Small class sizes for improving student achievement in primary and secondary schools: a systematic review

May 3, 2023Campbell systematic reviews

Small class sizes linked to better student achievement in primary and secondary schools

AI simplified

Abstract

The evidence suggests a small positive effect on reading achievement from reducing class size.

  • A total of 127 studies from 41 countries were included, analyzing various populations of students from kindergarten to grade 12.
  • The overall effect size for reading achievement was positive and statistically significant, with a weighted average of 0.11.
  • In contrast, the effect on mathematics achievement was negative and statistically insignificant, with a weighted average of -0.03.
  • Among studies analyzing data from the STAR experiment, all indicated positive effects for both reading and mathematics, but these were still small.
  • There is a 53% chance that a student in a smaller class will score higher in reading compared to a student in a larger class.
  • The findings suggest that while reducing class size may have some positive impact, the effects are modest and may not justify the associated costs.

AI simplified

Key numbers

0.11
Increase in Reading Achievement
Weighted average effect size from included studies.
-0.03
Decrease in Mathematics Achievement
Weighted average effect size from included studies.
53%
Probability of Benefit in Reading
Probability-of-benefit statistic for reading.

Full Text

What this is

  • This systematic review examines the effects of class size on academic achievement across various studies.
  • It includes 127 studies from 41 countries, focusing on children in grades K-12.
  • The review aims to clarify the impact of smaller class sizes on reading and mathematics performance.

Essence

  • Reducing class size has a small positive effect on reading achievement but no significant effect on mathematics. The evidence suggests that smaller classes may not be cost-effective.

Key takeaways

  • Smaller class sizes lead to a statistically significant positive effect on reading achievement, with an average effect size of 0.11.
  • Mathematics achievement shows a negative effect, with an average effect size of -0.03, which is statistically non-significant.
  • The STAR experiment results indicate a positive effect for both reading and mathematics, but the overall findings suggest only modest benefits from class size reductions.

Caveats

  • The review includes studies with a high risk of bias, limiting the reliability of the findings. Only a small number of studies contributed to the meta-analysis.
  • The geographical coverage is narrow, with most studies from the USA, France, and the Netherlands, potentially limiting generalizability.

Definitions

  • Standardised Mean Difference (SMD): A statistical measure used to quantify the effect size across studies, indicating the difference in means relative to the standard deviation.

AI simplified

what lands in your inbox each week:

  • 📚7 fresh studies
  • 📝plain-language summaries
  • direct links to original studies
  • 🏅top journal indicators
  • 📅weekly delivery
  • 🧘‍♂️always free