School to Work Transition in the Turkish Labor Market Keiichi Ogawa Graduate School of International Cooperation Studies Kobe University, Kobe, Japan Aysit Tansel Department of Economics Middle East Technical University, Ankara, Turkey and Institute for the Study of Labor(IZA), Bonn, Germany 1
Introduction This paper will provide evidence on various aspects of the relationship between schooling and employment in Turkey. The main questions asked are as follows. To what extent do the knowledge and skills the graduates obtained from years of schooling and training relate to their jobs? Do the leducation and skills acquired at schools have a favorable impact on the eventual job performance? Further investigations are carried out using differen types of vocational high schools which involve different programs of study. 2
Introduction All of the analysis are carried out for man and women separately to examine the differences by gender by using a special module of the 1997 Household Labor Force Survey. Such issues related to the vocational education programs are critically important for Turkey but they have never been systematically analyzed for vocational high schools in general and by the type of vocational high schools in particular. a For this reason, the results could generate important policy implications concerning Turkiy's education system, especially at a time when educators worldwide are increasingly advocating a greater emphasis on education and training. 3
Educational System Until the educational reform of 1997, the compulsory level of schooling was five years provided by the primary schools. In 1997, the compulsory level of schooling was extended d toeight years covering themiddle school. Since the survey data used in this study pertains to 1997 which is the year of educational reform, the relevant school structure is the pre-reform one. During the pre- reform period the middle school of three years of study followed the compulsory level of five years of primary schooling. 4
Educational System Further education could proceed along either at General High Schools or at Vocational High schools. Which take typically three but sometimes four years of study. University education provides two-to six years of training depending on the program of study. The extention of the compulsory schooling to eight years and the establishment of the 25 new universities i i since the early 1990s expanded the educational opportunities greatly. 5
Educational System Enrollments increased at all levels recently and the gender gap in education improved substantially. Adult literacy increased from 90 percent for men and 67 percen for women in 1990 to 95 percent for men and 80 percent for women in 2002. Secondary education enrollment rates increased from 46 percent for boys and 30 percent for girls in 1990 to 81 percent for boys and 60 percent for girls in 2002. 6
The Data ThedatausedinthisstudycomesfromtheOctober1997 Household Labor Force (HLFS) of Turkey conducted by the Statistical Institute of Turkey with the purpose of investigating the linkage. The October 1997 HLFS has a special section called Education and Training Supplementary Questionnaire which was implemented only in the October 1997 survey. This section elicited information on the formal and adult education of the population. The survey covers 21,004 households which included 15,530 households in the urban locations and 5,474 households in the rural locations in all of the geographical regions of Turkey. Urban locations are those with over 20,000 population. 7
The Data According to the October 1997 HLFS the labor force participation rate among individuals who completed formal education is about 52 percent. This rate is about 50 percent among those who completed primary schooling and rises consequtively for each level of education. For instance, 67 percent of vocational high school and 81 percent of university graduates participate in the labor force. 8
The Data The labor force participation rate differs by geographic location. Participation rates are higher in the rural areas than in the urban ones. There are vast differences in the participation rates by gender. The participation rate for females at 24 percent is significantly lower than that for males at 74 percent. However, this gender gap declines with level of education. Female participation rate among the university graduates is about 72 percent compared to the 87 percent for males. 9
Methodology This study relies on examination of distributions of various self-assessment questions. Pearson Chi-Square test statistics are reported at the bottom of the tables of relevant distributions. The analysis concentrates on vocational highh school and university it graduates' selfassessment of their education in two dimentions. The graduates are asked to assess the extent to which their education is related to their current job and enhanced their productivity in their job. The distributions are produced d for men and women separately. Further, the analysis is extended to the graduates of various types of vocational high schools. 10
Methodology In the 1997 HLFS, individuals are asked the following questions to evaluate several aspects of their education. 1. Is the department you have graduated from educational institution related to your present job? (Question no. 48). 2. Did your education taken in educational institution help you to do your job better? (Question no. 50). 11
Methodology 3. What is the reason of not working at a job related to your education? (Question no. 49). 4. Where did you learn main Occupation? (Question no. 41). 12
Methodology The answers to the first two questions ranged from 1- Yes, 2- Partially to 3- No. Question 1, essentially probes the match between content of education and skill requirements of the job. Therefore, this question can be related to the literature on educational mismatches that has received much attention recently 13
Methodology The second question essentially asks about the productivity contribution of the education. Question 3 is asked only to individuals who are not working at a job related to their education. In Questions 3 and 4 the respondent is confronted with several choices. In the latter part of our analysis the first two questions above are tabulated for the vocational high school graduates by the type of various vocational high schools attended. Finally, a logit analysis is performed to determine the factors that contribute to better job performance including level of education controlling for gender, marital status, and age. 14
Estimation Results In this section we first present the tables for the distributions in relation to the Questions 1-4 for vocational high school and university graduates. These are given in Tables 1-4. Tabulations are given for men and women separately. Next, Questions 1-2 are tabulated for vocational high school graduates in Tables 5-6 again by gender. In the notes section of each of the tables Pearson Chi-Square statistics i are reported for men and women separately. These statistics test the independence of vocational high school and university levels of education for the question asked in each table. 15
Table 1. Distribution According to Relation of Education to Current Job, Turkey, 1997 (N=3,334) (%). Gender Male Yes Partially Vocational secondary graduates 43.3 14.7 University graduates 67.8 14.1 No 42.0 18.11 All males 100.0 100.0 Female Yes 52.4 76.3 Partially No 12.4 35.2 All females 100.0 (273) 12.0 11.7 100.0 (657) TOTAL (1,318) 2,016 16
Table 2. Distribution According to Contribution of Education to better Job Performance, Turkey, 1997 (N=3,334) (%). Gender Male Contribution of Vocational High Education to School University Better job Graduates Graduates Performance Yes 46.3 70.77 Partially No 15.5 38.2 15.7 13.6 All males 100.0 100.0 Female Yes Partially No 55.3 15.4 29.3 79.4 12.5 8.1 All females 100.0 (273) 100.0 (657) TOTAL (1318) (1,318) (2016) (2,016) 17
Table 3. Distribution According to Reason for not Working at a Job Related to Education, Turkey, 1997 (N=858) (%). 18
Table 3. Distribution According to Reason for not Working at a Job Related to Education, Turkey, 1997 (N=858) (%). 19
Table 4. Distribution According to Where They Acquired their Occupational Skills, Turkey, 1997 (N=3,334) (%). 20
Table 4. Distribution According to Where They Acquired their Occupational Skills, Turkey, 1997 (N=3,334) (%). 21
Table 5. Relationship of Education and Current Job by Type of Vocational High Schools, Turkey, 1997 (N=1,318) (%). Types of Men Women Vocational High School Yes Partially No Yes Partially No Religious 26.7 13.3 60.0 16.7 16.7 66.6 Technical 39.1 16.3 44.6 46.9 13.3 39.8 Commercial/ Tourism 42.9 18.6 39.4 34.2 20.3 45.5 Military/ Police 88.1 -- ll.9 -- -- -- Health 88.2 5.9 5.9 93.0 1.8 5.2 Finance -- -- -- -- -- -- Other 44.0 9.5 66.6 50.0 5.6 44.4 Total 43.3 (453) 14.6 (153) 42.1 (439) 52.4 (143) 12.5 (34) 35.1 (96) 22
Table 6. Contribution of Education to Better Job Performance by Type of Vocational High Schools, Turkey, 1997 (N=1,318) (%). Types of Vocational Men Women High School Yes Partially No Yes Partially No Religious Technical Commercial/ Tourism Military/ Police 25.0 11.7 63.3 16.7 16.7 66.6 44.0 15.8 40.3 50.4 14.2 35.4 31.6 44.7 19.1 36.2 40.6 27.8 -- 86.6 1.5 11.9 -- -- Health 88.2 5.9 5.9 89.4 5.3 5.3 Finance 0.0 100.0 0.0 -- -- -- Other 43.1 19.0 37.9 55.6 0.0 44.4 Total 46.3 (484) 15.5 (162) 38.2 (399) 55.3 (151 15.4 (42) 29.3 (80) 23
Table 7. Logit Analysis of the Contribution of Education to better Job Performance. (N=2694) (%). Standard Wald Odds Independent Variables Coefficient Error Statistics Ratio Education Level 0.96* 0.38 6.48 2.60 Gender -0.71* 0.36 3.90 0.49 Marital Status Married -0.26 0.44 0.34 0.77 Divorced -2.81** 1.07 6.84 0.06 Widowed -2.06 1.65 1.55 0.13 Age 0.01 0.02 0.02 1.00 Constant -3.53 2.35 2.26 0.03 24
Conclusions This paper considered the link between education and labor market performance in Turkey. Several dimentions are investigated in this regard using the individuals self-assesment. The first pertains to the job related skills namely, whether vocational o a high school and university graduates are ae able to obtain jobs that are related to their education. The second pertains to productivity namely whether vocational high school and university education contribute to better job performance. 25
Conclusions There has been no previous attempts to evaluate the education system in Turkey empirically along these dimentions. For this purpose we used the special module of the October 1997 HLFS and Pearson Chi-Square tests and a logit regression. ess The findings provide support for the human capital theory in that both vocational high school and university education provide job related skills and improve productivity. 26
Conclusions Women tend to benefit more from schooling both in terms of provision of job related skills and in terms of improving their productivity. Another important result is that on the job training is the second important process for acquiring occupational skills s after formal education. These findings provide support for the human capital theory such that schooling provides job related skills and improves productivity. 27
Conclusions An examination of various vocational high school types indicate that military or police and health high schools are most efficient in importing the necessary job skills to their graduates while the religious high schools are the least efficient in doing this. Similarly, the former group of schools s contribute to higher productivity of their graduates on their jobs compared to the graduates of religious high schools. Thus, policy makers must be contious about increasing the numbers of religious high schools if the concern is efficiency and productivity in the labor market. 28
Conclusions Folllowing generalizations can be made from our analysis. First, the higher the level of education, the higher is the probability that individuals believe that their education is related to their job. Second, the higher the level of education the higher is the probability that individuals think that their education contributes to their job performance. Third, the higher the level of education, the lower is the probability that individuals can not find a job related to their education. Fourth, the higher the level of education, the higher is the probability that individuals rely on formal education for the acquisition of their occupational skills. 29
Conclusions These findings show that it is less likely for graduates from secondary vocational education in comparison with university graduates to find a job that is directly related to what they have learned in school. Additionally, in terms of job performance, it is more likely for the latter than the former that their education has helped them in their work. It is not surprising, therefore, that our study shows a greater reliance on on-the-job training for secondary vocational graduates than university graduates. 30
Conclusions In terms of policy development, several implications follow from the foregoing results. Ensuring that all graduates benefit from their education is a clear priority - we suggest that the government promote on-the-job training/apprenticeship program to students, sude s,especially at the secondary school level, e, and develop such a system between schools and prospective employers to further hone the skills of potential job candidates. 31
Conclusions Futhermore, areas such as relevance of curriculum and teaching (in terms of quality measure) should be reviewed, specifically in alignment with the labor market s demand. Experience learned from OECD countries such as United States Saes and Japan may be useful u and applicable to Turkish vocational secondary schooling. 32
Conclusions Vocational high schools in Turkey may need to integrate vocational training in academic coursework to provide students with the opportunity to learn academic skills in an alternative, applied context. In this way, vocational high school students will be able to cope and compete e in the contemporary labor market as they adjust their skills to suit its changing needs. 33