The end of the school day increasingly marks not a pause in learning, but a transition into a second educational space shaped by private tutoring. Private tutoring, often referred to as “shadow education”, derives its name from its tendency to mimic the curriculum of formal schools and is typically provided out of regular school hours in academic subjects in exchange for money (Bray, 2014). The popularity of this practice among students and parents, who seek it to compensate for school shortcomings or for enrichment purposes, is often fueled by the assumption that additional instructional time would lead to improved learning outcomes.

In Europe, shadow education has expanded markedly over recent decades, yet with diverse patterns of development across countries (Bray, 2021). Studies examining (non) academic benefits of shadow education in Europe have accumulated over time, yet yielding inconclusive results across countries. Some studies report positive impacts (e.g., Jansen et al., 2020; Neto-Mendes et al., 2013), while others point to negative associations (e.g., Guill et al., 2020; Zumbuehl et al., 2024), and some find no meaningful effects (e.g., Smyth, 2009). Review studies suggest that such inconsistencies can be partly attributed to variations in operational definitions and methodological designs, a pattern further complicated by differences in tutoring attributes such as type, duration, and intensity (Bray, 2014; Luo & Chan, 2022). These insights indicate that understanding benefits of private tutoring requires moving beyond the question of “whether tutoring works” to examining how its specific attributes shape students’ learning experiences in a European context.

In this regard, Bray (2014) argued shadow education is not a homogeneous intervention but a heterogeneous set of practices. It varies in when and how long it is taken, how often lessons occur, who provides the instruction, what is taught, and how strongly students themselves want to participate. In line with Bray’s (2014) observation, this study treats private tutoring not as a uniform intervention but as a set of differentiated educational experiences shaped by specific attributes.

Beyond variation in tutoring attributes, another key source of inconsistent findings in the literature is how educational benefits themselves are conceptualized. Research on private tutoring in Europe and elsewhere has increasingly questioned the adequacy of treating educational benefits as a single, objective outcome. Drawing on prior scholarship, educational benefits are better understood as multidimensional, encompassing not only academic performance but also students’ non-academic aspects, including motivation, confidence, and studying habits (Alam & Zhu, 2022; Benckwitz et al., 2022; Jansen et al., 2020). Reviews of the shadow education literature highlight that tutoring may generate diverse forms of benefits across behavioral and affective domains, which are not always captured by standardized achievement indicators (Luo & Chan, 2022).

Empirical studies have also shown that students and parents often perceive tutoring as highly beneficial even in the absence of measurable gains in academic achievement (Guill & Bos, 2014).From this perspective, students’ own evaluations of tutoring constitute a theoretically meaningful dimension of benefit rather than a mere proxy for achievement outcomes. In this study, perceived benefits refer to students’ own evaluations of how private tutoring supports their academic learning. Against this backdrop, this Czech context-specific study contributes to the European literature on shadow education by examining how different tutoring attributes (e.g., intensity, duration, organizational form) relate to students’ evaluations of its benefits across academic domains. More specifically, it aims to answer following research questions:

1)    How do Czech lower secondary students perceive the shadow education they took and its benefits?

2)    How are the attributes of shadow education (length, duration, intensity, provider etc.) related to students’ perceptions of benefits (before and after controlling for relevant factors)?

Data for this study were collected from Czech senior lower-secondary students during the 2018/2019 school year. A stratified two-stage probability sampling design was employed. In the first stage, lower-secondary schools were randomly selected from six strata defined by school type (regular track vs. academic track) and school size (number of enrolled students). A total of 68 schools were invited to participate, of which 43 agreed, yielding a school-level response rate of 63%. In the second step, senior grade students of lower-secondary education in the selected schools were asked to fill in a paper-and-pencil questionnaires during a personal visit of the members of the research team. The return rate at the second stage of sampling was 88 %, as 1280 of the 1455 eligible students returned the questionnaire and gave usable responses. The questionnaire was adapted from the international survey developed by Silova, Būdiene, and Bray (2006) and previously localized for the Czech context. The present study analyzed a subsample of students (N = 604) who reported having taken private tutoring during lower-secondary education. Students evaluated whether tutoring (1) improved their school grades, (2) enhanced their understanding of the subject, and (3) enabled them to learn something not covered in school; these three binary items served as the dependent variables. Drawing on relevant literature, the questionnaire collected information on key tutoring attributes, including intensity, length, provider, organizational format, and content. Background covariates such as locus of initiation, a composite socioeconomic status (SES) indicator, gender, and subject grades were included as independent variables. Data analysis was conducted in Statistical Package for Social Sciences (SPSS) and R. Descriptive statistics (frequencies, means, standard deviations, standard errors and ranges) were first generated to provide an overview of the sample and address first research question about how students perceived the benefits of private tutoring. To examine second research question about associations between tutoring attributes and perceived benefits, the analysis followed a staged approach. Correlational analyses were first conducted among key predictors to assess interrelationships and potential multicollinearity. Bivariate analyses were then used to explore unadjusted associations between tutoring attributes and perceived benefits and to inform subsequent model specification. Finally, multivariable regression models were estimated to assess adjusted associations. Binary logistic regression was applied to each of the three perceived benefit outcomes, while an ordinal regression model using a composite indicator (number of selected benefits) as dependent was employed to capture overall evaluations of tutoring benefits.

Regarding the first research question, students generally perceived private tutoring as beneficial. Better understanding of subject matter was the most frequently reported benefit (81%, n = 491), while perceived improvements in grades and learning something new not covered in school were reported at similar but lower levels (both around 61%; n = 366 and n = 367, respectively). The mean composite score was 2.03, indicating that students endorsed, on average, two of the three benefit statements. With respect to research question 2, regression analyses indicated that tutoring length and content were the most salient attributes associated with perceived benefits. Compared to students who took private tutoring only short term (≤ 1 month), students who participated in long-term tutoring (≥7 months) were about twice as likely to report improved grades (Exp (B) = 2.17, p < .05) and learning something new not covered in school (Exp (B) = 2.35, p < .05). Content-related attributes showed a more expected pattern of associations, with tutoring aligned with the school curriculum linked to perceived improvements in grades and understanding, and tutoring focused on new content beyond school linked to perceptions of learning something new. Background variables also showed significant associations. For example, students who themselves wanted to take private tutoring were about twice as likely to report learning something new through tutoring compared to those whose tutoring was initiated by others (Exp (B) = 2.06, p < .01). In contrast, no significant associations were found between perceived benefits and tutoring providers or tutoring intensity. These findings suggest that perception of tutoring benefits depends less on participation per se and more on tutoring length and content, underscoring the importance of considering tutoring attributes and student background when evaluating its educational benefits.

Alam, Md. B., & Zhu, Z. (2022). Shadow education and its academic effects in Bangladesh: A vygotskian perspective. Frontiers in Psychology, 13, 922743. https://doi.org/10.3389/fpsyg.2022.922743 Benckwitz, L., Guill, K., Roloff, J., Ömeroğulları, M., & Köller, O. (2022). Investigating the relationship between private tutoring, tutors’ use of an individual frame of reference, reasons for private tutoring, and students’ motivational-affective outcomes. Learning and Individual Differences, 95, 102137. https://doi.org/10.1016/j.lindif.2022.102137 Bray, M. (2014). The impact of shadow education on student academic achievement: Why the research is inconclusive and what can be done about it. Asia Pacific Education Review, 15(3). https://doi.org/10.1007/s12564-014-9326-9 Bray, M. (2021). Shadow education in Europe: Growing prevalence, underlying forces, and policy implications. ECNU Review of Education, 4(3), 442–475. https://doi.org/10.1177/2096531119890142 Guill, K., & Bos, W. (2014). Effectiveness of private tutoring in mathematics with regard to subjective and objective indicators of academic achievement. Journal for Educational Research Online, 6(1), 34–67. Guill, K., Lüdtke, O., & Köller, O. (2020). Assessing the instructional quality of private tutoring and its effects on student outcomes: Analyses from the German National Educational Panel Study. British Journal of Educational Psychology, 90(2), 282–300. https://doi.org/10.1111/bjep.12281 Jansen, D., Elffers, L., & Volman, M. L. L. (2020). A place between school and home: Exploring the place of shadow education in students’ academic lives in the Netherlands. ORBIS SCHOLAE, 14(2), 39–58. https://doi.org/10.14712/23363177.2020.11 Luo, J., & Chan, C. K. Y. (2022). Influences of shadow education on the ecology of education – A review of the literature. Educational Research Review, 36, 100450. https://doi.org/10.1016/j.edurev.2022.100450 Neto-Mendes, A., Costa, J. A., Ventura, A., Azevedo, S., & Gouveia, A. (2013). Private tutoring in Portugal: Patterns and impact at different levels of education. In M. Bray, A. E. Mazawi, & R. G. Sultana (Eds.), Private tutoring across the Mediterranean (pp. 151–165). SensePublishers. https://doi.org/10.1007/978-94-6209-237-2_9 Silova, I., Būdiene, V., & Bray, M. (2006). Education in a hidden marketplace: Monitoring of private tutoring. Open Society Institute. Smyth, E. (2009). Buying your way into college? Private tuition and the transition to higher education in Ireland. Oxford Review of Education, 35(1), 1–22. https://doi.org/10.1080/03054980801981426 Zumbuehl, M., Hof, S., & Wolter, S. C. (2024). Private tutoring and academic achievement in a selective education system. Education Economics, 33(4), 613–630. https://doi.org/10.1080/09645292.2024.2382990