Assessing the impact of expenditure on achievement: Some methodological and policy considerations

The article examines the relationship between achievement and expenditure in primary and secondary public education. The estimation methods are noteworthy in three respects: (1) socioeconomic effects are treated via an “unobservable variables” technique; (2) measures of school district expenditure are valued in constant dollars by constructing locality-level price indices; and (3) we correct for heteroscedasticity, a problem which may be common in estimations using aggregate data. The data are from Virginia, a four-year cross-sectional panel aggregated at the school district level. The results reveal the importance of heteroscedasticity correction and the usefulness of an unobservable variables approach to modeling socioeconomic effects. The results are relatively insensitive to cost-of-living adjustment. Among the policy inferences, we find that expenditure effects on achievement may be stronger at the primary school level.