Abstract
The purpose here is to demonstrate a methodology useful in revenue or expenditure models where interaction is believed present. The vehicle is a model of local educational efforts, state/federal subsidies to education, and, most important, the attempt to measure interaction between local effort and subsidies. It is argued that the correct model explaining local educational effort must be reasonably disaggregated (each subsidy or type of subsidy considered separately) and the system must be interdependent (evaluated by simultaneous equation techniques). Results of a model developed and estimated for Ohio school districts indicate that interaction is present in modest proportions at best. In this case, the one-equation model presented by earlier researchers would be accurate, but there is no guarantee that a single-equation model would generate adequate results for other states or even for other types of local expenditures in Ohio.
Get full access to this article
View all access options for this article.
