Abstract
Regression surface analysis was used to examine relations between the family environment and measures of academic achievement at different levels of school-related attitudes for 800 11-year-old children from different Australian social groups. The sample included lower social-status families from the following groups: Anglo Australian (250), Greek (170), recent English immigrants (120), and Southern Italian (120). In addition, there were 140 Anglo-Australian middle social-status families. A semi-structured family interview schedule was used to assess parents' aspirations for the child, press for English in the family, satisfaction with the child's school, and satisfaction with teaching in the school. Standardized tests were used to measure performance in mathematics, word knowledge, and word comprehension while an attitude schedule was devised to assess children's affective commitment to school and their academic adjustment to school. The regression surfaces were generated from models that examined possible linear, interaction, and curvilinear relations between the variables. Results indicate that there are differential relations between family environments, attitudes, and children's academic achievement within the ethnic and social-status groups. Tentative support is provided for a theoretical framework which suggests that there can be no change of social reality which is not the common effect of pre-existing social values and individual attitudes acting upon them.