Resource allocation and student achievement: A microlevel impact study of differential resource inputs on student achievement outcomes.

This study examined the relationships between resource allocation and student achievement using a modified version of a conceptual model designed by Bulcock (1989) within a general model proposed by Guthrie (1988). Five research questions were developed from a review of literature to investigate the relationship between microlevel student input variables and student output variables--both cognitive and affective. The mediating effects of the student perceptions of the quality of school life on student achievement outcomes were also examined. Multiple regression analyses were utilized and data were analyzed at both the individual and school levels. Models were used to investigate the indirect effects of the quality of school life on student achievement outcomes. Substantively meaningful relationships were identified between linguistic resources, language usage and reading outcomes; socioeconomic level, gender, linguistic resources, language usage, and mathematics achievement; gender, student attitudes, and student well-being. All grade eight Newfoundland students (10,146) were the subjects of the study. Participants in the study completed the Canadian Test of Basic Skills (CTBS) and the Bulcock Attitudinal Inventory (BAI). Females scored higher than males on every test of the CTBS and also had more favourable attitudes towards school as measured using the BAI. Urban students outperformed rural students by the equivalent of nearly one year on the CTBS scores. A variable was constructed to test Bernstein's (1961) theory of language discontinuity. Bernstein contended that the further an individual's language code departed from the standard language code in use in that society, the greater the difficulty that person would have in learning. The language code variable was constructed using the language usage score from the CTBS to create a continuous variable. This language code variable proved to be highly explanatory in that it explained a large percentage of the variance in reading achievement outcomes and in mathematics achievement outcomes. The measure for students' perceptions toward their schooling experiences explained a large percentage of the variance of student well-being. Two other noteworthy findings in the present study arose from relationships identified between mathematics achievement and independent variables. A strong relationship was identified between mathematics achievement and socioeconomic level. In general, the higher one's socioeconomic level the greater were the outcome measures in mathematics achievement. Indirect effects analyses produced a significant relationship between gender and mathematics achievement that favoured girls. The construction of the educational production function in the present study proved to be an accurate model. The present study contributed to research in several ways. This is one of the first studies that has employed Quality of School Life indicators as developed in the BAI in an educational production function model. A second contribution was the inclusion of microlevel student linguistic resources as predictors of cognitive achievement outcomes. The third contribution of the present study was the high percentage of variance of cognitive achievement outcomes explained by the modified Bulcock model