Fan and colleagues conducted a meta-analysis to review the homework-achievement relationship in math and science given that the effect of subject matters had not been carefully examined.

Among more than two thousand studies related to homework and achievement available on digital databases between 1986 and 2015, 28 studies that fulfilled the inclusion criteria were synthesized. Participants of the included studies were primary school, middle school or high school students. Key findings of those are as follows:

• Overall, there was a small and positive relationship between homework and achievement in math and science.
• The homework-achievement relationships were stronger in the studies operationally defined homework as “homework completion”, “homework grade”, and “homework effort” than the studies defined that as “homework frequency” and “time spent on homework”.
• The relationship for elementary and high school students was stronger than that for middle school students.
• The homework-achievement relationship was the strongest in studies involving US students while it was the weakest in studies involving Asian students. The authors suggested this seemingly counterintuitive result might be attributed to private tutoring that has been weakening the role of formal education in these countries.

The authors remarked that they only included studies written in English and there was an insufficient number of studies for them to consider the moderation effect of some important features, such as gender and complexity of homework assignments.

Source : Fan, H., Xu, J., Cai, Z., He, J., & Fan, X. (2017). Homework and students’ achievement in math and science: A 30-year meta-analysis, 1986–2015. Educational Research Review, 20, 35-54.