Employability of University Graduates as a Measure of Quality

Marek Rocki

Abstract

Keywords: monitoring graduates on the labour market, quality assurance, labour market, employability of graduates, universities

Introduction

Although there is no universally accepted definition of the quality of education, the problem of its measurement is the subject of many studies and analyses, both scientific and formal (cf. Rocki, 2005).

In practice, the assessment of the quality of education in higher education institutions is most often associated with the analysis of the effects of the external quality assurance system of education, with the assessment criteria defined in ESG (The Standards and guidelines for quality assurance in the European Higher Education Area) indicating in this context external stakeholders, including employers.

For example, standards 1 and 2 indicate stakeholders in connection with the creation and improvement of programmes and quality assurance. In particular, standard 2.2 states that for external quality assurance it is extremely important to have clear goals agreed upon by stakeholders, including employers¹. However, one may question the extent to which cooperation between universities and employers is genuine. It is potentially possible in the case of a small university operating for the needs of the local labour market, but in the case of large universities, whose students come from different regions of the country (and also from abroad), and whose graduates to a significant extent take up employment with diverse employers, including outside the place of study (in Poland this is over 20%²), such cooperation is symbolic. In ESG we can read that "quality, although not easy to define, is mainly the result of interactions between teachers, students and the institutional learning environment". It seems that for graduates, however, what is important is the result of this quality understood as their situation on the labour market. Information in this regard is collected and made available in Poland by the Polish Graduate Tracking System.

This text indicates the possibilities of quality assessment using data from this system and confronts the conclusions from this data with the assessments formulated by the Polish Accreditation Committee.

Literature Review

The formal assessment of the quality of education in Poland is currently the domain of the Polish Accreditation Committee (Polska Komisja Akredytacyjna, PKA)³, but as it is pointed out in the literature (cf. Chmielecka, 2015) in connection with the regulations governing the operation of the PKA, these assessments are largely related to control and administrative functions. Among the education quality standards used by the PKA is cooperation with the socio-economic environment, including employers. Standards in this area do not play a significant role in the final assessment, although existing studies indicate that PKA assessments are features of higher education institutions related to the situation of graduates on the labour market (cf. Grotkowska & Gaik, 2019; Rocki, 2018; 2024). Of course, the completion of studies itself gives rise to a salary premium (cf. Grotkowska & Pastuszka, 2019; Rocki, 2021; Rutkowski, 1996; Wincenciak, 2017).

In the literature, the quality of education is also associated with prestige and the quality of scientific research indicated by Black, Kermit, Smith (2005) in the case of US universities (graduating from a prestigious university is associated with higher salaries by 2 to 17%) and Hussain, McNally and Telhaja (2009) for British universities (the estimated difference between the salaries of graduates of the best and the weakest universities was from 10 to 16%). The latter text also indicates the relationship between graduates' salaries and quasi-accreditation, which is the university's membership in the Russell Group - an elite institution associating the best British universities. The relationship between university prestige and the employability of graduates in the case of Poland was discussed, among others, by Długosz (2013).

The impact of self-selection of candidates for studies on graduates' salaries, resulting from the selectivity of recruitment (related in turn to prestige), is discussed, among others, by Hoxby, Terry (1999) and Conlon, Chevalier (2003), Rocki (2025). Problems with measuring the quality of education are indicated, among others, by Black, Smith (2006), concluding that the existing literature underestimates the impact of university quality on graduates' earnings.

Polish Graduate Tracking System

The system was created at the request of the Ministry of Science and Higher Education. Data on graduates come from the state social insurance system (ZUS) and the POL-on system (the ministry's IT system, which collects data on science and higher education throughout Poland) and are of an administrative nature. The authors call the system the Polish Graduate Tracking System (the acronym PGTS will be used in this study).

Currently (June 2025) the PGTS includes data on 3,006,853 graduates from 2014-2022, including 1,741,533 from 2014-2018, about whom there is information from five consecutive years on the labour market (by definition, data is collected for five years from graduation). Data is available on graduates registered with ZUS, constituting 94.32% of the total⁴. The system automatically generates several standard reports for Poland, for each operating university and for each field of study, and also provides data in the form of a spreadsheet. In this form, data is provided on approximately 700 indicators showing, among others, unemployment, time spent looking for a job, and remuneration.

The basic unit of analysis is a cohort of graduates who received their diplomas in a given year from a given university and its organisational unit, in a particular field of study, at a defined level⁵ and mode⁶ of study. This means that for a given field of study (for example: economics) at a given university, in a given year, data on several groups of graduates may be available (for economics: four groups). It should be emphasised that although data is collected on all graduates of all universities registered with ZUS, information on groups of less than 10 people is not made available for anonymity. Data is currently available on over 90 thousand groups, and in some cases - as mentioned earlier - this is data for the five period after graduation. In practice, data sharing begins two years after obtaining diplomas, so for example for the year of 2022, information on the situation one year after graduation was made available in 2024.

Providing data on cohorts of graduates allows for their aggregation and obtaining information on graduates of the university, its faculties, groups of universities, and groups of fields of study (e.g. technical or natural sciences).

For the analysis of data from the PGTS, it is important to note that data from ZUS do not contain information on the profession performed, making it is impossible to determine whether there is a connection between the field of study completed and the work performed⁷.

The analyses omitted groups in which the share of graduates registered with ZUS was less than 60%, but due to the small number of graduates in such groups, this does not affect the conclusions presented below.

Only graduates of full-time second-cycle studies were taken into account in the analyses, due to the fact that:

• in most cases, graduates of first-cycle studies undertake further studies,
• in most cases, part-time students worked during their studies, which burdens the application (in particular, their salaries immediately after graduation are on average higher than the salaries of full-time graduates).

The situation on the labour market for the years 2014-2018 was taken into account, as for these years data is available for five years from obtaining diplomas. The choice of these years is also related to the fact that the values of the indicators selected for the analyses will be confronted with the assessments of the quality of education made by the PKA, and until 2018, this commission issued the following assessments: negative, positive, distinguished and conditional. Information about the assessments was taken from the database made available by the PKA via its website (see: https://pka.edu.pl/ocena/baza-uczelni-jednostek-i-kierunkow-ocenionych/). Since 2018, the PKA has not issued distinguished and conditional assessments, but instead certificates of excellence in education and positive assessments for a period of up to 2 years are issued.

It is worth adding here that positive assessments constitute (depending on the term of office of the PKA) from 78 to 88% of assessments, with distinguishing assessments from 2 to 5%, and conditional assessments from 7 to 19%⁸.

Assessment of Graduates by the Labour Market

The quality of education results from the composition, commitment and competence of academic staff, the quality and scope of research conducted, the comprehensiveness, structure, style of implementation and quality of teaching programmes, university infrastructure, the functioning of study support systems, as well as the effectiveness of the internal quality assurance system. All of these together constitute a set of variables that in econometrics, in soft modelling, are known as creating indicators (i.e. formative indicators). On the other hand, there are also reflecting indicators (reflective indicators) that indicate the effects of the quality of education, including information collected in the PGTS system, such as: the average time spent looking for the first job, the average number of months in which graduates were registered as unemployed, the share of unemployed graduates, the average monthly salary, etc.

For the analyses, two synthetic indicators characterising the situation of graduates on the labour market were selected from the PGTS:

WWB - relative unemployment rate, calculated in such a way that for individual graduates in the period covered by the study, an individual proportion of the risk of unemployment⁹ to the average registered unemployment in the county of residence (or counties, if the place of residence has been changed) is determined, with the value of the indicator being the average of these proportions. Values below 1 mean that on average the risk of unemployment among graduates is lower than the unemployment rate in their counties of residence, while values above 1 mean that on average the risk of unemployment among graduates is higher than the unemployment rate in their counties of residence,

WWZ - relative wage index, calculated by determining the proportion of each graduate’s average wage to the average wage in the county (counties) of residence in the period under review, with value of the index equal to the average of these proportions.

Analysed together, these two indicators characterise the employability of graduates: the probability of finding a job and a satisfactory salary. The presented analyses concern the values of WWB and WWZ for five years (total, i.e. for the entire study period) after obtaining diplomas.

Since the analyses are cross-sectional (comparing the situation of graduates of a given year in the same fields of study from different universities), the results are not influenced by data from the period when the economy was recovering from the pandemic.

It should be emphasised that the definition and method of calculating relative unemployment and wage rates allow for omitting the differences in the economic development of the regions of the country where graduates work, which has an impact on wages calculated in absolute terms, in the analyses. For example, a certain level of wages may be considered low in a large city and high in a small town. For this reason, in order to compare the situation of graduates on the labour market, their wages and the threat of unemployment should be compared with data for the local labour market.

Assumptions and Analysis Results

As mentioned earlier, graduates of second-cycle studies are considered in the analysis, with four fields of study with significant numbers of students, run by numerous universities of different types, selected. Since universities independently establish fields of study and set admission limits for studies guided (most often) by the demand expressed by the number of candidates, graduates of the same fields of study from different universities compete with each other for (theoretically) the same jobs on the labour market, meaning that differences in employability may in such a case be the basis for formulating conclusions about the quality of education (opinions of employers). On the other hand, the popularity of selected fields of study led to the development by the PKA of well-established quality assessment criteria (due to cyclicality, repeatability and comparability).

The selected fields of study are: construction (1.77% of all graduates of full-time second-cycle studies), economics (1.85% respectively), computer science (1.6%) and management (3.32%)¹⁰.

In Table 1, in addition to the WWB and WWZ values for these fields of study, the following are given:

• average time (in months) from graduation to first job after graduation,
• average percentage of months worked after graduation in any form,
• average monthly salary of graduates from all sources in the five years following graduation.

As can be seen from Table 1, the values of indicators for the surveyed groups are significantly diversified for graduates of selected fields of study: from a very short time of searching for a job to a dozen or so months, from less than 50% of months of work (within five years of obtaining the diploma) to almost 100%, from zero WWB values (indicating no threat of unemployment) to a threat several times higher than the average in the counties of residence, etc.

Table 1
Minimum and Maximum Values of Selected Indicators for Graduates from 2014–2018

Source: author's own work.

As the results presented below show, such significant differentiation is not reflected in the assessments made by the PKA, although this is, of course, related to the assessment scale used by the PKA.

Graphs 1-4 present the WWB values, and graphs 5-8 the WWZ values for graduates of the 2014-2018 cohorts for the analysed fields of study, separating graduates of universities that have ever¹¹ received a distinguished or conditional grade from the PKA for individual fields of study. The graphs also contain values for universities that have received positive grades or have not been assessed by the PKA (marked on the graphs as "other"). The tabular data used to prepare the graphs are included in the Appendix.

Figure 1
WWB Values for 2014–2018 Construction Engineering Graduates

Source: author's own work.

Figure 2
WWB Values for 2014–2018 Graduates of Economics

Source: author's own work.

Figure 3
WWB Values for 2014–2018 Computer Science Graduates

Source: author's own work.

Figure 4
WWB Values for 2014–2018 Management Graduates

Source: author's own work.

Figure 5
WWZ Values for 2014–2018 Construction Engineering Graduates

Source: author's own work.

Figure 6
WWZ Values for 2014–2018 Graduates of Economics

Source: author's own work.

Figure 7
WWZ Values for 2014–2018 Computer Science Graduates

Source: author's own work.

Figure 8
WWZ Values for 2014–2018 Management Graduates

Source: author's own work.

As can be seen in the graphs, in the vast majority of cases, the values of WWB and WWZ for graduates of universities that received distinguished grades differ positively from the values for other graduates, which, taking into account the definitions and interpretations of WWB and WWZ, means that on average, graduates of courses that received outstanding grades are assessed by employers better than graduates of courses that received conditional or positive grades.

Such a clear difference is not seen in the case of courses that have never received a conditional assessment in relation to the same courses that had received a positive assessment or had not received a PKA assessment, which can be for several reasons:

1) a conditional assessment obtained in a certain year meant that remedial actions had to be taken, which were then verified by the PKA (usually in the following academic year). The result of such verification had to be a positive assessment, as otherwise it would mean deprivation of the right to teach a given field of study;

2) the criteria for positive assessments are (were) formulated on the basis of regulations, defining the minimum level of meeting the required criteria. If the reasons for not meeting the criteria were removed and as a result the university received a positive assessment, this fact would not necessarily significantly affect the quality of education and thus significantly improve the values of the indicators;

3) the study examines the years of graduates for whom information is available for five years from the year of obtaining their diplomas. The values of the analysed indicators are averages for each of the five years, while the conditional assessment was ‘point-based’ – issued in a certain year. In the years following this assessment, the quality of education had to improve (otherwise, the university would lose the right to teach in a given field), but this did not have to significantly affect the values of the indicators.

Conclusions and Discussion of Results

The data from the system for monitoring the economic fate of graduates show that the labour market evaluates graduates of identical courses offered at different universities in a significantly different way. Although such significant differentiation is not reflected in the assessments made by the PKA, which is related to the criteria and grading scale used, the analyses conducted clearly indicate a positive relationship between distinction assessments and the situation of graduates on the labour market.

It should be noted, however, that in the case of positive assessments, data from the monitoring system provide more quantitative information indicating the effects of a specific state of education quality than reports from assessments made by accreditation committees (which by their nature contain rather qualitative information). On the other hand, descriptive reports from visits allow for a better interpretation of what results from the system for monitoring the economic fate of graduates.

It should be emphasised that the data provided by the PGTS are not burdened with the subjective opinions expressed by students during visits conducted by the PKA in the course of assessing the quality of education and in surveys conducted among graduates.

When assessing the possibility of using data from the system for monitoring the economic fate of graduates in assessing the quality of education, it should be noted that:

• the use of data related to employability is more consistent with the classification of better/worse than good/bad. In the case of many universities (fields of study), worst does not mean bad, but research indicates that there are fields of study with a distinguished rating, whose graduates have significant difficulties on the labour market (cf. Rocki 2018). On the other hand, the persistent, unfavourable values of the indicators (WWB > 1 and WWZ < 1) should encourage the university authorities to take action, since they indicate the poor situation of graduates on the labour market;
• the Polish higher education system provides freedom in naming fields of study¹², which means that fields of study with similar substantive content may have different names (for example: ‘finance and accounting’, ‘finance, auditing and investments’), and at the same time fields of study with the same name may differ in their programmes and specialisations offered. This obviously affects the assessment of the quality of education, as well as the situation of graduates on the labour market;
• the Polish higher education system offers many fields of study that have not been assessed by the PKA, as they are launched by universities that have the authority to create fields of study without ministerial approval (i.e. without ex ante accreditation) and are often incidental in nature (i.e. they exist only at one university). On the other hand, the monitoring system allows for ex post evaluation. For example¹³, for the field of study: trusted artificial intelligence systems (Wrocław University of Science and Technology, 21 persons) WWB = 0.4; WWZ = 1.13 and a salary of 9034.33 PLN indicate a very good situation of graduates on the labour market, with a similar situation observed for the field of study: financial management of an enterprise (SGH, 47 persons): WWB = 0, WWZ = 1.4, salary 10202.18 PLN, while for the field of study: knowledge of film and audiovisual culture (University of Gdańsk, 22 persons) WWB = 0.96; WWZ = 0.25 and an average salary of PLN 1986.25 indicate a very bad situation of graduates;
• data from the system for monitoring the economic fate of graduates are of an administrative nature, therefore providing objective information and covering practically all graduates, unlike information from surveys conducted by universities;
• because the data in the system for monitoring the economic fate of graduates are published annually, they provide more up-to-date information than the system for assessing the quality of education carried out by accreditation committees.

The significant differences in the situation of graduates of fields of study that received outstanding grades indicate, on the one hand, the durability of the features characterising the studied fields of study in the analysed universities, while on the other hand, confirm the hypothesis of the occurrence of the Matthew effect, which also been observed in other aspects of the operation of higher education institutions (cf. Farys & Wolbring, 2021; Kwiek & Roszka, 2023; 2024; 2025; Merton, 1968; Rocki, 2024).

To sum up: in order to enrich and supplement the information, internal and external assessment of the quality of education should use information from the system for monitoring the economic fate of graduates, and in the absence of such assessments, the situation of graduates on the labour market may be an effective approximation of such an assessment.¹⁴

Appendix

References

• Black, D. A., Kermit, D., & Smith, J. A. (2005). College quality and wages in the United States. German Economic Review, 6(3), 415-443. https://doi.org/10.1111/j.1468-0475.2005.00140.x
• Black, D. A., & Smith, J. A. (2006). Estimating returns to college quality with multiple proxies for quality. Journal of Labor Economics, 24(3),701-728. https://doi.org/10.1086/505067
• Chmielecka, E. (2015). Europejskie Standardy i Wskazówki zapewniania jakości w szkolnictwie wyższym a polski model akredytacji [European Standards and Guidelines for quality assurance in higher education and the Polish Accreditation Model]. Edukacja Ekonomistów i Menedżerów, 2(36), 29–40.https://doi.org/10.5604/01.3001.0009.4589
• Conlon, G., & Chevalier, A. (2003). Does it pay to attend a prestigious university? IZA Discussion Paper No. 848. https://ssrn.com/abstract=435300
• Długosz, P. (2013). Sytuacja absolwentów studiów wyższych na rynku pracy a jakość wykształcenia [Situation of high school graduates on job market and the quality of education]. Nierówności Społeczne a Wzrost Gospodarczy, 33, 43-55.
• Farys, R., & Wolbring, T. (2021). Matthew effects in science and the serial diffusion of ideas: Testing old ideas with new methods. Quantitative Science Studies 2(2), 5–526. https://doi.org/10.1162/qss_a_00129
• Grotkowska, G., & Gaik, A. (2019). Wpływ cech szkoły wyższej na sytuacje absolwentów na rynku pracy [The influence of university characteristics on the situation of graduates on the labor market]. Rynek Pracy, 2(169), 31–45.
• Grotkowska, G., & Pastuszka, S. (2019). Stopa zwrotu z edukacji w Polsce w świetle badań empirycznych [The rate of return on education in Poland in the light of empirical research]. Rynek Pracy, 2(169), 20-30.
• Hoxby, C. M., & Terry, B. (1999). Explaining rising income and wage inequality among the college-educated. Working Paper 6873. National Bureau of Economic Research. http://www.nber.org/papers/w6873
• Hussain, I., McNally, S., & Telhaja, S. (2009). University quality and graduate wages in the UK. IZA Discussion Paper No. 4043. https://doi.org/10.2139/ssrn.1355426
• Kwiek, M., & Roszka, W. (2023). Zawsze wysoce produktywni? Zawsze nisko produktywni? Dynamiczne ujęcie dorobku naukowego profesorów tytularnych wykorzystujące klasy produktywności [Always highly productive, always low productive? A longitudinal and class-based approach to research output of full professors]. Nauka, 1, 7–39. https://doi.org/10.24425/nauka.2023.144944
• Kwiek, M., & Roszka, W. (2024). Reguła 10/50. Analiza dorobku 150 tysięcy polskich naukowców pokazuje trwałość systemowych nierówności w nauce [The 10/50 rule. An analysis of research output of 150,000 Polish scientists shows permanent systemic inequalities in science]. Nauka 4, 37–61. https://doi.org/10.24425/nauka.2024.152910
• Kwiek, M., & Roszka, W. (2025). Are scientists changing their research productivity classes when they move up the academic ladder? Innovative Higher Education, 50, 329–367. https://doi.org/10.1007/s10755-024-09735-3
• Merton, R. (1968). The Matthew Effect in science. Science, 159(3810), 56–63. https://doi.org/10.1126/science.159.3810.56
• Rocki, M. (2005). Statistical and mathematical aspects of ranking: Lessons from Poland. Higher Education in Europe, 30(2), 173–181. https://doi.org/10.1080/03797720500260165
• Rocki, M. (2018). Jakość kształcenia a ekonomiczne losy absolwentów. Analiza przypadków [The quality of teaching and the economic future of graduates. Case analysis]. Nauka i Szkolnictwo Wyższe, 1(51), 219-239. https://doi.org/10.14746/nisw.2018.1.11
• Rocki, M. (2021). Wage premium to higher education based on data from Polish Graduate Tracking System. Gospodarka Narodowa, 3(307), 47–61. https://doi.org/10.33119/GN/140647
• Rocki, M. (2024). Ranking szkół wyższych jako predyktor zatrudnialności [Ranking of universities as a predictor of employability]. Nauka, 3, 33-54. https://doi.org/10.24425/nauka.2024.151943
• Rocki, M. (2025). O zjawisku spadku względnych zarobków absolwentów [The phenomenon of declining relative earnings of graduates]. Ekonomista, 2, 243-258. https://doi.org/10.52335/ekon/195856
• Rutkowski, J. (1996). High skills pay off: the changing wage structure during economic transition in Poland. Economics of Transition, 4(1), 89–112. https://doi.org/10.1111/j.1468-0351.1996.tb00163.x
• Wincenciak, L. (2017). Stopa zwrotu z wykształcenia wyższego w Polsce według grup kierunków studiów. Edukacja, 143(4), 5–16.