Achievement of Children in Orphan Homes as Revealed by the Stanford Achievement Test

Regression surface analysis was used to examine relationships between teacher perceptions of school related behavior and children's academic achievement at different levels of intelligence, social status, and family environment. Data were collected as part of a national survey of 11-, 12-, and 15-year-old English schoolchildren (N = approximately 3000). The regression surfaces were constructed from models which examined possible linear, curvilinear, and interaction relations between the variables. In the analysis the Jackknife technique was used to adjust the significance levels. Two propositions are suggested by the results that (a) at each level of intelligence and social environment, changes in the teacher perceptions of student behavior are associated with modest changes in academic achievement, and (b) at each level of teacher perceptions, increases in intelligence and social environment scores are related to sizable increments in academic achievement.