Collaboration between teachers and families is an important part of a teacher's daily practice. Successful collaboration with families builds strong school-family relationships and positively impacts students' attendance, test scores, and grades (Lasater et al., 2023). However, teachers often hold negative stereotypes – understood as general beliefs about social groups (Eagly & Chaiken, 1993) – against families with low socioeconomic status (SES), which could influence those partnerships. Those negative stereotypes focus on perceived low interest in education and a lack of parental responsibility (Civitillo & Jugert, 2022; Hannon & O’Donnell, 2022). In addition, German preservice teachers differentiate between different low SES by stereotyping welfare recipients more negatively than people from the working poor concerning commitment and sense of responsibility (Yendell et al., 2023). Welfare (German: Buergergeld) is defined in the cited study as benefits mainly targeted at the unemployed. In contrast, the working poor earn less than 60% of the median income without welfare. Moreover, German practicing teachers label parents receiving welfare as having little responsibility for their children's education (Yendell et al., 2024). Also, low-SES students are stereotyped more negatively than higher-SES students. These negative stereotypes include lower academic ability, lower motivation, and poorer social behavior than higher SES students (Shevchuk & Glock, 2022). These results lead to the question of whether teachers justify the school performance of students with different SES (low SES vs. middle SES) via the students themselves or their families and if this justification varies between different low SES (welfare recipients vs. working poor).

A theoretical framework for exploring this question is that of interpersonal stereotypical attributions, which can be used to examine the stereotypical causes teachers give for students' life events (Reyna, 2008). For a stereotypical attribution, an event must first occur, leading to teachers looking for a cause of this event. The (stereotypically) formulated cause can be classified according to the locus of causality, which can be internal (within the student) or external (outside the student). In the school context, it can further be distinguished between student-related, family-related, and school-related causes (Nemer et al., 2019). Additionally, the cause can vary based on the perceived control, where causes may be seen as controllable or uncontrollable by the student (Reyna, 2008).

Stereotypical attributions by teachers are particularly relevant for poor school performance, as an internally controllable attribution (e.g., low motivation) can lead to anger and punitive behavior (Wang & Hall, 2018). Teachers generally tend to attribute poor school performance internally but also see the family background as an external cause (Wang & Hall, 2018). Although a study showed that preservice teachers attribute poor school performance of students with low SES externally, it only recorded the education system as an external dimension (Glock & Kleen, 2020). Therefore, the authors recommend openly recording (stereotypical) attributions to capture the diversity of loci of causality.

Therefore, we use an open-ended survey to compare stereotypical attributions of teachers for the poor school performance of students with different SES:

• RQ1. What causes do teachers give for poor student performance depending on students' SES (welfare receipt vs. working poor vs. medium SES)?

In the second step, we relate these causes to the locus of causality and control:

• RQ2. Does the attributional style of teachers differ between students with different SES regarding the locus of causality (family-related, student-related, and school-related)?
• RQ3. Does the attribution style of teachers differ between students with different SES regarding perceived control?

For research questions 2 and 3, we formulate difference hypotheses between low and medium-SES backgrounds (welfare receipt vs. medium SES, working poor vs. medium SES) and between low-SES backgrounds (welfare receipt vs. working poor).

We developed three vignettes depicting a student with poor school performance but differing family SES. The first vignette featured a family receiving welfare, the second a family belonging to the working poor (cleaning staff), and the third a family with medium SES (administrative staff). Participating teachers read one vignette (between-subjects design) and openly provided up to five causes for the student's poor performance, then identified the most significant cause. Based on the most significant cause, they completed three items each for personal (by the student) and external (by others) control on the 10-point Causal Dimension Scale II (McAuley et al., 1992). For personal control of the selected cause, the extremes of the poles of an exemplary item were formulated as follows: “manageable by the student” and “not manageable by the student.” For an exemplary item on external control, the following formulations were used: “over which others have control” and “over which others have no control.” Teachers from 17 secondary schools in the German "Schule Macht Stark" project participated, with 149 responding fully and passing a manipulation check (50 in welfare, 50 in working poor, and 49 in medium SES condition). The average age was 43.01 years (SD = 9.28), with 14.03 years of teaching experience (SD = 8.78). Of the participants, 28.6% identified as male and 70.1% as female; 21.2% had a migration background. Participants rated their SES on a 10-point scale, averaging 7.33 (SD = 0.93), indicating medium-high SES. Additionally, 16.1% reported prior welfare receipt. Firstly, we conducted a quantitative content analysis of all the mentioned causes, resulting in 11 content categories: Special Educational Needs, Problems within the family, Disadvantageous external circumstances, Inadequate teaching, Irresponsible school-related behavior, Familial irresponsible school-related behavior, General emotional stress, Lack of school-related competencies, Physiological problems, Negative social interaction with classmates, Puberty-related influences. Next, two raters coded all mentioned causes independently to the 11 categories. They achieved an almost perfect agreement (K= .91). Furthermore, they resolved disagreements collaboratively. For the analysis of the most significant reason, we created three deductive overarching categories from the 11 previous ones: student-related, school-related, and family-related attribution (Nemer et al., 2019), plus an inductive category for system-related attribution. For RQ1, we performed Kruskal-Wallis tests on the 11 categories with Bonferroni-corrected post-Hoc tests. For RQ2, we used chi-square tests on the overarching categories, and for RQ3, we performed Kruskal-Wallis tests, again with Bonferroni-corrected post-Hoc tests.

Regarding RQ1, a Kruskal-Wallis test indicated significant differences in “disadvantageous external circumstances”: H(2) = 11.73, p = .003. Teachers cited more causes for the student on welfare (M = 0.74) than for the medium SES student (M = 0.22), z = 2.89, p = .012, r = .29. They also noted more causes for the working poor student (M = 0.72) compared to medium SES, z = 3.05, p = .007, r = .31. In "familial school-related behavior,” the Kruskal-Wallis test confirmed disparities (H(2) = 9.84, p = .007). Teachers attributed more causes to the student on welfare (M = 0.56) than to the medium SES student (M = 0.18), z = 3.08, p = .006, r = .31. We found similar insignificant directional differences between working poor vs. medium SES. For RQ2, the chi-square test for family-related causality was significant, χ²(2) = 7.02, p = .030, φ = .30. Eighteen teachers noted family causes for the student with welfare receipt (working poor: 17) compared to seven for medium SES. Odds ratios suggested a 3.38 times higher chance of family-related attribution for the student with welfare receipt (working poor: 3.01) than for medium SES. Regarding RQ3, The Kruskal-Wallis found significant differences in external control, H(2) = 14.14, p < .001. Teachers perceived more external control for the student with welfare receipt compared to the medium SES student, z = -29.99, p = .002, r = .35. They also reported higher external control for the working poor student versus medium SES, z = -26.12, p = .008, r = .30. We found no differences between low SES groups (welfare vs. working poor). Teachers are more likely to see low-SES families as responsible for their children's poor school performance than medium-SES families, which could provoke anger among teachers and affect teacher-parent cooperation (Lasater et al., 2023; Wang & Hall, 2018). We also discuss how follow-up studies could work on an internationally comparative basis.

Civitillo, S., & Jugert, P. (2022). ‚Sie kümmern sich nicht und haben es eh verdient‘ – Mythen über den Zusammenhang von Armut und Bildung. In G. Steins, B. Spinath, S. Dutke, M. Roth, & M. Limbourg (Eds.), Psychologie in Bildung und Erziehung: Vom Wissen zum Handeln. Mythen, Fehlvorstellungen, Fehlkonzepte und Irrtümer in Schule und Unterricht (pp. 181–196). Springer Fachmedien Wiesbaden. Eagly, A. H., & Chaiken, S. (1993). The psychology of attitudes. Harcourt Brace Jovanovich College Publishers. Glock, S., & Kleen, H. (2020). Preservice teachers’ attitudes, attributions, and stereotypes: Exploring the disadvantages of students from families with low socioeconomic status. Studies in Educational Evaluation, 67, 100929. Hannon, L., & O’Donnell, G. M. (2022). Teachers, parents, and family-school partnerships: emotions, experiences, and advocacy. Journal of Education for Teaching, 48(2), 241–255. Lasater, K., Crowe, T. C., & Pijanowski, J. (2023). Developing Family-School Partnerships in the Midst of Demographic Change: An Examination of Educators’ Attitudes, Values, and Beliefs and the Discourses They Shape. Leadership and Policy in Schools, 22(2), 347–368. McAuley, E., Duncan, T. E., & Russell, D. W. (1992). Measuring Causal Attributions: The Revised Causal Dimension Scale (CDSII). Personality and Social Psychology Bulletin, 18(5), 566–573. Nemer, S. L., Sutherland, K. S., Chow, J. C., & Kunemund, R. L. (2019). A Systematic Literature Review Identifying Dimensions of Teacher Attributions for Challenging Student Behavior. Education and Treatment of Children, 42(4), 557–578. Reyna, C. (2008). Ian is intelligent but Leshaun is lazy: Antecedents and consequences of attributional stereotypes in the classroom. European Journal of Psychology of Education, 23(4), 439–458. Shevchuk, A., & Glock, S. (2022). Pygmalion und die Rolle askriptiver Schüler*innenmerkmale auf Lehrkrafterwartungen und Erwartungseffekte: Damals und heute. In S. Glock (Ed.), Stereotype in der Schule II: Ursachen und Möglichkeiten der Intervention (pp. 1–48). Springer VS. Wang, H., & Hall, N. C. (2018). A Systematic Review of Teachers’ Causal Attributions: Prevalence, Correlates, and Consequences. Frontiers in Psychology, 9, 2305. Yendell, O., Claus, C., Bonefeld, M., & Karst, K. (2023). “I wish I could say, ‘Yeah, both the same’”: Cultural stereotypes and individual differentiations of preservice teachers about different low socioeconomic origins. Social Psychology of Education, 27(3), 777–812. Yendell, O., Claus, C., Budde, J., & Karst, K. (2024). Doing welfare in German secondary schools—How students’ welfare receipt becomes visible in teacher-student interactions and how teachers perceive these students. Zeitschrift Für Bildungsforschung. Advance online publication.