Quality teaching, not class size, raises student achievement

Diverse Elementary ClassYet another study has found that despite the enormous appeal of small classes among parents and educators, small class size seems to have no measurable benefits for students.

Two researchers from the University of Wisconsin at Madison conclude in a new study of kindergartners published in Educational Evaluation and Policy Analysis that they found “no evidence that class size affects reading or mathematics achievement in kindergarten.” However, they did find significant effects on student achievement from classroom instruction.

“Our results thus resonate with other studies that indicate that what happens inside classrooms matters much more than the structure in which those activities occur,” say University of Wisconsin at Madison researchers Carolina Milesi and Adam Gamoran.

Data on 21,260 children

To re-examine the issue of class size, the two researchers analyzed data on 21,260 children enrolled in approximately 1,000 kindergarten programs during the 1998-1999 school year. The data is from the Early Childhood Longitudinal Study–Kindergarten Class of 1998-99 (ECLS-K), the only nationally representative sample which assesses children’s status at kindergarten entrance, the researchers say. Besides class size, researchers analyzed data on class organization, teachers’ time allocation and type of instructional activities.

In reading and mathematics, the researchers found that both instructional orientations–whole language and phonics in reading and “teaching for understanding” and drill in mathematics–contributed to performance.

“In reading, we found that additional time spent on both whole class and small-group activities boosted achievement, a result that is consistent with prior research showing the value of instructional time in early reading,” the researchers say.

The results from this study are consistent with the emphasis on teacher quality under No Child Left Behind, the researchers note.

In California, one of the negative effects of a class-size reduction initiative was a drop in fully accredited teachers, the researchers note, a drop that disproportionately affected disadvantaged schools.  To meet the goal of having no more than 20 students in K-3 classes, schools hired less qualified teachers. The proportion of fully accredited teachers dropped by an average of 2% statewide while it dropped by 20% for the schools serving the most low-income students, the researchers say.

STAR study results

The major study providing evidence for small class size was an experimental study conducted in the 1980s in Tennessee called Student/Teacher Achievement Ratio (STAR). Other research has looked at whether there are class-size effects that impact achievement. For example, do teachers teach differently in small and large classes? Are certain instructional practices more effective in small classes? Do students behave differently in small classes?  The authors  did not find evidence of these class-size effects in their study, nor could they identify groups of students that seemed to benefit from small classes.

The results from STAR do not seem to have generalized to real-world settings.  An important consideration, the researchers say, is the other schooling conditions that occur with class size.  “More research is needed to identify the conditions under which experimental findings, such as Project STAR, can be generalized.”


“Effects of Class Size and Instruction on Kindergarten Achievement”, by Carolina Milesi and Adam Gamoran. Educational Evaluation and Policy Analysis, Winter 2006, Volume 28, Number 4, pp. 287-313.


Published in ERN March 2007 Volume 20 Number 3

Leave a Reply

  • (will not be published)