Peer Effects in Urban Schools: Assessing the Impact of Classroom Composition on Student Achievement
University of California, Santa Barbara
This study evaluates the effects of classroom peers on standardized testing achievement for all third- and fourth-grade students in the Philadelphia School District over 6 school years.
Journal Name or Institutional Affiliation:
Vol. 28 No. 5 pp.607-647
- Students who have repeated the current grade have a higher reading score in the next academic year whereas being young for the grade has no significant relationship with achievement.
- Lagged behavior grades are significant and positive the higher last year’s behavior grade, the higher the next year’s achievement score.
- Once being a free lunch student is accounted for (which has a coefficient that is significant and negative), neighborhood characteristics are not statistically significant in determining student achievement. Rather, it is the student’s individual characteristics and school environment that affect achievement outcomes.
- The classroom proportion of students receiving free lunch is negative and significantly related to student achievement. The count of students with behavior problems is also negative and significant, indicating that students with behavior problems detract from student achievement. An increase in the number of girls in a classroom raises individual student achievement.
- There are significant peer effects in reading and math standardized test achievement, even after holding constant student and neighborhood demographics, and teacher characteristics.
Journal Article Empirical Research
Academic Achievement, Elementary School, Gender, Peer Effects, SES Composition, Urban Schools
Secondary Survey Data
Method of Analysis:
ANOVA, Logistic Regression
3rd and 4th grade students in Philadelphia
Unit of Analysis:
Educator, Neighborhood, School, Student
- Student and teacher data were obtained from the School District of Philadelphia via the district’s office of student records and through the district’s personnel office. Neighborhood data were obtained from the Census flat files at the census residential block level.
- A measure of a standardized student achievement score including reading, math, lagged (last years) reading lagged (last years) math scores and late arrivals into classroom.
- School race in percentage (White, Black, Hispanic, Asian, other), gender percentage of school (male, female), academic variables in percentage (attended Philadelphia kindergarten, free lunch eligible, English language learner, special education, lagged (previous years) behavior grade), Students census block percentage (white, poverty, percent vacant houses), log of income (in dollars) of student’s census block. teacher race in percent (White, Black, Hispanic, Asian, other), teacher gender in percentage (male and female ), teacher skills (teacher experience in years, teacher state certified percent, teacher has masters degree percent), class size (head count), academic classroom characteristics (mean reading score, mean math score), other classroom characteristics head count (free lunch, behavior problems, English language learners, special education, female)