Students' behavioural problems are one of the greatest challenges facing teachers. At the same time, they have far-reaching consequences: There is a close connection between "deviant behaviour" in school, academic performance and later delinquent behaviour in adult live. According to these authors, an important driver is the disciplinary practice of teachers: the stricter and harsher it is, the more likely it is that deviant behaviour will intensify (Amemiya et al. 2020) and academic performance will decline (Del Toro & Wang 2022). This fact deserves attention because the disciplinary practices of teachers can vary greatly with comparable student behaviour. Although the nature of the behaviour itself has an influence on disciplinary responses, this is also influenced by contextual characteristics such as class composition (Rocque & Paternoster 2011; Payne & Welch, 2010; Welch & Payne, 2018). Our main interest in the present study is the question of how student behaviour develops over time under the influence of teacher behaviour and student composition.

Empirical studies on this question are generally rare, especially for primary schools, and only relate to partial aspects of the question. The available studies allow the following assessment: a) Deviant behaviour is higher in schools with a less privileged student body than in schools with a privileged student body, and this gap widens over time. B) Harsh punishments such as exclusion from school and lessons or a referral to the principal by teachers increase with the proportion of students from non-privileged families, even when controlling for student behaviour (Payne & Welch, 2010; Welch & Payne, 2018; Roque & Paternoster 2011 and Kinsler 2011 for primary school). C) Aggressive student behaviour occurs more frequently when minor deviant behaviour has previously been punished by official documentation (Amemiya et al., 2020). However, Amemiya et al. (2020) did not investigate the contextual influences on this development. These results suggest that deviant student behaviour is increasing as a result of harsher disciplinary practices by teachers, especially in schools with non-privileged students compared to schools with privileged students. However, as yet there is a lack of empirical evidence for this.

For a theoretical approach labelling theory can be used. According to this approach, teachers have to interpret behaviour in order to classify it as a rule violation and react to it. Primary school teachers, for example, react very differently to deviant behaviour, even to serious violations of school rules (Psunder, 2005; Skiba, 1997), ranging from ignoring to mild and harsh punishments. Punishments carried out in the presence of classmates can take on the function of labels. This means that a punishment labels a pupil as "deviant" and the following interactions refer to this label. This can result in teachers imposing increasingly strict sanctions (Bowditch 1993) and pupils describing themselves as deviant (Chiricos et al., 2007) and being more likely to join deviant groups of pupils (Bernburg et al. 2006). These effects of labelling lead to an increase in deviant behaviour over time. Although not explicitly modelled theoretically in the labelling approach, labelling processes could occur more frequently in schools with a less privileged student body. Teachers might expect more intense deviant behaviour in these schools, see their (cultural) authority at risk and believe they have to assert it through harsh punishments (Payne & Welch, 2010; Welch & Payne, 2010; Rocque and Paternoster, 2011). With this in mind, we expect the following: Harsher disciplinary practice is more common in schools with low-privileged compared to schools with privileged student bodies for the same student behaviour, which subsequently leads to higher proportions of deviant behaviour by students in schools with low-privileged compared to schools with privileged student bodies.

Methodology: Procedure: The sample consisted of 14 first grade classes at 14 primary schools in Germany. The deviant behaviour of the pupils and the reactions of the teachers were recorded by means of classroom observations during the first grade. These were carried out on four survey dates each. A total of 7,892 behavioural units of 314 pupils and 3085 disciplinary reactions of teachers were observed. The deviant behaviour was recorded in units of minutes per lesson (e.g. Amina runs through the class). Teacher behaviour (e.g. "Amina, stop it") was also recorded. Two trained observers were always present in a class. The material was then coded. A coding scheme was developed for coding deviant pupil behaviour, which was primarily based on the already validated instrument "Behavioural Observation of Students in Schools" (Shapiro, 2011). Student behaviour was assigned to the following categories: motor (unauthorised movement around the room or in the square), verbal (unauthorised utterances such as shouting, chatting), aggression (non-physical: insults, verbal abuse; physical aggression: destruction of objects, physical attacks on people) and passivity (passive inattention), other. Teacher behaviour in response to deviant pupil behaviour was also recorded. An inductive procedure was used to differentiate between 22 categories (e.g. "no reaction", "non-verbal reaction", "exclusion from ongoing lessons"). "Intensive" disciplining was coded if students they were excluded from lessons or classroom activities or punished by the removal of (sometimes symbolic) privileges. Variables: A sum score was calculated for each student per survey from the deviant behaviours, which represents the dependent variable. Furthermore, the individual proportion of general teacher reactions to individual deviant behaviour of a pupil and the proportion of "intensive disciplinary" teacher reactions to individual deviant behaviour were calculated for each wave. Teacher information was also available on how high the proportion of pupils from low-privileged families was at a school. A distinction was made between schools with a high (60-90%, N = 4 schools) and low (0-40%, n = 10 schools) proportion of low-privileged pupils. Method: Mixed-effects linear regressions were calculated taking into account the multi-level structure (behaviour nested in pupils nested in classes). The sum score of student behaviour at time tn was regressed on the individual proportion of intensive disciplinary teacher responses at time tn-1, controlling for individual student behaviour and the individual proportion of all teacher responses at time tn-1. Furthermore, interaction effects between the intensive disciplining practice and the student composition were calculated.

The results show that already at the beginning of primary school, the proportion of deviant behaviour is higher in schools with higher proportions of low-privileged pupils than in schools with higher proportions of privileged pupils; this applies in particular to motor and verbal deviant behaviour. Similarly, even when controlling for the type and frequency of pupil behaviour, the intensively punitive teacher reactions are more frequent at the former schools. However, according to preliminary evaluations, there are no indications that a) intensively punishing teacher behaviour increases the frequency of deviant student behaviour and b) that such a development is more likely to occur in schools with a less privileged student body than in schools with a privileged student body. However, there are indications that the more likely a teacher intensely disciplines a pupil's disruptive behaviour, the higher the individual share of disruptive behaviour, but not the frequency, and this development is more likely to occur with pupils who are already more disruptive to begin with and at schools with a less privileged pupil body than at schools with a privileged pupil body. This means that many pupils adapt to the behavioural expectations in the course of the first year and show more compliant behaviour, while pupils with higher initial values tend to maintain their behaviour in response to an intensive disciplinary practice and thus take up a higher proportion of the disruptive behaviour overall. This development in the course of the first class does not indicate either the theoretically expected general dynamic or a context-specific dynamic of labelling processes, as the quality of the behaviour does not change significantly either. The extent to which these developments will unfold in the further course of primary school will have to be shown by further surveys in the coming school years.

Amemiya, J., Mortenson, E. & Wang, M. (2020). Minor infractions are not minor: school infractions for minor misconduct may increase adolescents’ defiant behavior and contribute to racial disparities in school discipline. American Psychologist, 75(1), 23–36. https://doi.org/10.1037/amp0000475 Bernburg, J. G., Krohn, M. D., & Rivera, C. J. (2006). Official labeling, criminal embeddedness, and subsequent delinquency: A longitudinal test of labeling theory. Journal of Research in Crime and Delinquency, 43(1), 67–88. https://doi.org/10.1177/0022427805280068 Bowditch, C. (1993). Getting rid of troublemakers: High school disciplinary procedures and the production of dropouts. Social Problems, 40(4), 493–509. https://doi.org/10.2307/3096864 Chiricos, T., Barrick, K., Bales, W., & Bontrager, S. (2007). The labeling of convicted felons and its consequences for recidivism. Criminology, 45(3), 547–581. Kinsler, J. (2011). Understanding the black–white school discipline gap. Economics of Education Review, 30(6), 1370–1383. https://doi.org/10.1016/j.econedurev.2011.07.004 Payne, A. A., & Welch, K. (2010). Modeling the effects of racial threat on punitive and restorative school discipline practices. Criminology: An Interdisciplinary Journal, 48(4), 1019–1062. https://doi.org/10.1111/j.1745-9125.2010.00211.x Psunder, M. P. (2005). How effective is school discipline in preparing students to become responsible citizens? Slovenian teachers’ and students’ views. Teaching and Teacher Education, 21(3), 273–286. https://doi.org/10.1016/j.tate.2005.01.005 Rocque, M., & Paternoster, R. (2011). Understanding the antecedents of the „school-to-jail“ link: The relationship between race and school discipline. The Journal of Criminal Law and Criminology, 101(2), 633–666. Shapiro, E. S. (2010). Academic skills problems fourth edition workbook. Guilford Press. Skiba, R. J., Peterson, R. L., & Williams, T. (1997). Office referrals and suspension: disciplinary intervention in middle schools. Education and Treatment of Children, 20(3), 295–315. http://www.jstor.org/stable/42900491 Del Toro, J. & Wang, M. (2022). The roles of suspensions for minor infractions and school climate in predicting academic performance among adolescents. American Psychologist, 77(2), 173–185. https://doi.org/10.1037/amp0000854 Welch, K., & Payne, A. A. (2010). Racial threat and punitive school discipline. Social Problems, 57(1), 25–48. https://doi.org/10.1525/sp.2010.57.1.25 Welch, K. & Payne, A. A. (2018). Latino/a Student threat and school disciplinary policies and practices. Sociology of Education, 91(2), 91–110. https://doi.org/10.1177/0038040718757720