Are Education Cost Functions Ready for Prime Time? An Examination of Their Validity and Reliability

Abstract

This article makes the case that cost functions are the best available methodology for ensuring consistency between a state's educational accountability system and its education finance system. Because they are based on historical data and well-known statistical methods, cost functions are a particularly flexible and low-cost way to forecast what each school district must spend to meet the standards in a state's accountability system. However, the application of cost functions to education must confront several challenges in both data collection and estimation methodology. This article describes the strengths and weaknesses of various ways to address these challenges and illustrates how the reliability and forecasting accuracy of cost functions can be tested using data for Missouri school districts.Correspondence should be sent to John Yinger

The effect of charter schools on districts’ student composition, costs, and efficiency: The case of New York state

“…Because not all districts operate efficiently and we cannot directly observe which districts are and are not efficient, costs cannot be observed directly and cost functions have to be estimated using district spending as the dependent variable and indirect controls for efficiency. Spending is greater than costs when school districts are inefficient and deviate from current best practices (Downes & Pogue, 1994;Duncombe & Yinger, 1998, 2005, 2008, 2011a, 2011bReschovsky & Imazeki, 2001.…”

“…To estimate the effect of changes in enrollment and student composition on costs, we draw on the cost function literature (Downes & Pogue 1994;Duncombe & Yinger 1998, 2005, 2008, 2011a, 2011bEom, Duncombe, Nguyen-Hoang, & Yinger, 2014;Reschovsky & Imazeki 2001. Particularly, we estimate the following expenditure function:…”

“…Unlike the other analyses in this paper, which use district fixed effects to control for district factors that remain constant overtime and only districts in the treatment and control groups defined above, we estimate Equation (2) using cross sectional information for the 1989/99 school year and a sample that includes all districts in New York State outside of New York City that have an enrollment of at least 2000 students. It is difficult to estimate the effect of many cost factors on expenditures in panel data models that include district fixed effects because too much of the variation in these factors is differenced away by fixed effects estimators (Duncombe, Nguyen-Hoang, & Yinger, 2015;Duncombe & Yinger, 2011b). In addition, estimating equation (2) with only 28 school districts results in imprecise parameter estimates.…”

“…12 District enrollment counts are also drawn from NYSED Report Cards and represent official counts of students registered in the district as of October 1 of each school year. As in other work on cost functions, we use the log of student enrollment and the log of enrollment squared to allow for a nonlinear relationship between per pupil expenditures and enrollment (Duncombe & Yinger, 2008, 2011a, 2011b. We also include the percent of students eligible for free or reduced price lunch, students with limited observed new cut score, X NEW , and the associated proficiency rate.…”

Incentives and the Supply of Effective Charter Schools

“…Cost estimation in education settings is subject to a number of endogeneity concerns (Costrell, Hanushek, and Loeb 2008;Duncombe and Yinger 2011;Gronberg, Jansen, and Taylor 2011). For example, effectiveness or quality of education is often poorly measured due to nonrandom sorting of students across schools.…”