A1255
Title: Nonlinearity of spending effects on public education: Evidence of spatial dependence
Authors: Shengwu Shang - Kansas State University (United States) [presenting]
John Anderson - University of Nebraska-Lincoln (United States)
Juan Du - Kansas State University (United States)
Abstract: The effects of fiscal expenditures on educational outcomes in U.S. public education are studied, with particular attention to spatial dependence and nonlinearities. Using data from the 2010 Michigan educational assessment program (MEAP) on fourth-grade mathematics pass rates, the analysis replicates earlier findings on spending effects following Michigan's proposal a reforms and extends them by incorporating spatial dependence through the nonparametric covariance estimator. While prior research reports little difference between linear and nonlinear models in terms of average partial effects, it is shown that linear specifications fail to capture spatial interactions. Accounting for spatial dependence reveals meaningful nonlinear spending effects, and estimates from linear models using level pass rates overstate these effects by approximately 4-6\%.
