Characteristics of education production functions: An application of canonical regression analysis

This paper uses data from the State of Michigan and canonical regression analysis to investigate the effects of socioeconomic characteristics (SEC) of communities in the production of high school education. We find that SEC have positive and significant impacts - impacts that are independent of school resources - on the output of education. However, these independent effects are very hard to ascertain because of the high degree of correlation between SEC and school resources. We also find that education of parents is the only variable that can be used as a proxy for all SEC without misspecifying the education production function.