Monetary and Nonmonetary Student Incentives for Tutoring Services: A Randomized Controlled Trial

Abstract

In recent years, the largely punitive accountability measures imposed by the 2001 No Child Left Behind Act have given way to an emphasis on financial incentives. Although most policy interventions have focused primarily on linking teacher compensation to student test scores, several recent studies have examined the prospects for the use of financial incentives for students to improve student outcomes. This study builds on the existing literature by comparing student responses to monetary and nonmonetary (certificates of recognition) incentives to increase participation in federally funded supplemental educational services (SEdS). Approximately 300 eligible middle grade students who had signed up for SEdS were randomly selected and then assigned to either a control condition that received no incentive, a monetary reward group that could earn up to $100 for regular attendance, or a nonmonetary group that could receive a certificate of recognition signed by the district superintendent. Although the benefits of the monetary incentives were negligible, the students in the certificate group attended 42.5% more of their allotted tutoring hours than those assigned to control. The effect of the certificate was particularly strong for female students, who attended 26% more of their allocated tutoring hours compared to males who were also offered certificates. These results suggest the need for further research into the role of nonmonetary incentives in motivating student behaviors. Also, the findings could be useful to policymakers at the state or district level seeking cost-effective mechanisms to increase uptake of underutilized student supports.

Keywords:

• Financial incentives
• experiment
• accountability

ACKNOWLEDGMENTS

The authors are grateful to Dale Ballou, Chris Hulleman, Matthew Pepper, Morgan Polikoff, and Ron Zimmer for helpful comments and insight in developing this work. They also wish to acknowledge seminar participants at the University of Colorado–Colorado Springs, Vanderbilt University's Peabody College, and the annual meetings of the American Educational Research Association, Association for Public Policy Analysis and Management, Association for Education Finance and Policy, and U.S. Department of Education's Institute for Education Sciences. The usual disclaimers apply.

Notes

Studies finding positive impacts in mathematics and reading include Rickles and Barnhart (2007), Springer, Pepper, and Gosh-Dastidar (2014), Zimmer, Christina, Hamilton, and Prine (2006), and Zimmer et al. (2007). Studies with findings of mixed impacts include Heistad (2005) and Rickles and White (2006). Deke, Dragoset, Bogen, and Gill (2012), Heinrich, Meyer, and Whitten (2010), and Potter et al. (2007) had null findings.

In Chicago, the incentives had a positive, statistically significant impact on grade-point averages and student attendance.

Assuming a mean difference of.125 percentage points with a standard deviation of.315, 304 participants were needed to provide power of.801. See Lipsey (1990) for a detailed discussion on design sensitivity.

Figure 1 Consort diagram.

We randomized eligible participants at the individual student level blocked by timing for their signing up for tutoring services. We used a simple randomization procedure blocked by enrollment date because, as Bruhn and McKenzie (2009) demonstrate, different randomization methods (e.g., pairwise matching, stratification) perform similarly in populations of 300 or more.

Students are allotted different hours of tutoring because providers of SEdS can charge different hourly rates. Tutoring providers invoice the school district for the number of hours students attend, up to a maximum per-student, per-year dollar allocation.

We acknowledge that the assumption that students who are identified as needing special education services (or ELL services) would react the same to incentives is a broad one and most likely dependent on a wide variety of factors, which are outside the scope of this study. However, not surprisingly, given the small numbers in each treatment group, we are unable to detect any significant differential responses to the various incentives for these students. We also estimated models without special education and ELL populations and results are qualitatively similar.

Table 1 Summary statistics on student characteristics

The supplementary online materials contain a comparison of respondents and nonrespondents on observable characteristics.

As noted earlier, the supplementary online materials contains a comparison of the full and restricted survey samples.

Prior research suggests that students would prefer that different audiences receive different signals. Relative to peers, adolescents’ parents place greater value on time spent on homework (Fordham, 1996), are less tolerant of misbehavior in class (Berndt, Miller, & Park, 1989), and place different relative values on reputation, popularity, and academic success (Coleman, 1961). It is likely then that students know or perceive their parents to value academic achievement and behavior—including tutoring attendance—more than students’ peer groups. For this reason, incentives that aim to generate recognition or pride from a student's parents might be more effective than those awarded before students’ peers.