The Czech Republic, like many other European countries, has had long-term difficulty in providing quality education to Roma children. Roma children have a low participation rate in pre-school education, they finish compulsory education prematurely, and they have a low rate of completion of upper secondary education (e.g. FRA, 2021). The failure of the system in the education of Roma and other socio-economically disadvantaged pupils became even more pronounced during the COVID-19 pandemic, when the long closure of schools contributed to the further lagging behind of these pupils. As part of the National Recovery Plan, a programme to support schools with a higher proportion of socio-economically disadvantaged pupils was launched in the autumn of 2022. 265 basic schools (out of 4261 in the Czech Republic) with the highest proportion of pupils with socio-economic disadvantages carried out an initial analysis of their needs and chose support measures from the offer that was made available: 1. staff support (e.g. salaries of a teaching assistant, special pedagogue, social pedagogue, psychologist, leisure worker, career counsellor, tandem teacher,…); 2. direct support for pupils (team-building programmes, tutoring, programmes aiming at strengthening cooperation with parents, psychosocial interventions, preventive programmes, breakfast clubs…) ; 3. professional development of the teaching staff (professional development with a focus on inclusive education and innovative methods in education, internships, mentoring, coaching, supervision…). 

The schools most often used the funds that were provided for the services of school assistants for disadvantaged pupils (70% of the schools), for experiential programmes that were supposed to increase pupils’ school motivation (79% of the schools), and for tutoring (73% of the schools). Children at risk of school failure were to receive targeted support that would enable them to master the curriculum of compulsory education. If the support were effective, it should be reflected in improved motivation and academic results of these pupils.

Unfortunately, we cannot verify the progress in achievement, as there is no regular testing of pupils in compulsory education in the Czech Republic (testing takes place irregularly and does not take place in all schools in the same grades). Pupils receive grades for individual subjects (on a scale of 1-5), but these cannot be compared between individual schools. For that reason, indicators constructed from administrative data were used for the evaluation of the support programme. The study seeks an answer to the question of whether the support received by schools under the National Recovery Plan from September 2022 to September 2024 has resulted in a reduction of grade repetition, a reduction of non-completion of compulsory education, and an increase in aspirations for technical or academic track, which entitle students to apply for university studies.

The period after the end of the pandemic was very turbulent in schools. For that reason, it was not possible to directly compare the values of the indicators in 2022 and 2024, as it would have been difficult to distinguish the effects of the intervention and the effects of the development of the education system as a whole. Moreover, between 2022 and 2024, the methodology for applying for upper secondary education changed, so the aspiration indicators in 2022 and 2024 are not directly comparable. In order to find out whether the schools that were supported have undergone a desirable change, we compare them with schools working in similar conditions that did not receive support. We use the propensity score matching method to select comparable sets of schools (e.g. Austin, 2011; Ho et al., 2007). In the Czech Republic, compulsory basic education is provided by the so-called basic schools, which include five years of primary and four years of lower secondary education. The system contains many schools (especially in smaller settlements) that contain only the primary level. Given that the indicators used to evaluate the impact of the support were focused on lower secondary level, only schools with a lower secondary level were selected for comparison. At the same time, 117 schools that started receiving project support in 2023 were eliminated from the data set. The analysis included 2,295 schools, of which 196 were included in the support project in 2022. To identify similar cases, the nearest neighbor matching method was chosen with a ratio of 1:1. Logistic regression was used to determine the distance between cases. The matching was performed using the MatchIT procedure in R using the following variables: the school’s results in national testing in 2022 in mathematics and the Czech language, the proportions of pupils aspiring to academic or technical track, pupils who did not complete compulsory education (finished their compulsory schooling in the 7th or 8th grade), pupils with SEN, pupils with different cultural and living conditions, pupils with recognised support, asylum seekers, and Romas. Matching procedure identified a set of 270 schools (135 control and 135 experimental) with very good agreement in the input parameters. On this set, the values of the output variables were then compared using a t-test: aspirations for technical or academic tracks, non-completion of compulsory education, and repetition of a grade.

After two years of support, the comparison showed statistically significant differences between the supported and unsupported schools in terms of the non-completion of compulsory education and grade repetition, but to the disadvantage of the supported schools. A closer look at the results showed that the situation in the supported schools remained the same, but the situation in the control schools improved. A slight reduction of grade repetition and the early completion of compulsory education was observed in the entire set of basic schools. The aspiration for technical or academic tracks was comparable in both groups of schools. This is probably due to the fact that pupils who are at risk usually finish their compulsory education in a lower grade. Pupils who progress up to the 9th year no longer differ in their aspirations from pupils from other schools. It seems that the supported schools had some specifics that were not captured in the propensity score matching, which may be due to the way schools were selected for the support. The analysis shows that even intensive support was not able to reverse the unfavourable situation in the supported schools. It is also possible that changing the attitudes of teachers, students and parents requires more time than the project has provided. We cannot determine the causes of this condition from the available data. We know from contacts with supported schools that their conditions are continuously changing, with arrivals and departures of Roma families who move from eastern Slovakia or come from UK and whose children therefore do not receive continuous education. It is therefore probably not possible to evaluate the impact of the support using school indicators, but only on the basis of monitoring the progress of individual supported children. However, such data is not available from schools.

Austin, P. C. (2011). An Introduction to Propensity Score Methods for Reducing the Effects of Confounding in Observational Studies. Multivariate Behavioral Research, 46, 399–424. https://doi.org/10.1080/00273171.2011.568786 European Agency for Fundamental Rights (2023). Roma in 10 European countries. Main results. Luxembourg: Publication Office of the European Union. Gersten, R., Haymond, K., Newman-Gonchar, R., Dimino, J., & Jayantha,M. (2020). Meta-analysis of the impact of reading interventions for students in the primary grades. Journal of Research on Educational Effectiveness, 13(2), 401–427. https://doi.org/10.1080/19345747.2019.1689591 Ho, D. E., Imai, K., King, G., & Stuart, E. A. (2007). Matching as Nonparametric Preprocessing for Reducing Model Dependence in Parametric Causal Inference. Political Analysis, 15(3), 199–236. https://doi.org/10.1093/pan/mpl013 Johnson, K.F., Gupta, A., & Rosen, H. (2013). Supplemental Education Services Under No Child Left Behind: Who Signs Up, and What Do They Gain? Mentoring & Tutoring: Partnership in Learning, 21:4, 431-443, doi: 10.1080/13611267.2013.855861 Nickow, A., Oreopoulos, P., & Quan, V. (2024). The Promise of Tutoring for PreK–12 Learning: A Systematic Review and Meta-Analysis of the Experimental Evidence. American Educational Research Journal 61(1), 74–107, doi: 10.3102/00028312231208687.