Graduates Tracking with Administrative Data How it is Done in - - PDF document

1

Graduates’ Tracking with Administrative Data – How it is Done in Poland

Mikołaj Jasiński, University of Warsaw (mikolaj.jasinski@uw.edu.pl) Marek Bożykowski, University of Warsaw (mbozykowski@uw.edu.pl) Agnieszka Chłoń-Domińczak, Warsaw School of Economics (achlon@sgh.waw.pl) Tomasz Zając, University of Warsaw (t.zajac@uw.edu.pl) Abstract The massification of higher education in Poland means that labour market outcomes of graduates are an important perspective for future students, higher education institutions as well as the managers of higher education at national level. The Polish Graduate Tracking System, based on the administrative data, allows for monitoring of graduates’ outcomes on the labour market by type of studies, the higher education institution as well as individual

• curricula. The absolute and relative measures allow assessing the outcomes taking into

account the local labour market perspective. Results of the first two waves of graduate tracking show that the outcomes vary by study area, but also change in time. While in the short run, aspects such as prior experience on the labour market and the place of residence affect employment chances, in the longer run the labour market processes become more important.

2

In Intr troducti tion

The Polish Graduate Tracking System (ELA system) was introduced in 2014 by the rule of the Law on Higher Education. This followed the need to have better knowledge on the graduates’ transition from school to work in Poland. In the past years, Poland has undergone significant changes of its educational system, including higher education. At the beginning of the 1990s, only about 10% of the most talented youth completing upper secondary school were admitted to university each year and the Polish education institutions were considered elitist. Today, over half of each year’s upper secondary school graduating class pursues higher education studies, and the net enrolment ratio reached around 40% in the middle of 2000s and stayed at that level since then. (Central Statistical Office (GUS) 2015). The ‘massification’ of higher education induced in recent years a vivid public debate pointing out the need to assess the quality of ever more accessible tertiary education. The increase in academic enrolment was accompanied by significant changes in the entire system of higher education in Poland. Until 1990, all higher education institutions were state-

• wned, but in 1990 a new Law on Higher Education allowed creation of private higher

education institutions and introduction of paid part-time study programmes at public schools. This led to the increase in the total number of students, in particular in part-time courses at private institutions. This trend started reversing in early 2000s. Since 2002, the proportion of full-time students at public institution has been gradually increasing again, as a result of, among others, demographic processes and decline in the total number of students (Ministerstwo Nauki i Szkolnictwa Wyższego 2013). The ELA system introduced in Poland allows for monitoring of the economic outcomes of higher education understood as employment and earnings of graduates. In the paper, we present the design of the ELA system as well as its results, after the first two waves of the

• monitoring. The paper is structured as follows. In the first section, we briefly present the

education system in Poland, including the developments in the higher education area and

• verall situation of the graduates of higher education institutions in the labour market. In the

second section, we describe the main features of the graduate tracking system, which uses administrative data from two main registers: higher education register which includes the information on individual graduates, provided by all higher education institutions and the social insurance register, containing information on employment and wages of individuals. In the third section, we present the main results of the monitoring, including the absolute and relative indicators used for assessment of the economic situation of graduates. This is followed by the section on a probit model focusing on the characteristics affecting the employment and wages after graduation. In the analysis, we focus on the graduates from second cycle or master courses who finished education in 2014. This choice is motivated by the fact that the vast majority of the first cycle graduates continue their education and do not intend to become economically active yet. Section five presents the main conclusions related to the up-to-date experiences of tracing graduates in Poland.

3

1.

• 1. Educat

Education n sys ystem in n Po Poland and and and its out utco comes bas based d on n th the labour r mark rket t indicato tors rs

The education system in Poland comprises of several stages. The first stage is the 9-year general education in primary schools (6 years) and lower secondary schools (3 years). From September 2017, it will be gradually replaced with the 8-year primary education. The second stage is the upper secondary education, that lasts from 2 to 4 years, depending on the type of the upper secondary school. From September 2017, the secondary education will last from 3 to 5 years. School leavers from general upper secondary schools (licea ogólnokształcące) and technical upper secondary schools (technika), after passing external maturity examination, can access higher education institutions, which is shown in Figure 1.

Figure 1. Education system in Poland

Source: National Center on Education and the Economy (http://ncee.org)

4 The system of higher education in Poland is regulated by the Act – Law on Higher Education

• f 27 July 2015, as amended. According to this act, higher education is provided by public

and private institutions established for this purpose in the manner prescribed by law. Institutions providing public full-time curricula receive funding from the state budget. Part- time studies are financed from tuition paid by students. Higher education is organised, according to the Bologna framework as follows: first cycle, second cycle, long cycle master’s degree studies, doctoral studies, and postgraduate studies. Graduates can obtain the following diplomas (Marciniak et al. 2013):

• 1. Studies corresponding to the Bologna first cycle (BA):
• a diploma certifying the professional title of licentiate (licencjat),
• a diploma certifying the professional title of engineer (inżynier),
• a diploma certifying the professional title equivalent to licentiate or engineer (for

example engineer in fire prevention, licentiate in midwifery).

• 2. Studies corresponding to the Bologna second cycle (MA):
• a diploma certifying the professional title of master (magister),
• a diploma certifying the professional title of master engineer (magister inżynier),
• a diploma certifying the professional title of equivalent to master (for example, the

title of physician)

• 3. Doctoral studies
• A diploma certifying the title of doctor in a specific academic discipline or doctor
• f arts in a specific discipline of fine arts.
• 4. Certificates of completion of postgraduate studies.

The shape of the Polish higher education system is affected on the one hand, by demographic and social processes, related to the increased demand for the higher education and by the higher education policies on the other hand. The increased aspirations of young Poles were the most important social development driving the changes in the higher education. As it was mentioned in the introduction, an increased share of secondary school leavers enrolling into higher education programmes. Following this trend, the number of students in higher education institutions increased rapidly from about 400,000 in early 1990s to reach a peak in 2006 at the level of 1.93 million people. After 2006, the number of students started to fall, following the demographic changes. Namely, the fertility level in Poland dropped from above to 2.0 in early 1990s to 1.6 in 1995 and below 1.4 from 1999 (Kotowska 2014). As a result, the number of people in age group 19-23 years started to gradually decline and the number of students in 2016 dropped to 1.35 million with stable enrolment levels. The current number of students is still more than 4 times larger compared to the early 1990s. The increase of students from early 1990s led to the increased supply of the studies offered by newly emerging private higher education institutions. In 2008, the share of students in private higher education institutions was 34.5%, while the share of students in tuition-free,

5 state funded full-time programmes at public institution was 41.7%. Furthermore, the offer of part-time studies increased, both at public and private higher education institutions. The part- time programmes offer classes in the afternoon or during weekends, which allows students to work while pursuing education. The part-time studies are subject to a tuition, that is charged both in the case of public as well as private higher education institutions. The rapid development of the paid higher education studies in the part-time form as well as in private institutions raised some concerns. One of them is the suspicion that quality of education depends on the type of the institution (public vs non-public) and the type of the studies (full-time vs part-time). It is argued that students of part-time programmes receive education of lower quality (Herbst and Rok 2011). There are also concerns regarding equity as students of privileged backgrounds are more likely to enrol into a state-subsidised programme at a public institution (Herbst and Rok 2014). On the other hand, the emergence

• f the private higher education institutions allowed young people from rural areas and smaller

towns to access higher education (Kotowska, Chłoń-Domińczak, and Saczuk 2014). The decline in the overall number of students observed in the past decade led to the reduction

• f the number of students at private institutions as well as part time students, which means

that that the full-time studies at public institutions have regained their leading position. In 2016, around a quarter of all students were in private higher education institutions and those in public full-time studies made up 55.5% (Figure 2).

Figure 2. Number of students by ownership of the higher education institution and form of studies, 2002-2016.

Source: Local Data Bank, Central Statistical Office The introduction of the Bologna system in Poland led to the gradual change of the structure

• f students’ population in Poland. The share of students in the long cycle master’s degree

programmes declined, while the first-cycle students became the most numerous group. In

200 000 400 000 600 000 800 000 1 000 000 1 200 000 1 400 000 1 600 000 1 800 000 2 000 000 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 public full 3me public part 3me private full 3me private part 3me

6 2016, 64.8% of all students were enrolled into first cycle studies, 25.0% in the second cycle, and 10.8% in the long cycle master studies.

Figure 3. Number of students by the type of studies cycle, 2002-2016

Source: Local Data Bank, Central Statistical Office As a result of the above described processes, the share of young people with higher education attainment in Poland between 2002 and 2016 tripled, from 14.4% to 44.6%.

Figure 4. Tertiary education attainment in Poland and EU-28, 30-34 age group.

Source: Eurostat What seems to concern both the public opinion and politicians most about tertiary education is the graduates’ prospective employability. Although Polish economy fared well compared to other European economies and kept growing even in the years following the 2008 crisis, the labour market did not improve significantly at the same time (Boulhol 2014). After the

200 000 400 000 600 000 800 000 1 000 000 1 200 000 1 400 000 1 600 000 1 800 000 2 000 000 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 long cycle master studies first cycle studies second cycle studies 5 10 15 20 25 30 35 40 45 50 EU (28 countries) Poland

7 decline observed between 2004 and 2007, the unemployment rate among tertiary education graduates and young people in general rose between 2007 and 2013 (Rokicka et al. 2015).

Figure 5. Unemployment rate among recent graduates by type of e, moving average of quarterly data, 2010-2016

Source: authors’ calculations based on the Labour Force Survey data The unemployment rate among recent higher education graduates1 in the fourth quarter of 2016 was estimated to be 12,6%, compared to more than 21.8% among all school leavers (Central Statistical Office 2017), which is a stable tendency (Figure 5). Due to relatively favourable labour market condition, unemployment rate among recent graduates in Poland is also lower than EU average (Rokicka et al. 2015). Despite this overall positive assessment, there is a significant diversity of the labour market situation of graduates, which can be partially attributed to the type of studies and higher education institution. As pointed out by Marciniak et al. (2013, p. 2) the education system in Poland shifted from a purely elitist model to a model of diversified learning that needs to take into account to a much greater degree the diversity in the level of students’ abilities, as well as their interest and goals in life. While some students are undoubtedly still interested in and capable of pursuing research-oriented studies. The vast majority seek education which offers a solid base as well as flexibility that enable them to perform a various jobs and social roles. Given such goals, the monitoring of the economic outcomes of higher education studies, is an important aspect of assessing of the quality of outcomes of higher education in Poland.

1 People who graduated within 12 months before survey and are not in education.

10 20 30 40 50 60 I II III IV I II III IV I II III IV I II III IV I II III IV I II III IV I II III 2010 2011 2012 2013 2014 2015 2016 higher technical secondary and post-secondary general secondary basic voca3onal

8

2.

• 2. Gr

Graduate monitoring using registers in Pol

• land

The ELA system is a centralised system run by the Ministry of Science and Higher

• Education. The system was established in 2014. Its goal is to provide a detailed description of

graduates’ labour market performance for up to five years after they leave higher education institutions. The system uses individual data extracted from two administrative registers. The first register is the POL-on system which is owned by the Ministry of Science and Higher Education. The second one is the register of Social Insurance Institution (ZUS). Relying on administrative data allows the monitoring system to cover the entire population of graduates and greatly reduces the cost of such an endeavour. However, it also limits the analysis to the information collected by the respective administrations (United Nations Economic Commission for Europe, 2007; Wallgren & Wallgren, 2007). The POL-on system serves, among others, as a national register of students and graduates. The data exported to the ELA system include the following information on graduates: higher education institution, department or faculty, study programme, level of studies (BA vs. MA), the mode of delivery (full-time vs. part-time), the date of graduation, information on subsequent enrolment into a different study programme along with its characteristics. The ZUS register provides data on graduates’ monthly contributions to the national social insurance system. The contributions are mandatory for the vast majority of the economically active population. Records of contributions include the following data:

• Status in the labour market, including the type of work arrangement (i.e. salaried

worker, self-employed, unemployed, on maternity or parental leave),

• Basis for the payment of the social insurance contribution (podstawa wymiaru

składki) – the amount used to calculate the due contributions. For employed people, this figure equals the wage in PLN in each month. For self-employed individuals, the amount is declared in most cases fixed at the minimum required level of 60% of average wage in the economy and thus not indicative of the income. Data from these registers can be accurately and effortlessly linked due to the adoption of the national identification number PESEL as an ID in both databases. The PESEL number is a key component of the Polish system of administrative registers. It has been used primarily as an ID in The Common Electronic System of Population Register (PESEL). The PESEL database operates since 1979 and comprises personal details of Polish citizens and foreigners residing in Poland, including names, address, marital status, sex, the date and place of birth. The PESEL number is similar to personal identification numbers used in Scandinavian countries (Ludvigsson, Otterblad-Olausson, Pettersson, & Ekbom, 2009; Pedersen, 2011). It is an 11-digit numeric symbol which identifies a person, carries information on the date of birth and sex, and includes check digits (OECD, 2017). Poland has adopted an approach similar to the one taken be the Nordic countries (Poulain & Herm, 2013) and has been using

9 the PESEL number as an ID in multiple administrative registers. The implementation of the PESEL number makes the process of merging data simple and helps to avoid problems with data linking (mostly missed links and mislinks) experienced by other researchers who have to rely on combination of variables such as sex, address, and date of birth (Chowdry, Crawford, Dearden, Goodman, & Vignoles, 2013; Kim, Tamborini, & Sakamoto, 2015; Oreopoulos, von Wachter, & Heisz, 2006). The annual process of graduate tracking in the ELA system in Poland is organised as follows. First, the Ministry extracts the list of graduates from POL-on for a given date. The list is then delivered to ZUS which exports data on the social insurance contributions and labour market

• status. ZUS is also responsible for data anonymisation. PESEL numbers are replaced with

generic IDs. Data are subsequently transferred to the Ministry’s research team which computes the analytical database and oversees the production of reports. There are three types of reports. The first and most important kind describes the labour market performance of graduates of each programme. The second depicts the situation of graduates of each level of studies at every higher education institution. The third presents the results for all graduates of a certain level in entire country. All reports are generated automatically and published on a publicly available website. In 2017 a total of 38 448 reports were published. The reports covered 747,160 people who graduated from Polish universities in 2014 and 2015. The automatic reports consist of monitoring indicators’ values2 with short descriptions automatically generated depending on these values and guidelines for proper interpretations

• f those indicators. The reports include various indicators regarding three main areas:
• time spent looking for a job;
• job stability (periods of unemployment, types of contracts, time with specific types of

contracts, number of employers);

• wages.

Indicators are provided in absolute values and in relative terms. The relative indicators designed by the University of Warsaw expert team in cooperation with Educational Research Institute are an important component of the ELA system, which is a unique solution applied

• nly in Poland.

The content of reports is designed in a way that they cover information needs of main recipients – candidates, students, graduates, employees and policy makers. The reports are kept simple, so recipients are not required to have advanced knowledge of statistics. In the period of the system implementation, the content of the reports is broadened on a regular

• basis. Every year the report schemes take into account information on yet another year

following graduation, until they cover full five years after graduation of the followed cohorts.

2 See section 3 for the indicators description.

10 The same data are used also by the expert team for more advanced analyses following request

• f policy makers.

The system includes solutions developed to protect the privacy of graduates. An automatic report will not be generated if the group in question is not large enough (i.e., there were only a few graduates of a certain programme in the past year, hence there are chances of identifying individuals based on their records).

11

3.

• 3. Rel

Relative e indicator

• rs and

and labo abour ur mar arke ket out utco comes and and th their r use in th the gra raduate te monito tori ring

Poland is an economically diverse country encompassing both prosperous and struggling

• regions. The geographical economic diversity is visible in public statistics on unemployment

and income in Polish counties (powiat3). The distribution of the unemployment rate and the average wages in Polish counties in 2015 is presented in the Table 1 below.

Table 1. Economic diversity of Polish counties

Unemployment rate Average wage Minimum 3% 2 569 PLN (700 USD) 1st Quintile 10% 3 190 PLN (869 USD) 2nd Quintile 13% 3 344 PLN (911 USD) 3rd Quintile 16% 3 516 PLN (958 USD) 4th Quintile 20% 3 744 PLN (1 020 USD) Maximum 35% 6 956 PLN (1 895 USD) Source: Authors’ assessment based on Local Data Bank, GUS An adequate assessment of graduates’ labour market performance should take into account the economic context in which they are functioning. The same result can be considered as a success on a difficult job market and as a failure on an easy one. That is the reason why taking the situation on the local job market into consideration is needed. This need was the starting point for designing relative indicators that are used in the ELA

• system4. The indicators compare graduates’ the labour market situation in the given period to

the situation of their neighbours – the average citizens of their counties of residence. The percentage of months during which a graduate was unemployed is compared to the unemployment rate and his or her wage – to the average wage in his or her county. The possible changes in the place of residence in the analysed period are taken into account in calculations. Relative indicators correct the bias of an evaluation of the wage and risk of unemployment related to the location of HEIs and by graduates changing their place of residence. Without this correction, the analysis would lead to an obvious artefact – the graduates from low- quality HEIs in strong economic centres (e.g. the capital city) would in most cases do better than the graduates from the top provincial HEIs.

3 Poland is divided into 380 powiats. Powiat is the second-level unit of local government and administration,

equivalent to a county, district or prefecture in other countries. In the following text we use the term “county” as the translation for “powiat”.

4 Developed in the European Social Fund sponsored project conducted by the University of Warsaw expert team

in cooperation with Educational Research Institute

12 Examples of interpretation of the Relative Indicator of Unemployment (RIU) and the Relative Indicator of Wages (RIW) based on the data from the Polish Graduate Tracking System are presented below. Relative Indicator of Unemployment (RIU) For every graduate the proportion of his or her individual risk of unemployment (i.e. the percentage of months being registered as unemployed) to the average unemployment rate or registered unemployment rate in his or her county (or counties) of residence in the analysed period is calculated. The Relative Indicator of Unemployment for the group of graduates is the average of these proportions. The indicator has a very intuitive interpretation. Values below 1 mean that on average the graduates’ risk of unemployment is lower than the unemployment rate in their counties of residence, while values above 1 means that their risk of unemployment is higher than the unemployment rate in their counties of residence. The merits of the RIU are well illustrated with the example shown in Figure 6.

Figure 6. Relative Indicator of Unemployment – an example

The bar graphs above show that the absolute and the relative indicators of the risk of unemployment present different picture of the risk of unemployment among the graduates of economic studies from higher education institutions in different regions. RIU can be considered a measure of “vulnerability to unemployment”. The graduates of Poznań University of Life Sciences are in absolute terms less frequently unemployed compared to the graduates of the University of Technology and Humanities in

• Radom. However, the relative risk among the Poznań graduates is higher than the relative

risk of the Radom graduates. The latter face much more difficult situation on the local labour market (unemployment rate in their place of residence at the time of measurement was on average 20% compare to 7% for Poznań graduates). Therefore, given the local labour market context, the graduates of the University of Technology and Humanities in Radom perform better than the graduates of Poznań University of Life Sciences.

13.5% 5.0% University of Technology and Humanities in Radom Poznań University of Life Sciences

Risk of unemployment

0.71 1.16 University of Technology and Humanities in Radom Poznań University of Life Sciences

Rela%ve Indicator of Unemployment

13 Relative Indicator of Wages (RIW) Another relative indicator used in the graduate monitoring is Relative Indicator of Wages. For every graduate, the proportion of his or her average wage to the average wage in his or her county (or counties) of residence in the analysed period is calculated. The Relative Indicator of Wages for the group of graduates is the average of these proportions. This indicator also has an intuitive interpretation. Values below 1 means that on average the graduates earned less than average wage in their counties of residence, while values above 1 means they earned more than the local average wage. The interpretation of this indicator is shown in Figure 7 and explained below. The RIW can be considered a measure of “earnability”.

Figure 7. Relative Indicator of Wages – an example

Figure 7 presents the absolute and the relative indicators of unemployment among graduates from economy in two HEIs: the first one in the capital city (Warsaw), the other provincial. Although the graduates from AlmaMer HEI in Warsaw earn more, it’s the graduates from the High School of Commerce in Tarnobrzeg who are richer in comparison to their neighbours. The big part of AlmaMer graduates can be expected to live in Warsaw or in counties nearby, in a richer region, where it is much easier to find a well-paid job (as be said on the basis of average wage in graduates’ counties of residence: 1245 USD in counties of residence of AlmaMer graduates in comparison to 941 USD in the second group). Taking into account the local labour market, graduates from the HSC in Tarnobrzeg are better prepared for the job market than the graduates from AlmaMer. Dynamic analyses with relative indicators The relative indicators can take into account the dynamics of economic situation, both the long-term changes and the seasonal changes (e.g. in annual cycles). Therefore, the developed tool is applicable for a dynamic analysis without risk of taking economic fluctuations for changes in the economic situation of the analysed group.

1 044 860 AlmaMer HEI in Warsaw Higher School of Commerce in Tarnobrzeg

Average remunera+on (USD)

0.81 0.90 AlmaMer HEI in Warsaw Higher School of Commerce in Tarnobrzeg

Rela%ve Indicator of Remunera%on

14 The Figure 8 presents monthly dynamics of the Relative Indicator of Wages among Polish graduates from master-level programmes from 2014 for the 24 months after graduation. The data covers the entire population of Polish graduates who got their master’s degree in 2014.

Figure 8. Dynamics of the Relative Indicator of Wages among Polish graduates who got master’s degree in 2014

Source: Authors’ analyses based on ELA system As shown above, in the analysed period the graduates’ RIW has been systematically growing. In this period, the average wage in Poland also grew from 4004 PLN (1091 USD) in 2014 to 4226 PLN (1152 USD) in 2016. The growth of the graduates’ RIW means not only that their earnings grew but also that the growth was faster than in case of their neighbours. Full-time and part-time programmes graduates on the basis of the ELA system The ELA system can be used as a simple and effective tool for answering questions on higher education in Poland. An example of the problem solved with the ELA system was the fair assessment of part-time and full-time study programmes. As it was presented in Section 1, in Poland part-time studies are popular. When we take into account graduates the share of those who studied part-time is 44%. The graduates from part- time studies tend to achieve better results on the job market than their full-time counterparts. For instance, the graduates from part-time master-level studies achieve higher average wage than the graduates from full-time master-level studies – 3168 PLN (863 USD) in comparison to 2908 PLN (792 USD). Figure 9 shows dynamics of the Relative Indicator of Wages in both groups.

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Month a(er gradua.on

Rela.ve Indicator of Wages

15

Figure 9. Dynamics of the Relative Indicator of Wages among Polish graduates who graduated from full-time and part-time master-level studies in 2014

At the first sight, one might think that part-time programmes prepare their students better for the job market than full-time studies. However, the part-time students are much more likely to work during their studies, or even to have a job before entering the HEI. Only 25% of full- time studies graduates had a job experience before graduation, while among part-time studies graduates it was almost 75%. Previous job experience leads to higher pay, lower risk of unemployment, and shorter job search. That is the reason why part-time studies graduates achieve better results on the labour market. Figure 5 shows the significance of pre-graduation job experience for the RIW value.

Figure 10. Dynamics of the Relative Indicator of Wages among Polish graduates who graduated from master-level programmes in 2014 depending on the form of studies and job experience before graduation

As the Figure 10 above demonstrates, when we take the entire population of graduates, initial wages of the part-time studies graduates are higher compared to full-time studies graduates. With time, the latter manage to outperform the former. It takes 9 months in the case of the graduates with earlier job experience and just 2 months in the case those with no labour

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Rela%ve Indicator of Wages Month a6er gradua%on

Part-%me vs. full-%me programmes

Part-2me programmes Full-2me programmes

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Rela%ve Indicator of Wages Month a6er gradua%on

Part-%me vs. full-%me programmes - controlling for job experience

Part-2me programmes (with job exp.) Full-2me programmes (with job exp.) Part-2me programmes (no job exp.) Full-2me programmes (no job exp.)

16 market experience before. As a result, the full-time studies graduates achieve higher average wage than part-time studies graduates both among those with pre-graduation job experience – 3616 PLN (985 USD) in comparison to 3413 PLN (930 USD) – and without job experience – 2595 PLN (707 USD) vs. 2170 (591 USD). On top of that, the full-time studies graduates’ position on the job market improves faster. The explanation is that full-time programmes graduates are perceived by employers to have higher competences than the part-time programmes’ graduates. The reason for this may be the fact that full-time programmes are more selective than part-time ones and they are considered to be more prestigious. The differences between full-time and part-time programmes graduates are even bigger in the case of graduates with no previous job experience. Figure 11 below presents their average wages in consecutive months after finding employment.

Figure 11. Dynamics of the Relative Indicator of Wages in consecutive months after finding a first job among Polish graduates with no previous job experience who graduated from full-time and part-time master-level studies in 2014

As shown above, a full-time studies graduate receives better pay from the very beginning of his or her first job. Moreover, the pay gap keeps growing over time.

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Rela%ve Indicator of Wages Month a6er finding the first job a6er gradua%on

Part-%me vs. full-%me programmes a6er finding the first job

Part-2me programmes Full-2me programmes

17

4.

• 4. Wh

What affects the labour market performance of

• f Pol
• lish

gr graduat aduates?

The model Reshaping the educational and labour market system is not possible without a proper diagnosis of significant factors affecting labour market entry. We investigate the subject with a probit model based on the data from the ELA system. The model explores determinants of the graduates’ employability every three months for two years after graduation (technically, these are eight models, one for every quarter). The model is but one example of a more in- depth analysis of the graduates’ labour market performance based on the ELA data and which is not a part of the automatic reports in the ELA system. The dependent variable in the model is a dummy variable indicating whether a graduate has a steady job, i.e. an employment contract or self-employment. Polish regulations allow also

• ther forms of job contracts, including civil contracts or specific-task contracts, yet these

forms of contracts are meant for temporary work. In the public discourse, they are commonly called ‘junk contracts’. These contracts, even if systematic, are not considered steady, as they do not guarantee stability of employment. On top of that, some of these contracts, namely specific-task contracts, are not registered in ZUS. This kind of contract is relatively rare. It is a main form of employment only for less than 0.5% of working population (GUS, 2016). Information about these contracts do not appear in the data, while all employment contracts and self-employments are registered (Chłoń-Domińczak, Sowa, and Topińska 2017; Lewandowski and Keister 2015). We use a probit model to predict the employment status of the graduates. The model provides the probabilities for every graduate to be employed in a given month. The prediction is based on the independent variables mentioned below:

• Job experience – a dummy variable; informs if the graduate had any employment

contract or was self-employed during a few months before graduation.5

• Form of studies – informs if the graduate used to study full-time or part-time.
• Size of place of residence – the category of the size of the place of residence: over

500 000 citizens (big city); smaller than 500 000 but the city is a separate county (medium city); small town or village; unknown place of residence. The information of the place of residence is based on the records of ZUS. For some of the graduates the place of residence has not been noted in ZUS records, more

• ften if that person was not employed.
• Type of HEI – provides information if a specific person graduated from public

HEI, non-public HEI or ecclesiastical HEI. Only a small percent of graduates belongs to the last category. In general, public HEIs in Poland are considered to

5 The exact number of months taken into account depends on the date of graduation – it is the consequence of

data export procedures. In most cases it is about 6 months.

18 be more prestigious than non-public ones, and full-time studies in public HEIs require no tuition.

• Field of study – informs to what field of study belongs the programme that a

specific person graduated from. In Poland study programmes are grouped into 8 fields: humanities; medical and health sciences; natural sciences; agricultural sciences; social sciences; exact sciences; technical sciences; arts.

• Sex – male or female. 68% of graduates in the chosen population are female.
• Studying after graduation – a dummy variable; informs if the graduate studied on

some other study programme after graduation; it counts programmes started both after graduation as well as those started before that this person has not graduated from yet.

• Age category – informs on age category in the year of graduation; categories are

25 or less and 26 or more. In Polish education system if someone had no pause or delay in his or her education, he or she would get master’s degree at age of 25 or earlier .

Results

The estimated model is well fitted for the first quarter after graduation, but the fit declines for models that are estimated for the following quarters. It is not surprising: many of the independent variables in the model are related to the study programme. The more time passes from the moment of graduation, the smaller should be the impact of these features on the labour market performance. The decrease of the model fit is presented on the Figure 12 below.

Figure 12. Dynamics of the Model Fit Index every three months for two years after graduation

0.0 0.1 0.2 0.3 0.4 Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8

Quarter a(er gradua+on

Model Fit Index

19 The Model Fit Index6 decreases step by step from 0.37 three months after graduation to 0.13 two years after graduation. The following Figure 13 shows how much every single independent variable brings to the

• model. Partial correlation index for the Model Fit Index was assessed for each of the

independent variables.7 The detailed results can be found in the table in the annex part.

Figure 13. Dynamics of the partial correlation index for the chosen independent variables every three months for two years after graduation

The most important factor was pre-graduation job experience. Its impact decreases rapidly

• ver time, as one could expect. Nevertheless, it remains a noticeable factor even two years

after graduation. In the case of other factors, the only one with substantial contribution to the model is the size of place of residence. However, the contribution of this variable is quite low, and does not change significantly over the analysed period. The third factor is the field

• f study, which influence is much weaker than of the previous two.

Surprisingly, all the other factors, i.e. form of study, type of HEI, sex, age and further studying have barely any effect on the prediction of employment – they add almost nothing to the model with the three important factors. Apparently, in Poland these features have no influence on one’s chances to find a job. The contribution of a variable to the model needs to be complemented by the parameters of the model, so one could tell which features go hand in hand with higher chances of

• employment. The parameters of the model are presented in the Table 2 below.

6 Based on the likelihood ratio (increase from the null model to the alternative model). 7 The partial correlation index for MFI of the variable controlling for remaining variables shows how much

better is the full model in comparison to the model lacking that variable (to what extend the likelihood ratio has increased after adding that variable to the model).

0.0 0.1 0.2 0.3 Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8

Quarter a(er gradua+on

Model Fit Index - par+al correla+on

Job experience Size of place of residence Field of study

20

Table 2. Parameters of the model predicting employment situation every three months for two years after graduation

Quarters after graduation Independent variables Quarter 1 Quarter 2 Quarter 3 Quarter 4 Quarter 5 Quarter 6 Quarter 7 Quarter 8 Job experience 2.1 1.7 1.4 1.2 1.1 1.0 0.9 0.8 Sex (F vs. M)

• 0.1
• 0.1
• 0.1
• 0.1

0.0 0.0 0.0 0.0 Graduated at 26+ 0.0 0.0 0.0 0.0

• 0.1
• 0.1
• 0.1
• 0.1

Part-time 0.1 0.0 0.0 0.0 0.0 0.0

• 0.1
• 0.1

Non-public HEI (vs. public)

• 0.1
• 0.1
• 0.1
• 0.1
• 0.1
• 0.1
• 0.1
• 0.1

Ecclesiastical HEI (vs. public)

• 0.1
• 0.1
• 0.1
• 0.1
• 0.1
• 0.1

0.0 0.0 Studying after graduation 0.1 0.0 0.0

• 0.1
• 0.1
• 0.1
• 0.1
• 0.2

Big cities (vs. small towns or villages) 0.3 0.3 0.3 0.2 0.2 0.1 0.1 0.1 Medium cities (vs. small towns or villages) 0.1 0.1 0.0 0.1 0.1 0.0 0.0 0.0 Unknown residence (vs. small towns or villages)

• 0.8
• 0.7
• 0.7
• 0.8
• 0.8
• 0.9
• 0.9
• 0.9

Medical & health sciences (vs. humanities) 0.2 0.5 0.5 0.5 0.4 0.2 0.3 0.4 Natural sciences (vs. humanities)

• 0.3
• 0.2
• 0.2
• 0.1
• 0.1
• 0.1
• 0.1
• 0.1

Agricultural sciences (vs. humanities)

• 0.1

0.0 0.0 0.1 0.0 0.0 0.1 0.0 Social sciences (vs. humanities) 0.0 0.1 0.1 0.1 0.1 0.1 0.2 0.2 Exact sciences (vs. humanities) 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.0 Technical sciences (vs. humanities) 0.2 0.3 0.4 0.4 0.4 0.4 0.4 0.4 Arts (vs. humanities)

• 0.1
• 0.1
• 0.1
• 0.1
• 0.2
• 0.2
• 0.2
• 0.2

Intercept

• 0.4
• 0.1

0.1 0.3 0.5 0.6 0.7 0.8

The table above presents the full set of parameters of the model. For the illustration purposes the most important results are presented in Figures 14 and 15 below.

21

Figure 14. Dynamics of the parameters for the chosen independent variables every three months for two years after graduation

As shown on Figure 14, people with job experience before graduation had much bigger chance of being employed. This ‘bonus’ gets smaller in time, dropping from 2.1 three months after graduation to 0.8 two years after graduation. Still, even after the decrease it was the factor that improved the most graduates’ chances of employment after graduation. Another factor that increased chances of being employed was living in a big city (in comparison to living in a small town or a village). Big cities offer more job opportunities, which goes hand in hand with smaller unemployment rate. The estimated bonus for living in a big city gets smaller over time, as people from smaller localities manage to find employment. Women’s chances for being employed were slightly smaller in the first year after graduation, while in the second year there were the same as men’s. People pursuing further education were a bit more likely to be employed three months after

• graduation. This changes later to the opposite. In the first months after graduation, this group

is mostly people who were continuing a programme they studied in parallel to those they graduate from. In later periods, these are also those who enrolled to another programme after graduation – which might suggest they were not satisfied with the received education. To sum up, the socio-demographic variables’ significance decreases over time while the people who have entered the job market accumulate more work experience.

• 0.5

0.0 0.5 1.0 1.5 2.0 2.5 Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8

Quarter a(er gradua+on

Job experience Sex (F vs M) Studying aEer graduaFon Big ciFes (vs small towns or villages)

22

Figure 15. Dynamics of the parameters for the fields of studies every three months for two years after graduation

The first surprising result of the more detailed analysis on the impact of the field of studies is that graduates in natural sciences were among those who were least likely to be employed – their results are much closer to graduates from arts than to graduates in exact sciences. The atypical shape of the curve for medical and health sciences requires some explanation. It starts with a strong increase, then after a while it decreases rapidly, reaching the lowest value 1.5 years after graduation, and then it rises again. This is a result of Polish regulations regarding physicians and dentists. In 3-4 months after graduation they all start a training period in a hospital (that is why the chance for being employed goes up in the 2nd quarter), which lasts for 13 months. Then they have a gap in employment lasting few months (in the 6th quarter) during which they take licensure exams. If they manage to pass, they start a regular employment or residency. That is the reason for the increase in the 7th and 8th quarter. While the socio-demographic features tend to have lower and lower significance for employability, the influence of the field of study remains more or less stable. It is due to the fact that the field of studies to some extend outlines the kind of job. For instance, only the graduates from medical programmes can work as physicians and it is very unlikely for these graduates to compete for jobs with the graduates in technical or social sciences.

• 0.3
• 0.2
• 0.1

0.0 0.1 0.2 0.3 0.4 0.5 0.6 Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8

Quarter a(er gradua+on

Medical & health sciences (vs humani>es) Natural sciences (vs humani>es) Agricultural sciences (vs humani>es) Social sciences (vs humani>es) Exact sciences (vs humani>es) Technical sciences (vs humani>es) Arts (vs humani>es)

23

5.

• 5. Con

Conclusion

• ns

The Polish Graduate Tracking System provides a new insight into the labour market

• utcomes of the higher education graduates in Poland. While the up-to-date information gave

some insight into the overall performance of graduates, more detailed assessment of particular curricula or types of studies was difficult. Merging administrative data from two major administrative registers: of higher education graduates (part of the POL-on register) and Social Insurance Institution, allows for more effective and cost-efficient assessment of the labour market outcomes of graduates. Given the high participation of young people in higher education, the results of the monitoring can support the decisions on the choice of curricula for the secondary school leavers. It also provides a very rich source of information for deans, rectors and other representatives of higher education institutions, as well as the Minister of Higher Education. The design of the monitoring system, including the set of indicators used to assess the labour market outcomes allow comparing the situation of graduates, taking into account also local labour market conditions. The further waves of the ELA system will shed more light on the outcomes, not only right after graduation, but also in further months. The Polish approach is also very cost-effective. As administrative register data are used, the data collection costs are non-existent, while the coverage is very high. While the official monitoring portal grants public access to automatic reports, the dataset generated in the monitoring allows for a more in-depth assessment. The probit model indicates that with time the impact of individual characteristics diminishes. This is related in particular to the pre-graduation job experience as well as the place of residence. This means that the study content and work experience prior to graduation are important at the start of economic activity, while other factors – related to gaining more job experience – become more important over time.

Ref Refer eren ences es

Boulhol, Hervé. 2014. Making the Labour Market Work Better in Poland. Central Statistical Office. 2015. Labour Force Survey in Poland - I Quarter 2015. Warsaw. Central Statistical Office (GUS). 2015. Higher Education Institutions and Their Finances in

• 2014. Warsaw: GUS.

Chłoń-Domińczak, Agnieszka, Agnieszka Sowa, and Irena Topińska. 2017. “ESPN Thematic Report on Access to Social Protection of People Working as Self-Employed or on Non- Standard Contracts Poland.” (February). Chowdry, Haroon, Claire Crawford, Lorraine Dearden, Alissa Goodman, and Anna Vignoles.

• 2013. “Widening Participation in Higher Education: Analysis Using Linked

Administrative Data.” Journal of the Royal Statistical Society. Series A: Statistics in Society 176(2):431–57.

24

• GUS. (2016). Pracujący w nietypowych formach zatrudnienia. Warsaw.

Herbst, Mikolaj and Jakub Rok. 2011. “Equity of Access to Higher Education in the Transforming Economy. Evidence from Poland.” Pp. 475–94 in Investigaciones de Economía de la Educación 6, edited by A. C. Ruiz. Herbst, Mikolaj and Jakub Rok. 2014. “Equity in an Educational Boom: Lessons from the Expansion and Marketisation of Tertiary Schooling in Poland.” European Journal of Education 49(3):435–50. Kim, ChangHwan, Christopher R. Tamborini, and Arthur Sakamoto. 2015. “Field of Study in College and Lifetime Earnings in the United States.” Sociology of Education 88(4):320– 39. Kotowska, Irena E. 2014. Niska Dzietnośc W Polsce W Kontekście Percepcji Polaków. Kotowska, Irena E., Agnieszka Chłoń-Domińczak, and Katarzyna Saczuk. 2014. “Uwarunkowania Decyzji Edukacyjnych: Wnioski Dla Przyszłości Szkolnictwa Wyższego.” Lewandowski, Piotr and Roma Keister. 2015. “Segmentacja Rynku Pracy a Emerytury W Polskim Systemie O Zdefiniowanej Składce.” Ludvigsson, Jonas F., Petra Otterblad-Olausson, Birgitta U. Pettersson, and Anders Ekbom.

• 2009. “The Swedish Personal Identity Number: Possibilities and Pitfalls in Healthcare

and Medical Research.” European Journal of Epidemiology 24(11):659–67. Marciniak, Zbigniew, Ewa Chmielecka, Andrzej Kraśniewski, and Tomasz Saryusz-Wolski.

• 2013. Self-Certification Report of the National Qualifications Framework for Higher

Education. Mettler, Suzanne. 2014. Degrees of Inequality. How the Politics of Higher Education Sabotaged the American Dream. New York: Basic Books. Ministerstwo Nauki i Szkolnictwa Wyższego. 2013. Szkolnictwo Wyższe W Polsce 2013. Warszawa.

• OECD. 2017. “Poland - Information on Tax Identification Numbers.”

Oreopoulos, Philip, Till von Wachter, and Andrew Heisz. 2006. The Short- and Long-Term Career Effects of Graduating in a Recession: Hysteresis and Heterogeneity in the Market for College Graduates. Cambridge, MA. Pedersen, Carsten Bøcker. 2011. “The Danish Civil Registration System.” Scandinavian Journal of Public Health 39(Suppl 7):22–25. Poulain, Michel and Anne Herm. 2013. “Central Population Registers as a Source of Demographic Statistics in Europe.” Population-E 68(2):183–212. Rokicka, Magdalena, Małgorzata Kłobuszewska, Marta Palczyńska, Nataliia Shapoval, and Jędrzej Stasiowski. 2015. Composition and Cumulative Disadvantage of Youth across

• Europe. Tallinn.

United Nations Economic Commission for Europe. 2007. Register-Based Statistics in the

25 Nordic Countries. Review of Best Practices with Focus on Population and Social

• Statistics. Geneva: United Nations.

Wallgren, Andreas and Britt Wallgren. 2007. Register-Based Statistics. Administrative Data for Statistical Purposes. John Wiley & Sons.

26

An Annex.

Partial correlation index for the independent variables in the probit model every three months for two years after graduation

Quarters after graduation Independent variables Quarter 1 Quarter 2 Quarter 3 Quarter 4 Quarter 5 Quarter 6 Quarter 7 Quarter 8 Job experience 0.26 0.16 0.12 0.09 0.07 0.06 0.05 0.05 Form of studies 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Size of place of residence 0.08 0.06 0.05 0.06 0.06 0.07 0.07 0.07 Type of HEI 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Field of study 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 Sex 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Studying after graduation 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Age category 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00