The results of my meta-analysis for teacher student relationships are reported in Figure 4.1. What is perhaps most striking about the findings is not the average effect size of –.869
(which is sizable and impressive in itself), but the extremely large effect sizes for the middle school/junior high level and the upper elementary level. (Note that data were not available to compute an average effect size for the primary level.) I should caution that if more studies of the impact of teacher-student relationships on student behavior were available, these large average effect sizes would probably go down. They are simply much larger than are commonly found in the social sciences. Nevertheless, it makes sense that teacher-student relationships would be extremely important to students below the high school level and, therefore, would have a profound impact on student behavior.