Pathways of Learning: Do Immigrant Students Choose Vocational Education?

I continue exploring PISA data for my research grant on immigrants in Balkan countries. Analyzing PISA 2018 and 2022 data, I noticed that in most countries, pupils enrolled in vocational tracks demonstrated lower PISA scores. Even after controlling for socioeconomic status and parents’ educational and occupational background, vocational students still lagged behind their peers in general education tracks.

To properly estimate the effect of general versus vocational tracks, more advanced data analysis is required beyond multiple regression. Many years ago, we published a paper examining the impact of one additional year of education in different tracks for countries with early tracking systems  (Kuzmina, J., & Carnoy, M. (2016). The effectiveness of vocational versus general secondary education: Evidence from PISA 2012 for countries with early tracking. International Journal of Manpower, 37(1), 2-24). We used Fuzzy Discontinuity regression and showed that although vocational students had lower overall performance, their year-on-year learning gains in PISA scores were not necessarily lower than those of students in general tracks.

Now, I am turning my attention to immigrants and their educational trajectories. Since vocational tracks allow students to acquire professional qualifications earlier than general tracks, and given that immigrants in many countries often face barriers to accessing higher education, it is possible that immigrants may either choose vocational education voluntarily or be directed towards it.

To explore this issue, I initially examined the proportion of pupils in different tracks and grades across selected countries (Austria, France, Italy, Greece, Croatia, Montenegro, Serbia, and Slovenia). Below, you can see the distribution of students by track and grade.

PISA 2022: Distribution of Pupils by Grade and Track

In Serbia and Croatia, most pupils in the sample were in the 9th grade, whereas in other countries, the majority were in the 10th grade, except in Austria, where the proportions of 9th and 10th graders were nearly equal.

Next, I inspected the PISA math and reading scores for general and vocational students in the selected grades for each country (for Austria and Croatia, both grades are included).

Key Observations

1. Performance Gap Between Tracks: In every selected country except Austria (9-th grade), students in the general track had significantly higher scores than their vocational-track peers. In Slovenia and France, the difference reached 100 points—a substantial gap, exceeding the average standard deviation for OECD countries. The smallest difference between general and vocational students was observed in Italy.

2. Cross-Country Comparisons: When considering both grade level and track, Croatia and Serbia had relatively high results compared to other countries. For example, 9th-grade students in Serbia (equivalent to the first year of gymnasium) achieved scores comparable to 10th-grade students in other countries, such as Slovenia or France. Notably, Serbia’s average PISA math performance was lower than that of France or Slovenia, highlighting the importance of considering grade level and educational track when interpreting PISA results.

   
   

   The patterns for reading performance were similar to those observed for math.

   
   

   

   
   

It is important to note that these results are not adjusted for additional variables such as socioeconomic status or school characteristics. When adjusted for SES, the differences between general and vocational tracks would likely be smaller.

Immigrants and the Probability of Enrolling in Vocational Tracks

My primary goal was to determine whether immigrant students have a higher likelihood of being placed in vocational tracks after controlling for SES, gender, parental education, and home language. To test this, I ran a logistic regression analysis.

Findings:

·       In Montenegro, both first- and second-generation immigrants had significantly higher odds of being in a vocational track.

·       In Serbia and Slovenia, neither first- nor second-generation immigrants showed significantly different odds of enrolling in vocational education compared to native students.

·       In France and Austria, both first- and second-generation immigrants had lower odds of being in a vocational track than native students.

·       In Italy and Greece, only second-generation immigrants had lower odds than native students of being in a vocational track.

In every selected country, higher SES reduced the likelihood of being in a vocational track, and girls were also less likely to enroll in vocational education. Interestingly, speaking a language other than the national language at home was associated with higher odds of vocational enrollment in Serbia, Slovenia, Italy, France, and Austria but lower odds in Montenegro.

Below are several graphs displaying the odds ratio coefficients from the logistic regression analysis (with the dependent variable being enrollment in the vocational track) for three Balkan countries, along with Austria for comparison.

Odds Ratios from Logistic Regression for Enrollment in the Vocational Track

Conclusion

This analysis provides two key insights:

1.     Understanding PISA Results Requires Context – Cross-country differences in PISA scores should be interpreted in relation to grade level and educational track structure. Simply comparing national averages without considering these factors may lead to misleading conclusions.

2.     Immigrants Are Not Necessarily Overrepresented in Vocational Tracks – Despite their lower academic performance in most selected countries (except Serbia and Montenegro), immigrant students do not always have a higher probability of enrolling in vocational tracks. In fact, in Austria, France, and Greece—where the native-immigrant achievement gap is relatively large—immigrants are less likely than native students to be in vocational education.

It is important to emphasize that this analysis does not aim to establish causal effects of educational track selection or the mechanisms driving student placement. However, it highlights that vocational tracking is not necessarily a default pathway for immigrant students, even in countries where their academic performance is lower.

The implications of educational track selection extend beyond academic performance and immediate employment prospects. One of the key concerns regarding vocational education is its potential long-term impact on career trajectories. While vocational programs may provide students with practical skills and a quicker transition into the labor market, their long-term adaptability and opportunities for career advancement remain in question.

Hampf and Woessmann (2017) provide valuable insights into this issue. Their research, based on PIAAC data from multiple countries, demonstrates that while individuals with vocational education initially have better employment opportunities than those with general education, this advantage declines over time. Around the age of 45–50, the employment probability of individuals with general education surpasses that of those with vocational training. This pattern is particularly pronounced in apprenticeship-based systems, where industry-specific training limits workers' ability to adapt to economic changes later in life. These findings emphasize the importance of considering both the short-term and long-term consequences of educational track selection, especially for populations, such as immigrants, who may already face structural barriers in the labor market.

Reference:

Franziska Hampf, Ludger Woessmann, Vocational vs. General Education and Employment over the Life Cycle: New Evidence from PIAAC, CESifo Economic Studies, Volume 63, Issue 3, September 2017, Pages 255–269, https://doi.org/10.1093/cesifo/ifx012