Measuring the Impact of Lotteries on State per Pupil Expenditures for Education: Assessing the National Evidence

Abstract

State-operated lotteries have recently been asserted by public administrators and academicians as panaceas for eradicating revenue disparities existing across public school districts in the American states. The purpose of this research project is to empirically test the hypothesis that lottery revenues raise the state expenditures for public education. A state-level national dataset, which includes fifty American states over the period 1977-1997, was used for the analysis. Pooled time-series cross-sectional and ARIMA modeling was employed to test the hypothesis. This study finds that lottery revenues had a positive influence on state per pupil expenditures for education. The evidence for the impact of lotteries on state per pupil expenditures for education was robust and statistically significant. Copyright 2005 by The Policy Studies Association..

Using intervention-oriented evaluation to diagnose and correct students’ persistent climate change misconceptions: A Singapore case study

“…Its usefulness had also been recognized in special education studies (Horner et al, 2005;Marston, 1988), in measuring teaching effectiveness (Lin & Lawrenz, 1999), and self-regulated learning (Schmitz & Wiese, 2006). Time-series is preferred in longitudinal inquiries such as in the assessment of the effectiveness of school reforms (Bloom, 2003;May & Supovitz, 2006), the effect of financial aid to educational quality (Henry & Rubenstein, 2002) and the impact of lotteries to state expenditure for education (Moon, Stanley, & Shin, 2006). Ortega and Iberri-Shea (2005) posit that time-series ''constitutes the single best formal strategy for investigating effects of instruction longitudinally'' (p. 33) in their meta-analysis of approaches used for second-language acquisition.…”

Randomized Experiments and Quasi-Experimental Designs in Educational Research

“…However, if time series are long enough and if the required assumptions hold, then interrupted time series designs are among the strongest quasi-experimental designs. Nonetheless, time series designs in educational evaluations are rare and typically not long (Henry & Rubenstein, 2002;Kearney & Kim, 1990;Lin & Lawrenz, 1999;May & Supovitz, 2006;Moon, Stanley, & Shin, 2005).…”