What 20 Years of MDRC RCTs Suggest About Predictive Relationships Between Intervention Features and Intervention Impacts for Community College Students

Abstract

This article provides evidence about predictive relationships between features of community college interventions and their impacts on student progress. This evidence is based on analyses of student-level data from large-scale randomized trials of 39 (mostly) community college interventions. Specifically, the evidence consistently indicates that impacts tend to be larger as the number of intervention components increases and with the degree to which interventions promote full-time enrollment in fall and spring and/or summer enrollment. Less consistent evidence suggests that impacts tend to be larger for interventions that increase advising, tutoring, and, to a lesser extent, financial support. These results provide food for thought about the design of future community college innovations to be tested by researchers or tried by practitioners.

Report on the Implementation, Costs, and Impacts of the Findhelp Platform in the North Carolina Community College System

“…We hypothesized that providing students access to Findhelp would help them better meet their basic needs and that this would enable greater student success. Figure 1.1 summarizes this theory of change, which is supported by research documenting links between student resources, basic needs, and academic success (Blankstein and Wolff-Eisenberg, 2021;Daugherty, Johnston, and Berglund, 2020;Weiss, Bloom, and Singh, 2022).…”

“…Other research on supports among community college students suggests larger benefits to more-intensive programs. For instance, programs like the City University of New York (CUNY) Accelerated Study in Associate Programs (ASAP) that pair transportation support with advising, enrollment requirements, and tutoring indicate large benefits (but also larger costs) (Bettinger and Baker, 2014;Evans et al, 2020;Weiss, Bloom, and Singh, 2022;Weiss et al, 2019). 26 The variation in our estimates across college campuses also indicates that variations in implementation, student populations, or college contexts can be important determinants of program effects.…”

“…Only 38 percent of food insecure students reported using their campus food pantry, and many reported being deterred by social stigma, insufficient information, and inconvenience (El Zein et al, 2018). Some programs focused specifically on providing free or reduced-price transit and housing indicate more positive effects (Clay and Valentine, 2021;Perez-Felkner, Baker-Smith, and Goldrick-Rab, 2022;Weiss et al, 2019) but encouraging take-up of these resources is still a challenge (Goldrick-Rab et al, 2021;Umaña et al, 2022;Weiss, Bloom, and Singh, 2022). Evidence on case management, counseling, or advising provided by colleges is also mixed (Evans et al, 2020;Karp et al, 2021).…”

Improving the Usefulness and Use of Meta-Analysis to Inform Policy and Practice

“…This makes it possible to examine heterogeneity of impacts across programs that vary in the types and qualities of services offered. 11 For example, Weiss et al (2022) used the prevalence and number of components to create a measure of “comprehensiveness” (Figure 3, horizontal axis). The study team interacted the number of components with Treatment Status to estimate impacts for participants who were exposed to different numbers of components (Figure 3, vertical axis).

Figure 3.Exploring Variation in Impacts: Micro-Regression Results from THE-RCT.

…”

Funding adequacy and the community college

“…The three Century Foundation reports summarized in this article all cite the ASAP program as evidence that money matters. This program has been carefully evaluated and has been replicated in Ohio, California, and upstate New York with varying degrees of positive results (Weiss & Bloom, 2022). All of these efforts required more money than college budgets could support and had to rely on large amounts of outside grant funding.…”