View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Apollo 84

Postdoctoral research 4. that postdoctoral research positions provide an positions as preparation for opportunity for the incumbents an academic career to concentrate solely on research. GERLESE S. ÅKERLIND The Australian National University Introduction

Centre for Educational Development and Whilst studies of doctoral research Academic Methods, training have been growing apace since The Australian National University, 0200 Australia the 1990s, studies of postdoctoral E-mail: Gerlese.Å[email protected] research training are still relatively Phone: + 61 2 61250056 rare. Nevertheless, the nature of Fax: + 61 2 61254023 postdoctoral training and the career prospects of postdoctoral contract Abstract researchers (PDRs) has been receiving a growing degree of attention over the Discussions of the nature and purpose past decade (Åkerlind, 2005; of postdoctoral contract research Thompson et al, 2001; Nerad and positions is an area where assumptions Cerny, 1999; Helbing et al, 1998; and stereotypes tend to predominate. Science journal special issue, 1999). This is due to (a) recent changes in the This is primarily in response to higher education sector that have reduced academic employment impacted on postdoctoral positions in opportunities for PDRs, as an outcome a way that conflicts with traditional of dramatic rises in the numbers of expectations, and (b) a relative lack of PhDs and PDRs without a data and publications on postdoctoral corresponding rise in the number of positions, which creates a climate in academic positions. which stereotypes can continue relatively unchallenged. This is In the UK and Australia, government unfortunate, because it limits the policy papers have emphasised the ability of supervisors to provide sound need for broader skills development for career advice to their postdocs as well those in postdoctoral positions, as as the ability of postdocs to make career preparation for both academic informed career decisions. and non-academic employment (Marceau and Preston, 1996; HM Based on an extensive study of PDRs Treasury, 2000; Borthwick and in Australia, this paper challenges four Wissler, 2003). In line with this, commonly held assumptions: research funding bodies in both 1. that postdoctoral researchers countries have been supporting want an academic career; opportunities for broader skills 2. that postdoctoral research development of PDRs. The primary positions provide a stepping research funding body in Australia, the stone to academic careers; Australian Research Council (ARC) 3. that postdoctoral research offers a teaching and research positions provide an fellowship, on a 75% ARC funding opportunity for novice and 25% host institution funding basis. researchers to become In the UK, the research funding increasingly independent; and councils have agreed on a joint skills training statement for the PDRs (and 85

PhDs) they fund (Research Councils researchers to become UK, 2001). This requires 10 days increasingly independent; and training per year in personal and 4. that postdoctoral research professional skills for each student, positions provide an and provides some £500 per PDR to opportunity for the incumbents fund such training activities. In to concentrate solely on addition, the European Community has research. established a European Charter for These assumptions derive from Researchers and Code of Conduct for traditional views of the nature of the Recruitment of Researchers, with postdoctoral positions, but this paper the aim of providing more open and will argue that each of these sustainable career prospects for assumptions needs to reassessed in the researchers (European Commission, current higher education climate. 2005). Methods These policy emphases acknowledge that there are not enough academic The arguments presented here are positions available for the number of based on three sources of data: PDRs seeking them, that many PDRs 1. an online survey of 1011 PDRs move into non-academic careers, and from 38 universities in that broader skills training is required Australia; to address the breadth of career options 2. an in-depth interview with a for PDRs. However, they do not sample of 22 PDRs, selected consider the way in which these from those who completed the changes are being experienced by survey; and PDRs. 3. an in-depth interview with 10 supervisors of PDRs, This paper is based on extensive study nominated by the PDRs who of postdoctoral training and were interviewed. employment outcomes in Australia. The study consisted of a questionnaire Survey of PDRs survey of some 1011 PDRs, selected In common with other countries, the from the range of universities in first problem with a survey of PDRs Australia, accompanied by follow-up was, first, defining, then second, interviews with 22 PDRs, plus 10 locating the population of PDRs. postdoctoral supervisors. The overall results of this study have been reported A consistent problem besetting elsewhere (Thompson et al, 2001). research into PDRs is the lack of an The purpose of this paper is to use the agreed definition as to what constitutes data to explore four commonly held a PDR (Åkerlind, 2005; Thompson et assumptions about the nature of al, 2001). There is variation between postdoctoral positions: universities in the titles assigned to 1. that postdoctoral researchers PDRs, and in whether these want an academic career; researchers are classified as academic 2. that postdoctoral research or non-academic staff. Further, many positions provide a stepping PDRs work outside of universities, stone to academic careers; conducting research for industry or 3. that postdoctoral research government. As the focus of this study positions provide an was on postdoctoral research in opportunity for novice universities (also representing those 86 researchers most likely to be interested these variations,, issues and in an academic career), and within expectations facing postdoctoral Australian universities postdoctoral researchers are similar internationally contract research positions are most (Thompson et al, 2001; Nerad and commonly classified as academic Cerny, 1999; Helbing et al, 1998; appointments, the definition used for Science journal special issue, 1999). this study was: a fixed term, research only, base-grade academic Once defined, the problem of appointment, with PhD qualification. identifying the population was addressed. In no case was there a It should be noted that this definition separate institutional record of PDRs; excludes three categories of contract in all cases PDRs had to be identified researchers that other studies may from a larger base of university include, i.e. fixed-term researchers on personnel, often on the basis of their academic appointments but without a position name. This means that the PhD, fixed-term researchers with a number of PDRs identified was PhD but not classified as an academic probably an underestimate of the appointment, and contract researchers population. The primary position working outside of a university names of the selected population were: setting. The Australian context is Postdoctoral Fellow, Postdoctoral important here, in that within Australia Research Fellow, Research Associate, it is common for postdoctoral research Research Officer, Research Fellow, positions to be academic positions, Senior Research Associate.

Discipline In total, 2800 PDRs Engineering/Architecture/Building 85 (8.4%) were identified. PDRs Mathematics 34 (3.4%) were invited by e-mail Physical sciences 68 (6.7%) to participate in the Chemistry 80 (7.9%) survey by completing Earth sciences 57 (5.6%) an online Biology 257 (25.4%) questionnaire. Of the Information technology/Computing 22 (2.2%) original 2800, 428 did Agriculture/Vet/Environment 68 (6.7%) not have a valid e-mail Medical health 206 (20.4%) address, and 63 Social sciences 79 (7.8%) responded saying that Humanities 55 (5.4%) they did not fit the Total 1011 definition. Of the (100%) remaining 2309, 1011 Table 1: Research field and level of appointment of participating (43%) completed the PDRs questionnaire.

Responses by field of with academic employment rights and study are presented in Table 1. Not responsibilities. This is in contrast to surprisingly, there is a predominance the UK, where it is more usual for of science disciplines, particularly postdoctoral positions to be non- medical and earth sciences, but also a academic positions, representing non- representation of the humanities and academic employment, and North social sciences. America, where it is common for postdoc positions to be regarded as non-employment training positions, without employee rights. Despite 87

University type Identified PDRs Total responses Response rate* A: Group-of-eight 2051 (73.3%) 716 (70.8%) 42% B: post-1960 univ 522 (18.6%) 190 (18.8%) 40% C: ex-Colleges 62 (2.2%) 37 (3.7%) 65% D: Unitechs 165 (5.9) 68 (6.7%) 45% Total 2800 1011 43% Table 2: University type of participating PDRs *response rate is based on number of valid e-mail addresses, not on number of PDRs identified.

Australian universities are commonly Interviews with PDRs grouped into four institutional types. While all Australian universities A sample of 22 PDRs was interviewed, engage in both teaching and research, selected from those who completed a they differ in age, disciplinary foci, survey questionnaire. Each respondent research-intensiveness and associated selected had indicated on the prestige: questionnaire that they were willing to A. the oldest and most research- undertake a follow-up interview. The intensive universities (Group- interviews were semi-structured, of-Eight); consisting of a series of core questions, B. newer but still research- typically followed by additional intensive universities (post- questions seeking further information 1960 universities); in response to interviewees’ answers. C. previous teaching-oriented In general, the interviews were of colleges that were transformed approximately 60 minutes duration. into universities by government The interviews were taped and policy in the late 1980s (ex- transcribed verbatim, except in one colleges); and instance where the interviewee D. vocationally-oriented preferred not to be taped. The universities of technology transcripts were then content analysed, (unitechs). searching for common themes running across the interviews, as well as Table 2 shows the population and similarities and differences in response response rate from each university to the issues raised. The resulting type. The great majority of PDRs are themes arose from the data analysis, employed in group A or B universities, and were not predetermined. with 73.3% in group A alone. Nevertheless, PDR employment in Interviewees were all selected from group C and D universities has been group-of-eight universities, as these increasing in recent times, so it seemed host the vast majority (approx 75%) of important to capture the experiences of PDRs in the Australian higher that segment of the population. While education system. However, within there was a higher response rate from this parameter interviewees were then group C than from the other university- selected to represent, as far as possible, types, this was counterbalanced by the the variation in the postdoctoral fact that group C respondents population found in the questionnaire constituted only 3.7% of the total data, with an emphasis on exploring sample. Although all of the PDRs were the range of postdoctoral research currently employed in an Australian experience. The PDRs interviewed university, 30% had had their PhD varied along the dimensions outlined awarded in another country. below:

88

• Institution -- Located in five area. As a consequence, approximately Group A universities in four half of the supervisors who were different states of Australia. interviewed were currently supervising • Gender -- Male (12); female one or more of the PDRs interviewed. (10). Of the 18 supervisors invited to • Age -- 25–29 (3); 30–34 (6); participate, five did not respond to the 35–39 (7); 40–44 (1); 45–49 invitation, and four agreed to the (3); 50–54 (1); 55–60 (1). interview but were not available at a • Citizenship -- Australian (15); suitable time. This left a sample of 10 other (7). supervisors interviewed, varying along • Research field -- Mathematics the dimensions outlined below: (1); physics (2); chemistry (2); • Institution -- Interviewees were earth sciences (2); engineering selected from the same five (3); biological sciences (2); universities as the PDRs. agriculture (3); health sciences • Gender -- Male (7); female (3). (4); social sciences (1); • Research field -- Physics (1); humanities (2). chemistry (1); earth sciences • University where PhD awarded (1); engineering (2); agriculture -- Same university as postdoc (2); health sciences (2); position (6); other Australian humanities (1). university (7); overseas • Level of appointment – university (9). Professor (5, including 3 • Years since award of PhD -- 0- departmental heads); 1 (5); 2-4 (8); 5-10 (4); 11-15 Reader/Associate Professor (3); (4); 23 (1). other (2). • Term of current contract -- 12 • Experience as a PDR months (3); 2 years (4); 3 years supervisor -- From 1 PDR over (8); 4 years (3); 5 years (2); no 1 year to 30 PDRs over 20 response (2). years. Interview questions focused on what participants do as postdoctoral Findings related to the four researchers, what they think a assumptions postdoctoral researcher is, their career expectations and their experiences of Assumption 1 – PDRs want an career support. academic career.

Interviews with supervisors The first question to consider when thinking about postdoctoral positions The sample of supervisors interviewed as preparation for academic careers is was similarly selected to enable whether the holders of such positions exploration of variation in the range of actually want an academic career. views of postdoctoral research, and Here the issue can become confused by were asked similar questions to PDRs, the definition of an academic position. but from a supervisory perspective. As For instance, there is variation between there was no preceding survey of countries, and between institutions supervisors, and thus no collection of within countries, as to whether broad demographic parameters of postdoctoral positions are classified as supervisors, they were selected by academic – so some PDRs may already asking those PDRs interviewed to be regarded as having an academic suggest names of supervisors in their career. 89

In the survey of PDRs, respondents positions as "teaching positions" in were asked "Ideally, what type of contrast to the research positions that position would you like next?". they currently held. The shift from a Although, 73% of PDRs said that they postdoctoral research position to a wanted an academic position (see traditional academic position was seen Table 3), when their response was as representing a shift from research to limited to academic teaching and teaching, with the associated teaching research positions only (i.e., with workload seen as impinging heavily on academic research-only positions time to do research. excluded), the percentage reduced to 41% -- less than half of the sample. In Ideally, I would like to contrast, those wanting a research-only continue researching, but career – either university research am aware that a full (32%), or Government or industry career path for a research (16%) – amounted to 48% of researcher is not the sample, with 11% interested in an available in the social ‘other’ career, such as consultancy, etc. sciences and humanities in this country. For that It is clear that continuing with a reason, a position that research-only career is highly combines teaching and attractive to many PDRs. This result researching is the long- should not be surprising, in the sense term aim (anonymous that most PDRs will have completed PDR survey respondent). many years of intensely research- oriented activity, starting with their Assumption 2 – Postdoctoral research doctoral work and followed by 1-3 positions provide a stepping stone to positions as a PDR by the time they academic careers. seek more permanent work. This sort As shown in Figure 1, based on the PDR survey, a substantial number of Ideal position N % Academic -- teaching and research 408 41.1 Academic -- research only (incl PDR) 320 32.2 Government research agency or Institute 76 7.7 Industry – research 78 7.9 Industry – non-research 18 1.8 Public service 2 0.2 Self-employed 14 1.4 Other 13 1.3 Undecided 64 6.4 Total 993 100 Table 3: PDRs ideal next position of preparation for academic work the PDRs in this sample were still in a inevitably selects for those who enjoy contract research position six or more research-based work. years after the award of their Ph.D., with a mean period of 5.4 years. In line with this, during the interviews Similarly, Figure 2 shows many PDRs with PDRs they frequently described in their fourth and fifth contract academic teaching and research 90 position following their Ph.D., with a mean of 2.7 positions.

Figure 1: Years since award of PhD Figure 3: Ratings of long term job prospects

Based on the interviews with PDRs, viewing their position as a stepping stone towards a permanent position was only one perception of the nature of contract research positions. PDRs alternatively described contract positions as (a) being their career, with

ongoing contract research positions the nature of their career, or as (b) an opportunity to engage in research, with Figure 2: No of contract positions following career implications of lesser PhD significance. (These different perspectives on the nature of PDR Asked to rate their perceptions of their positions are described in more detail long-term job prospects (in any area of in Åkerlind, 2005). Such researchers work) on a five-point likert scale seem motivated to undertake (where 1 indicated ‘very poor’ postdoctoral research by the nature of prospects, 3 ‘average’, and 5 ‘very the research question itself, and the good’ prospects), only 45% saw their opportunity to investigate this job prospects as good or very good, question, irrespective of career and 23% as poor or very poor (see implications. For example, Figure 3), with an overall mean of 3.3. Interestingly, as shown in the original It [a PDR position] means survey report (Thompson et al, 2001, the chance to focus on pp. 55-56), there were no significant some of my own research. differences in the perceived prospects And it means a job, of those from more prestigious, because I'm not likely to research intensive universities than have a career after it... I those from less prestigious universities, don't see what comes next. or those appointed on more prestigious Theoretically, they [PDR Fellowship positions than those on less positions] are for you to prestigious grant-funded positions. lead into academia or an academic appointment somewhere, but I don't see how at this point (PDR 91

interviewee from the postdoc is someone who social sciences). has finished their doctorate and then Assumption 3 – Postdoctoral research worked in that area. So, positions provide an opportunity for when do you become not a novice researchers to become postdoc? (supervisor increasingly independent interviewee from Agriculture). The majority of PDRs in the survey sample did not see their position as a Also, PDRs on fellowships are already training position, nor did they see operating independently, in the sense themselves as being supervised (Table that they have independently designed 4). their research project and secured the

Training Not No Independent Supervised No Training Resp Resp 269 698 44 623 276 12 26.6% 69.0% 4.4% 61.6% 27.3% 11.1% Table 4: Ratings of PDR positions as training/not training and independent/supervised

Many PDRs did not see their position funding of their salary through as a training position because they felt applications for research funding. that they were already operating as independent researchers. Perhaps this Independence also varies with is not surprising, given the increasing disciplinary area. In many areas, periods of time spent in postdoctoral science disciplines in particular, a positions, with most PDRs on their period of postdoctoral research is a third or later appointment. virtual requirement in order to achieve academic employment. However, in I guess in medical humanities and social science research, you do a PhD disciplines, academic appointment and then you do a postdoc immediately post-PhD is more the position, and usually do a norm. series of them... I know people of 55-60 who are I would emphasise that in still in postdoc research my discipline positions (PDR (philosophy), postdocs are interviewee from not considered training medicine). positions. I moved from a being a [higher salaried Again, a postdoc position teaching and research] is a very broad term. … I academic to a [lower always wonder what a salaried research only postdoc is, and how do position] in order to take you tell that someone is a up this postdoc, partly postdoc?... Because a because the opportunity 92

for full time research is am concerned… I give very attractive. Given my them 20% freedom to do previous experience, my anything that they like. approach to my But that is me. It doesn’t postdoctoral work is that mean that that is normal of undertaking a large practice (supervisor research project, rather interviewee from than ‘training’ to do Engineering). research (anonymous PDR survey respondent). Assumption 4 – Postdoctoral research positions provide an opportunity to Another reason given by PDRs for not concentrate solely on research seeing their position as a training position related to a sense of being Well, it [a PDR position] exploited, with little time allowed for is a luxurious, beautiful developmental activities within their entity, where one is able appointment. For example, to, by and large, focus on a specific research What training? As far as I project. So that most of am aware postdocs come your field of view is straight out of their PhD, occupied by the single attempt to generate task of performing papers and a career, and research in a specified are not trained any area (supervisor further. In fact, there is interviewee from not time for such training Engineering). because most of the time is spent on research in the During interviews, the PDR lab (anonymous PDR supervisors were asked to describe the survey respondent). role played by PDRs within their department. While supervisors from Some supervisors also acknowledged all disciplines described PDRs as the variability in developmental ‘enriching the intellectual life of their opportunities between different PDR department’, and supervisors from positions. For example, laboratory-based disciplines emphasised the role played by PDRs as It depends very much on day-to-day managers of the research the supervisor how much lab and in informally guiding and freedom the supervisor supervising research students and more gives in running a lab and junior PDRs, other roles were rarely whether the supervisor mentioned. treats the postdoc as an equal colleague as In contrast, the online survey explicitly opposed to just a slave … investigated duties carried out by the majority of postdocs, PDRs in addition to their research. they are tied to a Respondents were asked about their particular grant and involvement in four areas: supervision project and that is their (formally, as a member of the job. However, I always supervisory panel, and informally, in tell them that as far as I terms of day-to-day training and 93 advice), lecturing, tutoring/demonstrating and conference organisation. Respondents were also Figure 4: Proportion of time spent on asked to list any other duties that they additional duties undertook, and to indicate the percentage of their time spent on these As shown in Figure 4, only 127 duties. Table 5 shows that just over respondents (13%), did not report 50% reported being involved in formal engaging in additional duties, and for supervision of research students, just over 50% of the sample such almost 40% in lecturing, and a quarter additional duties were estimated as in tutoring/demonstrating and taking up more than 20% of their time. conference organisation. Perhaps even more surprisingly is the existence of some postdocs who Additional Duty report spending the majority of their Formal Supervision (N = 52.6% time on ‘additional’ duties. In the 1,001) interviews, some PDRs reported acting Informal Supervision (N 78.7% as more of a supervisor's aide (i.e. = 988) undertaking Lecturing (N = 997) 39.0% any duties requested of them by their Tutoring/Demonstrating 24.8% supervisor) than a researcher, which (N = 976) may explain this unexpected finding. Conference Organisation 25.5% (N = 977)

Table 5: Additional duties performed by Discussion PDRs – fixed choice response It has often been said that the intensive research nature of a PDR position provides poor preparation for the range of activities that typically constitute an academic position, particularly due to Table 6 shows the large range of other the absence of preparation for duties also reported, though less teaching. However, this argument is frequently than the duties in Table 5. normally a skills-based one, that is, that PDR positions are often not providing training in teaching and other academic skills. While this point is valid, an additional point that has emerged from this research is that the intensive research nature of PDR

94 positions may not be a good way to substantial variation between different select for people who are interested in PDR positions. . a broader range of academic activities. Indeed, 48% of the PDR sample Another implication of the large regarded a research-only position number of PDRs engaged in non- (whether or not in academia) as their research activities is that, while the contribution made by PDRs to research productivity within the higher Additional duty No. of education sector is well-known, they respondents are making a much more substantial reporting contribution to teaching and University administration 252 supervision across the sector than is Group or project manager 129 typically recognised. Grant applications 87 Committee work 69 Another issue highlighted here is that Staff or group 56 some PDRs do not see contract supervision research as an interim stage to a Reviewer/editor/examiner 43 permanent position. Given their External administration 28 commitment to research, some have Conference/seminar 23 the intention of continuing to preparation undertake ongoing contract research Consultancies 23 positions indefinitely, as the only way Clinical duties 4 in which they can maintain a research Table 6: Additional duties performed by PDRs career, despite the obvious insecurity – open choice response and other disadvantages. Others choose to take a PDR position in order ideal employment, with only 41% to focus on research for as long as they preferring a teaching and research can, even though they do not believe position. that ongoing contract research positions will be available long-term. At the same time, a surprising number of PDRs are engaged in significant Over the last decade in particular, PhD amounts of non-research duties. Based awards and subsequent PDR positions on the sample described here, over have risen at a much faster rate than 50% are involved in formal academic positions. The knock-on supervision of research students, effect of this, in terms of a reducing almost 40% in lecturing, and a quarter proportion of PhDs and PDRs finding in tutoring/demonstrating and also academic employment, is well-known. conference organisation. Furthermore, Under the current situation of intense for over 50% of the sample, such non- competition for a limited number of research duties were estimated as academic positions, one would expect taking up more than 20% of their time. that PDRs in more prestigious This indicates that PDRs may universities and appointed on more commonly receive a higher degree of prestigious research fellowships would experience in a breadth of academic- have a decided advantage. It is related duties than is usually interesting then that there was no recognised. Unfortunately, the significant difference in perceptions of opportunities for such experience seem their long-term job prospects amongst to be more commonly available in an the PDRs in these more prestigious ad hoc than systematic way, with positions. Of course, perceptions of 95 job prospects and actual job prospects Australian Department of Education, need not coincide, but at the least these Training and Youth Affairs and by the findings indicate widespread Centre for Educational Development uncertainty amongst PDRs with and Academic Methods of the respect to their career prospects. Australian National University. This paper draws on the original EIP report Perhaps the most surprising outcome (Thomson et al., 2001), however, the of this study is the high proportion of conclusions are those of the present PDRs who consider that they are author. A modified version of this already engaged in independent paper has previously been presented at research and do not see themselves as the 2007 annual conference of the being in a training position. This may Society for Research in Higher be due to both the large number who Education, Brighton UK. are in their third or later postdoctoral position, and thus who have References accumulated extensive experience as a Åkerlind, G.S. (2005) PDRs: Roles, Functions researcher, and to the variable nature and career prospects, Higher Education Research and Development, 24, pp. 21-40. of the development and training opportunities available across Borthwick, J. and Wissler, R. (2003) postdoctoral positions. Either way, if Postgraduate Research Students and Generic contract research positions are not Capabilities: Online Directions, Research training positions, then it is hard to Evaluation Programme, Higher Education Group, DEST, Commonwealth of Australia: regard them as anything other than Canberra. exploitative in nature since they lack permanency. European Commission (2005) The European Charter for Researchers and The Code of The key implications of this research Conduct for the Recruitment of Researchers. EUR21620www.europa.eu.int/eracareers/euro for preparation of PDRs for academic peancharter careers lies in highlighting the hidden variation in the nature of PDR HM Treasury (2000) SET for Success, Roberts positions. The research duties report undertaken by PDRs, the amount of www.hmtreasury.gov.uk/Documents/Enterpris e_and_Productivity/Researchand supervision and training received, Enterprise/ent_res_roberts.cfm. opportunities and obligations to engage in non-research duties, motivations of Helbing, C. C., Verhoef, M. J., & Wellington, PDRs, and future career aspirations are C. L. (1998). Finding identity and voice: A highly variable in a way that cannot be national survey of Canadian postdoctoral fellows. Research Evaluation, 7(1), 53–60. predicted by the position title, institutional setting, funding type or Marceau, J., & Preston, H. (1996). Taking the discipline of PDRs. This makes a one lead: The ARC Fellowship Scheme in size fits all approach to PDR Australia. Canberra: Australian Government development unlikely to be effective, Publishing Service. and emphasises the need for provisions Nerad, M., & Cerny, J. (1999). Postdoctoral that are as varied and flexible as patterns, career advancement and problems. possible. Science, 285, 1533–1535.

Acknowledgments Research Councils UK (2001) Joint Statement of Skills Training Requirements of Research The research reported here was funded Postgraduates-- by the Evaluations and Investigation http://www.grad.ac.uk/cms/ShowPage/Home_ Programme (EIP) of the then 96 page/Policy/National_policy/Research_Counci ls_training_requirements/p!eaLXeFl

Science journal special issue on postdoctoral researchers, 285(3), September 1999.

Thompson J., Pearson, M., Åkerlind, G., Hooper, J., & Mazur, N. (2001). Postdoctoral training and employment outcomes, EIP Report 01/10. Canberra: Higher Education Division, DETYA. 97