For decades, public school educators have claimed that smaller class sizes would significantly improve educational achievement. However, the national data fails to support this premise. Reviewing 152 studies of teacher/pupil ratios, Hanushek found only 27 were statistically significant. Fourteen studies found a positive correlation and 13 studies found a negative correlation between smaller teacher/pupil ratios. 
In our sample, we found that Detroit private schools tend to have significantly smaller class sizes than Detroit Public Schools. Class sizes among private schools tended to range between 14 and 2.5 pupils. The average class size in the Detroit Public Schools in 1989 was approximately 34.
On the other hand, the Boston School System which spends more than $7,000 a year per enrolled student, has 15 students per teacher, yet its dropout rate at 47% was nearly identical to that of the Detroit Public School District.  In sum, class size is not likely to be a significant determinant of educational achievement.