Bates, Littisha A., & Glick, Jennifer E.
Does It Matter If Teachers and Schools Match the Students? Racial and Ethnic Disparities in Problem Behaviors
University of Cinncinnati
The present study asks whether tacher-student racial/ethnic mathces result in evaluations of student behaviors that are different from instances in which children are taught and assessed by a teacher from outside their racial or ethnic group.
Journal Name or Institutional Affiliation:
Social Science Research
Vol.42 Pp. 1180-1190
- The analyses presented here identify the extent to which children receive different evaluations from their teachers depending on the racial/ethnic match of teachers and students.
This study is distinct from previous work because they examine the assessment of an individual child by multiple teachers.
The results indicate that Black children receive worse assessments of their externalizing behaviors (e.g. arguing in class and disrupting instruction) when they have a non-Hispanic white teacher than when they have a Black teacher.
Further, these results exist net of school context and the teacher’s own ratings of the behavior of the class overall.
Overall, minority students are more likely to attend public schools and schools that receive Title I funding than non-Hispanic white students. Minority students are also more likely than their non-Hispanic white counterparts to attend schools with a higher percentage of minority teachers.
Minority students (with the exception of Asians) are the ones most likely to be rated as having more externalizing behaviors.
Children who attend schools that receive Title I funding receive poorer behavior ratings than their counterparts in schools that do not receive these funds.
- The results indicate there are important differences in the behavioral ratings children receive based on their ascribed characteristics.
There are persistent racial/ethnic differences in the ratings of student’s behaviors and, generally speaking, teachers’ ratings tend to be consistent with the societal stereotypes associated with the racial and ethnic groups when rating students’ externalizing behavior.
Black students are more likely to be rated as exhibiting more externalizing or problematic behaviors in school, while Asian students are perceived as exhibiting fewer of these behaviors.
In a model distinguished by its multi-racial emphasis, student’s blackness is still shown to be a significant predictor of unfavorable teacher-perceptions.
However, these results also indicate that students do not receive the same ratings from all teachers. If teachers are of the same racial/ethnic group as the student, the ratings are less consistent with these expectations.
Teacher-student racial congruence is again highlighted as a contextual factor helping to counter balance stereotypes.
Black students suffer most as they are stereotyped most negatively, and are the least likely to have a minority teacher who might be able to sympathize with their plight and look past these stereotypes when assessing the student’s behavior.
These differences persist even when they control for classroom and school characteristics.
School type is significantly associated with the behavioral assessments children receive. Students in private schools tend to receive less positive behavioral assessments compared to their peers in private religious and public schools.
Minority children in majority-dominated schools are the most likely to experience this racial mismatch and to be rated less favorably by their teachers. Thus, the effects of teacher-student mismatch presented here are disproportionately experienced by minority children nationwide.
Journal Article Empirical Research
Behavior, Classroom Composition, Discipline, SES Composition, Student Characteristics
Method of Analysis:
Descriptive Statistics, Multilevel Models
Kindergarten Cohort of 1998/1999
Unit of Analysis:
Educator, School, Student
They used multiple waves of the Early Childhood Longitudinal Study Kindergarten Cohort of 1998-1999 (ECLS-K). The ECLS-K begins with a nationally representative sample of children who entered Kindergarten in the U.S. between 1998 and 1999.
To account for student’s grade level, teachers are asked different questions across the waves. For example, Kindergarten teachers are asked questions to reflect disruptive behavior in Kindergarten, such as "does the child disrupt nap time?" Fifth grade teachers, in contrast, are asked about students disrupting reading time.
The sample consists of 16,701 children ( non-Hispanic white = 10.366, Hispanic White = 1,640, black = 3,021 and Asain = 1,674) Because there are four observations per child, the effective sample size is 66,804
The dependent variable is reported student behavior. The assessments that make up their dependent variable, externalizing behavior, come from the Teacher Social Rating Scale (SRS), in which teachers were asked to report the frequency of students’ externalizing behavior. Externalizing behaviors are those behaviors often perceived as ‘‘acting out’’ behaviors which may interrupt classroom activities (Tourangeau et al., 2006). This problem behavior scale was adapted from Gresham and Elliott’s (1990) Elementary Scale A: ‘‘How Often?’’. Teachers were asked to report on a scale of one (not often) to four (very often) the frequency with which individual children exhibited five behaviors: child argues, fights, gets angry, acts impulsively, and disturbs ongoing activities (Tourangeau et al., 2006).
Their primary independent variable is whether or not the teacher and the child being assessed are of the same race or ethnicity.
Other independent variables were school context, teacher ratings of the class as a whole, and teachers level of education.
School characteristics inlcuded; School type (private school, private religious school, or public school), overall racial/ethnic composition of the school (measured as a dummy variable- if the school is composed of over 50% or more minority students) , and weather the school recieves title 1 funding.
They also controlled for child characteristics and family background characteristics including; child’s gender (males are the reference group) and age, family structure (step family, single parent family, some other family form and two biological parent family (reference group)), family income (measured in quintiles, the fifth quintile is used as the reference group), home language (English is the reference group) and a parental rating of the child’s behavior (that higher scores reflect less control).