands. Iredale, R. and R. Appleyard (2001) International Migration of the Highly Skilled: Introduction, 39(5): 7-24. Johnson, J. and M. Regets (1998) International Mobility of Scientists and Engineers to the United States: Brain Drain or Brain Circulation, NSF Issue Brief 98-316, National Science Foundation, Washington.
12 Khoo, S-E, McDonald, P. and G. Hugo (2006) Temporary Skilled Migrants Employment and Residence Outcomes, Third Report prepared for DIMA, August 2006. Laudel, G. (2005) Migration currents among the scientific elite, Minerva 43: 377-395. Mahroum, S. (2000) Scientific Mobility, Science Communication, 21(4): 367-378. Meyer, J-B., Kaplan, D. and J. Charum (2001) Scientific nomadism and the new geopolitics of knowledge, International Social Science Journal, 53, 168: 309-321. National Science Board (NSB) (2006) Science and Engineering Indicators 2006, 2 Volumes, Arlington VA, National Science Foundation. OECD (Organisation for Economic Cooperation and Development) (1997) National Innovation Systems, OECD, Paris. OECD (2002) International Migration of the Highly Skilled, Paris, OECD. Ozden, C. and M. Schiff (eds) (2005) International Migration, Remittances and the brain drain, World Bank and Palgrave MacMillan, New York. Powell, W. W. and K. Snellman (2004) The Knowledge Economy, Annual Review of Sociology, 30: 199-220. Shinn, T. and G. Benguigui (1997) Physicists and intellectual mobility, Studies of Science, 36(2):293-309. Turpin, T., Xielin, L., Garrett-Jones, S. and P. Burns (eds) (2002) Innovation, Technology Policy and Regional Development, Edward Elgar, Cheltenham. Whitley, R. (2000) The intellectual and social organization of the sciences, (2 nd Ed.), OUP, Oxford.