In the school year 1994/1995 homework characteristics and effectiveness in math education among secondary schools were compared. The study examined the 1st year of secondary education (average age level: 12.3 years). The sample consisted of 28 schools, 56 classes and 1,394 students. The results show a lack of policy directed at the co-ordination of and increase in the amount of homework in Dutch schools. Nevertheless most math teachers give homework every lesson. Multi-level analysis shows that the amount of homework was the only homework variable related to achievement. It explained 2.4% of the variance. Frequency of homework and homework time were not related to achievement. There is a sharp contrast between Dutch and US teachers in ‘checking and grading’ of homework. However, in this study this variable was not related to math achievement.