**Author:**
Opdenakker, Marie-Christine, Van Damme, Jan, De Fraine, Bieke, Van Landeghem, Georges, & Onghena, Patrick

**Title:**
*The Effect of Schools and Classes on Mathematics Achievement*

**University Affiliation:**
K.U. Leuven, Belgium

**Email:**
Marie-Christine.Opdenakker@ped.kuleuven.ac.be

**Research Question:**
What are the effects of secondary schools, teachers, and classes on mathematics achievement.

**Published:**
Yes

**Journal Name or Institutional Affiliation:**
School Effectiveness and School Improvement

**Journal Entry:**
Vol. 13, No. 4, pp. 399-427

**Year:**
2002

**Findings:**

- Schools and classes are important , although much of the variation in mathematics achievement is due to individual student characteristics.
- Results indicate that both the school and class/teacher level are relevant for the mathematics achievement of students. Especially the class composition in terms of the level of initial cognitive ability and SES and -to a lesser extent- the learning climate were important for explaining variance at the class and school levels.
- Results report important recruitment differences between schools and classes within schools. The group composition at the class level remained very important for the explanation of differences in mathematics achievement.
- The learning climate in the class, which correlates positively with group composition, explained an additional part of the variance.
- At the school level, the proportion of girls was positively related to mathematics achievement.
- Indications of differential effectiveness of classes and schools and of heteroscedasticity related to SES, average class SES, sex, and learning climate were found.
- Results indicate that ability grouping is often beneficial for talented students. But often harmful for the less able ones.

**Scholarship Type**
Journal Article Empirical Research

**Keywords:**
Academic Achievement, Classroom Composition, Gender, Math, SES

**Regions**
International

**Methodologies:**
Quantitative

**Research Designs:**
N/A

**Method of Analysis:**
Multilevel Models

**Sampling Frame:**
Math classes students

**Sample Types:**
Nonrandom

**Unit of Analysis:**
School, Student

**Data Types:**
Quantitative-Cross Sectional

**Data Description:**

- A subsample of 2,552 students belonging to 150 mathematics classes and to 57 secondary schools of the data from the LOSO-project of Van Damme and colleagues.
- The analysis with the relevant student, class, and school variables included in the model of mathematics achievement is based on a dataset with 1,936 students, 131 classes and 47 schools.
- Analysis included student level and school level variables.
- DV: Mathematics achievement test (end of first grade)
- IV: Student level: initial cognitive ability, family SES, achievement motivation, immunity to stress, sex, language spoken at home; school-level variables: mean initial cognitive ability, mean family SES, mean achievement motivation, mean immunity to stress, proportion of females, proportion speaking Dutch at home

**Relevance:**