Predictors of unemployment in refugees
Main Article Content
Refugees have an increased risk of unemployment. We initially examined 83 unemployed refugees in Austria (mean age 34.01 years, s = 10.04), 42 of whom attended vocational training courses and 41 of whom did not. At the time of assessment some months later, 37 participants (44.6%) were employed and 46 (55.4%) were unemployed. We used sociodemographic variables as well as collectivistic values, clinical symptoms, command of the German language, perceived social support, and self-efficacy as possible predictors of unemployment. Logistic regression revealed that attending a training course, previous work experience, and a shorter time in Austria increased chances of employment. The logistic regression model we presented will enable practitioners to predict chances of employment in individual cases and to tailor refugees’ vocational rehabilitation to their personal needs. The results suggest that vocational rehabilitation for refugees should be provided as soon as possible after asylum has been granted.
A high percentage of refugees depend on social security in Austria. Once granted asylum, unemployed refugees can receive welfare support for an unlimited period. Welfare authorities may assign refugees to training courses that teach a basic knowledge of German and job-finding skills.
Hansen and Lofstrom (2006) have shown that refugees contribute more than do other immigrants to the “immigrant-native welfare gap” (p. 23) in Sweden. In a Norwegian longitudinal study, Ekhaugen (2005) found that 63% of refugees still depended on social benefits in the third year of their stay and 55% of refugees still depended on social benefits in the eighth year of their stay. Jäckering (2007) reported similar results for Germany.
Poor knowledge of the host country’s language and lack of formal qualifications can be risk factors for unemployment (Kogan, 2003). Accordingly, Beiser and Hou (2001) pointed to the importance of command of the host country’s language for finding work. Ward, Bochner, and Furnham (2006) indicated that language proficiency, a higher educational level, perceived social support, and a younger age increased refugees’ chances of vocational integration. Ward et al. also emphasized that psychological symptoms, especially posttraumatic stress can be enhanced by unemployment and, in turn, reduce a refugee’s chances of finding work. According to Faelli and Carless (1999), a longer time in Austria and a higher degree of perceived self-efficacy improved job prospects in a sample of immigrants from India to Austria.
Women are especially disadvantaged among refugees, as they are frequently denied access to the labor market and higher education in their countries of origin (Bloch, Galvin, & Harrell-Bond, 2000). Thus, as many women remain in their traditional roles as mothers and housewives, they suffer from social isolation (Bloch et al., 2000). However, women may also be more successful than men at overcoming initial vocational difficulties (Kofman, Phizacklea, Raghuram, & Sales, 2000).
Collectivist attitudes may also be a risk factor for unemployment, because collectivism mitigates the psychological consequences of being unemployed (Martella & Maass, 2000), and decreases the readiness to relocate to attain a job (Otto & Dalbert, 2012).
Our aim in the present study was to integrate the results reported in the literature into a more complex statistical model that would allow practitioners to predict an individual refugee’s risk of continued unemployment. In accordance with the above findings, we expected a lower degree of collectivist values, a lower degree of clinical – and especially posttraumatic – stress symptoms, a younger age, a higher level of formal education, greater knowledge of the host country’s language, a higher degree of perceived social support, a higher degree of self-efficacy, and a longer time in the country, to predict employment in a sample of jobless refugees in Austria. In addition, a number of sociodemographic variables (e.g., gender or country of origin) were used as possible predictors and we expected that attending training courses would improve refugees’ chances of finding work.
Method
Participants
We recruited 83 unemployed legal refugees in Austria (41 men, 34 women, and eight participants who did not state their gender) with the help of welfare officers and training course providers. Among this group 42 were attending training courses (with varying content, structure, and duration) for language and job-finding skills over several months and at three different sites. The remaining 41 did not attend such courses, but received social welfare. There was no significant difference between men and women attending courses. The participants’ religion was diverse. In terms of country of origin, 34 came from Chechnya, 16 from Afghanistan, 10 from Africa, 10 from former communist countries (apart from Chechnya), and 13 from other countries. The mean age was 34.01 years (s = 10.04, range 17 to 59 years) and the average time in Austria was 7.01 years (s = 6.68, range 1 to 32 years). The average work experience in Austria amounted to 21.60 months (s = 54.36, range 0 to 264 months), including 45 participants who had never worked in Austria.
For the participants who were attending the training courses, predictor variables were assessed as part of the courses and data on employment versus unemployment were collected at the end of the courses. The vocational status of participants not attending a training course was assessed by social welfare officers approximately six months after they had completed the questionnaires.
Measures
Collectivist values were assessed using the Cultural Orientation Scale, Part II (COS; Bierbrauer, Meyer, & Wolfradt, 1994). Clinical symptoms of anxiety, depression, and somatization were assessed using the Hopkins Symptom Checklist-25 (HSCL-25; Mollica, Wyshak, de Marneffe, Khuon, & Lavelle, 1987). Posttraumatic symptoms were assessed using Part IV of the Harvard Trauma Questionnaire (HTQ; Mollica et al., 1992). Language command was assessed using a multiple-choice test of German vocabulary (Lehrl, 1977). Perceived social support was measured using a slightly modified (to meet the requirements of a refugee sample) 18-item questionnaire suggested by Ong and Ward (2005). We assessed self-efficacy with respect to interpersonal abilities in regard to finding work using the Assertive Job-Hunting Survey (AJHS; Becker, 1980) and self-efficacy or self-esteem (SE) with respect to task-oriented abilities using a 10-item questionnaire developed by Ellis and Taylor (1983). All instruments were reported to have good to excellent reliability and validity. Russian, Farsi, Turkish, and English versions of the questionnaires had been prepared.
Results
At the time of assessment, 37 participants (44.6%) were employed and 46 (55.4%) were unemployed. On the COS, ranging from 1 = very bad to 7 = very good, the mean score was 4.92 (s = .58, range 2.85 to 6.10), suggesting an average moderate collectivist orientation. The mean score on the HSCL-25, ranging from 0 = not at all to 3 = extremely, was .88 (s = .70, range .00 to 2.57) and on the HTQ, also ranging from 0 = not at all to 3 = extremely, the mean score was 1.15 (s = 0.76, range 0 to 3). A maximum of 13 correct answers was possible in the German vocabulary test, where the mean score was 7.47 (s = 3.75, range 0 to 13). On the measure of social support ranging from 0 = no one to 4 = many people, the mean score was 1.70 (s = 1.04, range 0 to 4). On the AJHS, measuring SE with respect to job-hunting ability and ranging from 0 = very unlikely to 5 = very likely, the mean score amounted to 2.99 (s = .76, range 1.50 to 5.00) and on the measure of task-oriented SE, ranging from 0 = does not apply at all to 4 = applies extremely, the mean score was 2.11 (s = .62, range 1.10 to 3.70). Symptom measures correlated moderately and negatively with social support and command of German, were correlated highly to each other, and uncorrelated with age, COS, AJHS, and SE scores. Command of German correlated positively with SE scores, age, and time in Austria. Men and women did not differ significantly on any of the predictor variables.
The sociodemographic variables (i.e., attending a training course or not, age, gender, time in Austria, country of origin, years of schooling, and duration of work experience in Austria) were first entered into a binary logistic regression model (backward stepwise method) with employment versus unemployment as the dependent variable. The variables of training course, time in Austria, and duration of work experience significantly contributed to the prediction. Next, the scores obtained from the psychometric questionnaires were used as possible predictors of the dependent variable, again using the backward stepwise method of binary logistic regression. In this case, no significant predictors of outcome were identified. Thus, to attain a final predictive model, the variables of training course, time in Austria, and duration of work experience were entered into a binary logistic regression using the enter method, the results of which are shown in Table 1.
This model classified 75.6% of unemployed and 63.9% of employed individuals correctly. Overall, 70.1% of the refugee group were identified correctly and Nagelkerke’s R2 = .278, pointing to moderate predictive power. The results indicated that attending a training course (vs. staying on social security without attending a training course) and reporting a longer duration of previous work experience in Austria increased the chances of finding work, whereas an overall longer time in Austria decreased the probability. It is seen from Table 1 that an individual’s probability of finding employment can be estimated using the formula p = 1 / (1 + e-z), where e = 2.718 (Euler’s number) and z = 1.372 * x1 -.292 * x2 + .031 * x3 + .544 (Bühl, 2011). For participants in a training course x1 = 1, for those not participating x11 = 0; x2 is the individual’s number of years in Austria and x3 is the number of months he or she has already worked in Austria. For example, for a person not attending a course (x1 = 0), who had stayed in Austria for five years (x2 = 5), and who had worked in Austria for three months (x3 = 3), the probability of finding work within a maximum of six months amounts to p = 1 / (1 + 2.718 - (1.372*0 – .292 * 5 + .031 * 3 + .544)) = .31. Whereas for an individual who attends a training course (x1 = 1), who has stayed in Austria for two years (x2 = 2), and who has worked in Austria for eight months (x3 = 8) this probability is p = 1 / (1 + 2.718 - (1.372 * 1 – .292 * 2 + .031 * 8 + .544)) = .83.
Table 1. Binary Logistic Regression Variables
Discussion
Attending a training course was found to improve the participants’ chances of finding work. This result can be interpreted as a direct result of the training courses, but it may also reflect the better motivational prerequisites of the courses’ participants. A longer duration of work experience in Austria increased the chance of finding employment. However, our finding that a longer duration of stay in Austria decreased the chance of gaining employment contradicts the above-mentioned results by Faelli and Carless (1999). In the present sample, still being unemployed after a long time in the host country may be indicative of especially poor individual prerequisites for finding work (e.g., health-related problems or, in the case of mothers from collectivist cultures, pressure from the family not to take a job).
As various other independent variables, contrary to expectations, did not contribute significantly to the prediction of unemployment, the results also indicate a fair chance of finding work for apparently disadvantaged refugees (e.g., those who are older, those with a lower educational level, and those with a poor command of German). Similarly, according to our findings, refugees from highly collectivist cultures should not prematurely be considered problematic candidates for the labor market.
The cross-sectional design and the nonrepresentative sample point to limitations in the present study, the results of which may not be generalized to the total population of refugees in Austria. They can, however, be cautiously generalized to the population of currently unemployed refugees, who have already lived in Austria for a longer period of time. For this section of the refugee population, the results emphasize the positive results obtained from vocational rehabilitation through training courses and show the importance of offering jobs to refugees as soon as possible. Whereas work experience in the host country promotes later vocational chances, a long duration of unemployment is a risk factor. Refugees at risk of long-term unemployment can be identified and given specific assistance through the use of the logistic regression model. In addition, the effects that attending training courses are likely to have on their chances of finding employment can be computed.
With respect to the exploratory nature of the present study, additional investigations on a larger scale are advocated. Ideally, such investigations should be longitudinal studies, with representative samples of refugees participating immediately after they have been granted asylum and allowed to work in the host country.
References
Becker, H. A. (1980). The assertive job-hunting survey. Measurement and Evaluation in Guidance, 13, 43-48.
Beiser, M., & Hou, F. (2001). Language acquisition, unemployment, and depressive disorder among Southeast Asian refugees: A 10-year study. Social Science & Medicine, 53, 1321-1334. http://doi.org/fn5pp5
Bierbrauer, G., Meyer, H., & Wolfradt, U. (1994). Measurement of normative and evaluative aspects in individualistic and collectivistic orientations: The Cultural Orientation Scale (COS). In U. Kim, H. C. Triandis, Ç. Kâğitçibaşi, S. C. Choi, & G. Yoon (Eds.), Individualism and collectivism: Theory, methods, and applications (pp. 189-199). Thousand Oaks, CA: Sage.
Bloch, A., Galvin, T., & Harrell-Bond, B. (2000). Refugee women in Europe: Some aspects of the legal and policy dimensions. International Migration, 38, 169-190. http://doi.org/fjwzvq
Bühl, A. (2011). SPSS 20: Einführung in die moderne Datenanalyse [Introduction to modern data analysis] (13th ed.). Munich, Germany: Pearson.
Ekhaugen, T. (2005). Immigrants on welfare: Assimilation and benefit substitution. Memorandum No. 18. Retrieved from http://www.sv.uio.no/econ/english/research/memorandum/pdf-files/2005/Memo-18-2005.pdf
Ellis, R. A., & Taylor, M. S. (1983). Role of self-esteem within the job search process. Journal of Applied Psychology, 68, 632-640. http://doi.org/bkpvwq
Faelli, R. C., & Carless, S. A. (1999). A comparison of psychological characteristics of employed and unemployed professionally qualified immigrants. Australian Psychologist, 34, 183-187. http://doi.org/fddhj9
Hansen, J., & Lofstrom, M. (2006). Immigrant-native differences in welfare participation: The role of entry and exit rates. IZA Discussion Paper No. 2261. Bonn, Germany: Forschungsinstitut zur Zukunft der Arbeit. Retrieved from http://ftp.iza.org/dp2261.pdf
Jäckering, N. (2007). Die Integration von Migranten auf dem Arbeitsmarkt [Migrants‘ integration on the labor market]. Retrieved from http://www.grin.com/e-book/81918/die-integration-von-migranten-auf-dem-arbeitsmarkt
Kofman, E., Phizacklea, A., Raghuram, P., & Sales, R. (2000). Gender and international migration in Europe: Employment, welfare, and politics. New York: Routledge.
Kogan, I. (2003). Ex-Yugoslavs in the Austrian and Swedish labour markets: The significance of the period of migration and the effect of citizenship acquisition. Journal of Ethnic and Migration Studies, 29, 595-622. http://doi.org/dvxr55
Lehrl, S. (1977). Mehrfachwahl-Wortschatz-Intelligenztest MWT-B [Multiple choice vocabulary intelligence test (MWT-B)]. Erlangen, Germany: Straube.
Martella, D., & Maass, A. (2000). Unemployment and life satisfaction: The moderating role of time structure and collectivism. Journal of Applied Social Psychology, 30, 1095-1108. http://doi.org/bxwhfw
Mollica, R. F., Caspi-Yavin, Y., Bollini, P., Truong, T., Tor, S., & Lavelle, J. (1992). The Harvard Trauma Questionnaire: Validating a cross-cultural instrument for measuring torture, trauma, and posttraumatic stress disorder in Indochinese refugees. Journal of Nervous and Mental Disease, 180, 111-116. http://doi.org/cb7
Mollica, R. F., Wyshak, G., de Marneffe, D., Khuon, F., & Lavelle, J. (1987). Indochinese versions of the Hopkins Symptom Checklist-25: A screening instrument for the psychiatric care of refugees. The American Journal of Psychiatry, 144, 497-500.
Ong, A. S. J., & Ward, C. (2005). The construction and validation of a social support measure for sojourners: The Index of Sojourner Social Support (ISSS) Scale. Journal of Cross-Cultural Psychology, 36, 637-661. http://doi.org/ccb
Otto, K., & Dalbert, C. (2012). Individual differences in job-related relocation readiness: The impact of personality dispositions and social orientations. Career Development International, 17, 168-186. http://doi.org/jf2
Ward, C., Bochner, S., & Furnham, A. (2006). The psychology of culture shock (2nd ed.). New York: Routledge.
Becker, H. A. (1980). The assertive job-hunting survey. Measurement and Evaluation in Guidance, 13, 43-48.
Beiser, M., & Hou, F. (2001). Language acquisition, unemployment, and depressive disorder among Southeast Asian refugees: A 10-year study. Social Science & Medicine, 53, 1321-1334. http://doi.org/fn5pp5
Bierbrauer, G., Meyer, H., & Wolfradt, U. (1994). Measurement of normative and evaluative aspects in individualistic and collectivistic orientations: The Cultural Orientation Scale (COS). In U. Kim, H. C. Triandis, Ç. Kâğitçibaşi, S. C. Choi, & G. Yoon (Eds.), Individualism and collectivism: Theory, methods, and applications (pp. 189-199). Thousand Oaks, CA: Sage.
Bloch, A., Galvin, T., & Harrell-Bond, B. (2000). Refugee women in Europe: Some aspects of the legal and policy dimensions. International Migration, 38, 169-190. http://doi.org/fjwzvq
Bühl, A. (2011). SPSS 20: Einführung in die moderne Datenanalyse [Introduction to modern data analysis] (13th ed.). Munich, Germany: Pearson.
Ekhaugen, T. (2005). Immigrants on welfare: Assimilation and benefit substitution. Memorandum No. 18. Retrieved from http://www.sv.uio.no/econ/english/research/memorandum/pdf-files/2005/Memo-18-2005.pdf
Ellis, R. A., & Taylor, M. S. (1983). Role of self-esteem within the job search process. Journal of Applied Psychology, 68, 632-640. http://doi.org/bkpvwq
Faelli, R. C., & Carless, S. A. (1999). A comparison of psychological characteristics of employed and unemployed professionally qualified immigrants. Australian Psychologist, 34, 183-187. http://doi.org/fddhj9
Hansen, J., & Lofstrom, M. (2006). Immigrant-native differences in welfare participation: The role of entry and exit rates. IZA Discussion Paper No. 2261. Bonn, Germany: Forschungsinstitut zur Zukunft der Arbeit. Retrieved from http://ftp.iza.org/dp2261.pdf
Jäckering, N. (2007). Die Integration von Migranten auf dem Arbeitsmarkt [Migrants‘ integration on the labor market]. Retrieved from http://www.grin.com/e-book/81918/die-integration-von-migranten-auf-dem-arbeitsmarkt
Kofman, E., Phizacklea, A., Raghuram, P., & Sales, R. (2000). Gender and international migration in Europe: Employment, welfare, and politics. New York: Routledge.
Kogan, I. (2003). Ex-Yugoslavs in the Austrian and Swedish labour markets: The significance of the period of migration and the effect of citizenship acquisition. Journal of Ethnic and Migration Studies, 29, 595-622. http://doi.org/dvxr55
Lehrl, S. (1977). Mehrfachwahl-Wortschatz-Intelligenztest MWT-B [Multiple choice vocabulary intelligence test (MWT-B)]. Erlangen, Germany: Straube.
Martella, D., & Maass, A. (2000). Unemployment and life satisfaction: The moderating role of time structure and collectivism. Journal of Applied Social Psychology, 30, 1095-1108. http://doi.org/bxwhfw
Mollica, R. F., Caspi-Yavin, Y., Bollini, P., Truong, T., Tor, S., & Lavelle, J. (1992). The Harvard Trauma Questionnaire: Validating a cross-cultural instrument for measuring torture, trauma, and posttraumatic stress disorder in Indochinese refugees. Journal of Nervous and Mental Disease, 180, 111-116. http://doi.org/cb7
Mollica, R. F., Wyshak, G., de Marneffe, D., Khuon, F., & Lavelle, J. (1987). Indochinese versions of the Hopkins Symptom Checklist-25: A screening instrument for the psychiatric care of refugees. The American Journal of Psychiatry, 144, 497-500.
Ong, A. S. J., & Ward, C. (2005). The construction and validation of a social support measure for sojourners: The Index of Sojourner Social Support (ISSS) Scale. Journal of Cross-Cultural Psychology, 36, 637-661. http://doi.org/ccb
Otto, K., & Dalbert, C. (2012). Individual differences in job-related relocation readiness: The impact of personality dispositions and social orientations. Career Development International, 17, 168-186. http://doi.org/jf2
Ward, C., Bochner, S., & Furnham, A. (2006). The psychology of culture shock (2nd ed.). New York: Routledge.
Table 1. Binary Logistic Regression Variables
The authors gratefully acknowledge the funding of this study by Tiroler Wissenschaftsfonds (TWF).
Walter Renner, Department of Psychology, Clinical Psychology, Psychotherapy, and Psychoanalysis Unit, Alpen-Adria - Universität Klagenfurt, Universitätsstraße 65-67, 9020 Klagenfurt am Wörthersee, Austria. Email: [email protected]