POTENTIAL EMIGRATION OF SCIENTISTS FROM SLOVENIA IN THE MID 1990s1 Milena Bevc 1. INTRODUCTION Emigration is in most countries as a rule a poorly registered phenomenon, especially emigration of the highly educated - brain drain. In Slovenia, a small upper-middle-income country, the statistical basis (register of population, censuses) in this field is much better than in many other former socialist countries as well as in comparison to some developed countries. But despite this fact, on the basis of these sources no uniform conclusion is possible regarding the extent of total emigration and emigration of the highly educated from the country since the late 1980s when the great political and economic changes started (for the 80s as a whole there are signs that the propensity of those with higher education for emigration increased in comparison to the previous decade and was higher in comparison to other educational categories of population).2 Even more rare is the registration of external migration in different countries on the basis of the same methodology - that is the existence of the internationally comparable data on this migration. Furthermore, there is no statistical basis for the investigation of potential emigration. A few years ago, the European Commission initiated the international research project on brain drain (of researchers) from former socialist countries (COST A2 project: Europe’s Integration and the Labour Force Brain Drain). 1 This paper was presented in the International Conference on Applied Statistics, organised by the Slovene Institute of Social Sciences and Statistical Society of Slovenia in 1997 (Preddvor, September, 15-17, 1998). 2 Source: Bevc, Logar, 1992; Bevc, Malačič, 1995. Dve domovini / Two Homelands — 9 — 1998, 167—189 Different dimensions of this brain drain were investigated - external (abroad) and internal (to non-scientific sectors) on the one hand, and real (in period 1988— 94) and potential (probable in the 90s) on the other. Slovenia was included in this project together with the following nine countries in transition: Bulgaria, the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania and the Slovak Republic. In this paper we use the term »region« when all the ten countries mentioned are considered as one group. The main purpose of the project was to estimate the extent, the reasons, motives and characteristics of the outflow of researchers abroad and to non-scientific sphere within the country. The main source of data in this project were surveys conducted in the science sector (surveying of researchers and institutions with researchers). No similar project exists for developed countries. The work of the Slovene research group was cofinanced by the Slovene Ministry of Labour, Family and Social Affairs. The results are presented in four research reports.3 This paper depicts the methodology and results of analysing the potential emigration of Slovene researchers in the mid 90s within the mentioned international research project. It also includes the comparison of the »population«, sample and respondents. As a »potential emigrant« we defined a researcher who wishes, intends or would under special conditions go abroad for more than one year. 2. METHODOLOGY Sample - some main characteristics and the method of selection of surveyed persons. - We surveyed 1012 researchers with master’s and doctor’s degrees. The basis for the selection of the surveyed persons was the population of 3542 researchers (with a doctor’s or master’s degree), registered at the Slovene Ministry of Science and Technology. We received the list with names and addresses of all these persons from the above-mentioned Ministry. The data on their education (level and field) and age could not be obtained by reason of the security of personal data; for this reason we could not exclude older researchers. As regards the data the Ministry demanded that the survey should be totally anonymous. Random sampling was used since we had data on names and addresses at our disposal (in other countries systematic sampling was used). The process of selection of the surveyed persons was the following: The (220) institutions/firms 3 Bevc, Malačič, 1995; Bevc, et.al., 1996; Bevc, 1996; Malačič, 1996. with researchers (with doctor’s and master’s degrees) registered at the Ministry of Science and Technology were classified alphabetically and the researchers within each institution/firm were also classified in the same way. For the purpose of getting the sample of approximately 1000 scientists each third to fourth (3.5) researcher was selected (3500/1000 = 3.5) starting at the beginning of the list. For that reason some institutions/firms with fewer than 3 researchers were excluded from the sample. All scientists who informed us (by phone or by post) that they would not fill in the questionnaire for different reasons were replaced by others from the list of 3542 researchers. Since the questionnaire had to be anonymous, it was sent to each individual participant (at the beginning of June 1995) separately by post to the address of his/her employment. The anonymity in returning the questionnaires was achieved by enclosing an envelope to each questionnaire addressed to the Institute for Economic Research. To assure as great response as possible we did the following: We promised to give the results of the survey and the project to those who expressed their interest in them (by phone or by post). The letter from the Slovene Minister of Science and Technology was enclosed with each questionnaire (next to the letter from the Institute for Economic Research). After two weeks the first and after the following two weeks the second reminder was sent to each surveyed person (with an exception of those for whom we knew in one way or another that they returned the answered questionnaire). Different signs led to the conclusion that all of these efforts had an important impact on the response. There were lots of telephone and written correspondence with the surveyed researchers. Many of them expressed the wish to get the results after finishing the project. They got its summarised version. By the deadline (the same for all countries included in the above-mentioned international project) 648 questionnaires were returned, that is 64% of the total number of those to which the questionnaire was sent and 19% of the total »population« which represented the basis for the selection of the sample. The response could have been even greater if the process of surveying had started two or more weeks earlier. Some main problems in the process of surveying and in the analysis of questionnaires. - We can summarise these problems as follows: 1. The Ministry’s list of researchers was not entirely up to date as regards some institutions, since some of the addressed persons were already retired or dead. 2. Since we did not have the data on the age of the surveyed persons, the questionnaire was sent to many older researchers (just before retirement), too. 3. For the question on research field (discipline) the original uniform code book (used in all 10 countries) was not enclosed to the questionnaire which had some advantages and disadvantages: - Advantages: This code book would most probably dissuade some surveyed persons from filling in the questionnaire, since the classification included in the code book differs in some aspects from some other internationally or nationally valid classifications. - Disadvantages: Since some surveyed persons defined their scientific discipline too broadly or too narrowly, we had a lot of problems correcting the grouping of such examples. In the process of coding the answers to this question a lot of scientists from different disciplines helped us. Methods used in the analysis of the data. - The factors of potential external (and internal) brain drain were obtained by the statistical analysis (SPSS) using the Spearman and Cramer coefficients. The methodology was mainly the same in all countries included in the international project. 3. THE COMPARISON OF THE SAMPLE, THE POPULATION AND THE PART OF IT THAT RESPONDED TO THE QUESTIONNAIRE The comparison includes the scope, age, gender and education (data on age and education were obtained from the Ministry of Science and Technology after the surveying). The comparison of the sample and the population (of scientists): The institutions/firms by the scope of the »scientific basis« and their inclusion in the sample. - Since one of the requests in the process of selection of the surveyed persons was to include the scientists from all the biggest scien- tiflc institutions, the following explanation is important: In Slovenia in 1995 the majority (57%) of scientists with master’s and doctor’s degrees were concentrated in 17 biggest scientific institutions with 50 and more such scientists per institution (among them there were 13 institutions of higher education). The distribution of the rest of the researchers with master’s and doctor’s degrees (registered at the Ministry of Science and Technology) was the following (in 1995 -see Table 1): 27% in (35) institutions with 15 to 49 such scientists (among them a half of these institutions presented the institutions of higher education), 10% in (47) institutions with 5 to 14 such scientists (among these institutions the majority presented state research institutes), and 6% in 121 different institutions/enterprises with fewer than 5 mentioned researchers per »unit«. For the sake of the method of selection of the surveyed persons (each third to fourth researcher within particular organisations which were ranked by alphabetical order - similarly as researchers) the same proportion of scientists with master’s and doctor’s degrees from all types of institutions/firms was included in the sample (29% - see Table 2). The only exception were institutions/firms with fewer than 3 such scientists. The comparison of some structural characteristics. - The structure of the sample and the population by the level of education (master’s and doctor’s degrees) and gender were almost the same, and the structure by the age was very similar as well (Table 3). The comparison of the surveyed persons who responded to the questionnaire with the sample. - 64% of scientists included in the survey returned the answered questionnaires as it has already been mentioned. The structure of the sample and that of participants who answered the questionnaire were almost the same by the level of education and gender, and the structure by the age differed in the direction expected. On average, those who responded were slightly younger than all scientists included in the sample. The response to the questionnaire was the greatest among the youngest and it decreases with the age of participants. The comparison of the population with those scientists who responded to the questionnaire. - As a result of the size of the sample (29%) and the willingness of the surveyed persons to collaborate in the survey (64% response) we got the answers to the questionnaire from a very large part of all the Slovene scientists (almost one fifth - 19%). The structure of those scientists who filled in the questionnaire and that of the whole population of scientists are almost the same as regards the level of education and the gender. The structure by the age differs in the following: on average the total population is slightly older than those scientists who answered the questionnaire. Table 1: »Population« - institutions/enterprises by the average and total number of researchers with master’s and doctor’s degree - May 1995 Number of researchers with M.A., M.Sc., Ph.D. Number of institutions/ enterprises with researchers observed Structure (%) Total number of researchers with M.A., M.Sc.,Ph.D. Structure (%) - less than 5 121 55.0 211 5.9 - from 5 to 14 47 21.4 362 10.2 - from 15 to 29 23 10.4 486 13.7 - from 30 to 49 12 5.5 474 13.4 - from 50 to 99 10 4.5 785 22.2 - 100 and more 7 3.2 1224 34.6 Altogether 220 100 3542 100 Calculated on the basis of data of the Slovene Ministry of Science and Technology. Table 2: The share of »population« in the sample (%) POPULATION SAMPLE % OF POPULATION IN THE SAMPLE Number of researchers with M.A., M.Sc., Ph.D Number of institutions/ enterprises (indirectly through researchers) Number of researchers in the sample Institutions/ enterprises Researchers - less than 5 59 61 49 28.9 - from 5 to 14 47 102 100 28.2 - from 15 to 29 23 140 100 28.8 - from 30 to 49 12 134 100 28.3 - from 50 to 99 10 224 100 28.5 - 100 and more 7 351 100 28.7 Altogether 158 1012 72 28.6 Calculated on the basis of data of the Slovene Ministry of Science and Technology and 648 questionnaires. Table 3: The comparison of researchers who answered the questionnaire to the sample and the »population« by some structural characteristics (%) Population Sample Researchers, who answered the questionnaire Education 100 100 100 -Ph.D. 54.6 55.9 56.6 - M.A., M.Sc. 45.4 44.1 43.4 Sex 100 100 100 - female 28.5 28.2 28.3 - male 71.5 71.8 71.7 Year of birth 100 100 100 - 1965 or later 5.5 4.8 5.9 - from 1955 to 1964 35.6 34.7 37.5 - from 1945 to 1954 29.7 33.2 33.7 - from 1935 to 1944 19.5 19.0 16.2 - 1934 or earlier - 9.7 8.3 6.7 Calculated on the basis of data of the Slovene Ministry of Science and Technology and 648 questionnaires. 4. SOME MAIN CHARACTERISTICS OF SURVEYED RESEARCHERS IN COMPARISON TO OTHER COUNTRIES IN TRANSITION Through the questionnaire, the following dimensions of scientists were observed: demographic characteristics, professional characteristics, work conditions, hierarchy of values and the estimation of their achievement in the next five years under different circumstances, demand for scientific work within the country, economic situation, professional contacts with other countries (previous, current, planned future). Some main conclusions for Slovenia in comparison to other countries in transition that were observed are: Demographic characteristics: higher percentage of men (72% to 64% for the »region«). Professional characteristics: the structure by the scientific field was the following: 47% from natural, medical and biotechnical sciences, 29% from technical and 23% from social sciences and humanities. The first group was larger and the latter two smaller than in the »region«. The majority of scientists (54%) were employed in institutions of higher education (more than in the »region«). 40% of the surveyed scientists had worked in the field of science up to 10 years (region - 30%), and 60% of them for a longer period (region -70%); 38% of the surveyed persons had a leading position in the institution of their employment (region - 35%). Work conditions: on average better than in the majority of the other countries observed. Hierarchy of values and their achievement in the next five years in different circumstances (continuation of scientific work within the country, change of the field of activity within the country, continuation of the work abroad): Among the most important values (»very important«) the Slovene scientists listed the following: professional fulfilment, topical scientific information, availability of key publications and good research infrastructure. The comparison with the average for the »region« is the following: a higher share of the respondents thought they would most probably achieve the majority of values (there are two exceptions: good research infrastructure and financial prosperity) in the next five years to the highest degree by the continuation of their scientific work within the country. Economic situation: on average better than in the »region«; they estimated their financial situation in the time of surveying to be somewhere between »we can survive« and »we can save some money from our income«. 5. POTENTIAL EMIGRATION OF SLOVENE RESEARCHERS 5.1. THE EXTENT OF THE PHENOMENON The structure of the respondents by the probability of going abroad for the period of more than 1 year. - In the context of joint methodology (within the international project) we defined the scientists as »sure« migrants, »sure nonmigrants« and others who are somewhere between both categories (»hesitant« -less probable migrants) on the basis of their answers to two broader questions: What would you do if you received an offer for going abroad in the course of the next few months (fellowship for more than 1 year, research work for more than 1 year, non-research work for more than 1 year, other); possible answers were: I would accept without hesitation, I would accept under certain circumstances, I would accept, but I would try to postpone it for some time, I would decline the offer, I do not know). This question refers to the probability of the accepting the offer to go abroad for more than 1 year. Do you intend to leave for the foreign country for more than 1 year; possible answers were: yes, I am arranging the departure; yes, at the moment I intend to leave but I have not undertaken any specific steps; yes, but not now; no, I do not intend to leave the country. This is the question about the level of concretisation of an intention of going abroad for more than 1 year. The above-mentioned probability of going abroad for more than 1 year reveals the following structure of the Slovene respondents: sure migrants 7%, hesitants 69%, sure non-migrants 24%. The structure is similar to the structure of 7300 surveyed scientists from all 10 countries altogether (10%, 68%, 22%). Among the Slovene respondents about 76% were potential emigrants, but most of them were hesitants (see Figure 1). Figure 1 The structure of the surv eyed researchers by the probability of going abroad for more than 1 year (%) 90% 21.4 21,3 28.6 37.4 80% 70% 60% 50% 67.6 69.3 64.7 40% 30% 20% 21,4 10% 11.7 4.7 0% M o Cl M ■o u, s 'c H Sure nonmigrants □ Hesitants □ Sure migrants______ The structure ofpotential (e)migrants regarding the intended period of staying abroad4 (see Figures 2, 3 and 4 and Table 4). - It is possible to distinguish the following three groups: short-term (1-3 years) emigrants: 75% (sure 20%, hesitants 55%); in all other countries observed except Bulgaria the share of these emigrants is higher; 4 Only a part of potential emigrants is considered (36% of all potential Slovene emigrants) since only a part of potential emigrants was »expected« to answer the question of the intended duration of being abroad. medium-term (4-5 years) emigrants: 10% (sure 1%, hesitants 9%); a higher share than in all other countries except Bulgaria, Romania and Lithuania; long-term (more than 5 years) emigrants: 15% (sure 6%, hesitants 9%); a higher share than in all other countries except Bulgaria. The propensity of the surveyed Slovene scientists to medium-term and long-term emigration was in the mid 90s higher than in the »region«, which is mainly the consequence of the structure of less probable emigrants (hesitants). Despite the fact that the majority of potential emigrants present less possible (hesitant) short-term emigrants (42% of all respondents), the extent of potential long-term and medium-term emigration (both presenting 25% of total potential emigration and 19% of all respondents) is also relatively high. The percentage of very probable (sure) medium-term and long-term emigrants presents 7% of potential emigrants and 5% of all respondents. The structure of potential emigrants could be (due to the method used for surveying - random sampling — and the similarity of the population and the respondents by many dimensions) generalised to the total »population« of researchers with master’s and doctor’s degrees. If so estimated extent of sure long-term emigration of scientists was approximately close to the reality, the extent of potential emigration in the mid 90s would be much larger than was the real emigration in the period 1988— 94 (in this period: it accounted to less than 0.5% of the »population« per year). F iglire 2! Tl* structure of the |»tential emigrants by Hie intended duration of being abroad (%) KD% m ED% m 60% m