This study synthesizes findings from 888 articles published in peer-reviewed journals or by the National Bureau of Economic Research since 1990. We focus on studies that explore relationships between inputs or policies (such as per-pupil expenditure, average teacher salary, class size, performance pay, and school choice) and student outcomes (such as test scores, dropout rates, and college attendance).
In our survey, we found 116 studies that explored the link between per-pupil spending and student outcomes. Higher spending was associated with higher student performance only 32 percent of the time, after adjusting for student background and other factors mostly or entirely outside the control of schools. Similarly, of the 90 studies examining the link between average teacher salaries and student outcomes, 56 percent found mixed or statistically insignificant results.
Of the 132 studies probing the relationship between years of teaching experience and effectiveness in the classroom, only 41 percent found a positive effect. The results were far worse for teachers with graduate degrees, a variable included in 114 studies. There was a positive relationship with student outcomes only 16 percent of the time.
The available evidence on performance pay shows promise. Of 34 studies examining teacher incentives for individual performance, school-wide performance, or both, 61 percent found statistically significant gains in student outcomes. The design of performance-pay models seems to matter a great deal, however.
Based on these findings, we can say that in recent years North Carolina has been moving in the right direction on school reform, enacting policies that will likely improve the efficiency and efficacy of our education system in the coming years. Policymakers should continue the momentum and resist attempts to backtrack from the significant accomplishments already achieved.