The COVID-19 pandemic has had grave implications for students’ learning. According to the National Assessment of Education Progress (NAEP), K-12 students across the nation lost ground academically between 2019 and 2022 in both mathematics and reading. In particular, the pandemic affected access to educational resources and opportunities in ways that exacerbated already persistent inequities in students’ educational experiences and student outcomes. While math and reading proficiency began to rebound in 2023, recent research suggests that rebound has been uneven, leaving students from lower-income communities and students of color at a greater disadvantage than they were before the pandemic.

Students have always entered classrooms with vastly different needs and abilities—and differentiating instruction to meet diverse learning needs is nothing new for educators—but the pressure on schools is greater than ever. Teachers are now tasked with helping all their students master grade-level standards while also helping an increased share of their students catch up. Supplemental academic supports that tailor instruction to each student’s individual strengths and needs is a promising solution, especially if those supports are delivered in ways that do not place additional burden on already over-taxed classroom teachers. MDRC is currently exploring two particularly promising delivery mechanisms for this type of tailored instruction: tutoring and adaptive educational technology (or adaptive ed tech).

Tutoring is a form of one-on-one or small-group instruction that tailors the academic content and pacing to a student’s individual strengths and needs. Tutoring has consistently been shown to be effective in rigorous research studies, including research conducted by MDRC. Despite this promise, studies have shown that challenges arise when implementing tutoring at scale, such as low tutor and student participation and inconsistency in the quality of the academic content and pedagogy. In addition, the high price tag and the challenges associated with identifying and maintaining a tutor workforce have been barriers to scaling tutoring programs.

Adaptive Ed Tech products use assessments and sometimes artificial intelligence (AI) technology to adjust the content and pacing of instructional material to meet students’ specific learning needs. Studies have shown that these digital learning platforms can accelerate student learning and are particularly promising in math and for students who are far behind. Compared to tutoring, ed tech products are not difficult to scale from a cost or labor market perspective. However, research has shown that a lack of consistent integration of the products into classroom and curriculum (that is, incorporating online and teacher-led instruction so that they complement each other) can be a barrier to achieving meaningful impacts. The products can only be beneficial if students use them for a significant amount of time, and achieving sufficient “dosage” is often cited as a challenge.

A new generation of research is needed to advance our understanding of these two promising types of tailored individualized instruction—in particular, to determine whether models found to be effective in pre-pandemic settings can be adapted to meet the current urgent needs of K-12 educators and students. MDRC is playing leading roles on three large national studies, all using experimental evaluation designs. They offer students tailored individualized instruction that supplements core classroom instruction using different delivery models: the Personalized Learning Initiative, the Reading Partners Connects Evaluation, and the ReSolve Math Study.

Personalized Learning Initiative: The Personalized Learning Initiative (PLI) is a large-scale, national demonstration to determine how to provide both high-dosage tutoring and other less resource-intensive personalized learning interventions to many students in a cost-effective manner. To accomplish this, the PLI team is providing technical assistance to districts, a charter management organization, and one state in: (1) scaling traditional live high-dosage tutoring (tailored instruction provided to students at least three times a week for a significant portion of the school year) and (2) developing and implementing approaches that are potentially more sustainable, some of which include adaptive ed tech. The PLI project is supporting and testing models that provide tutoring during school time, some in person and others that are virtual (see this brief from the Education Trust and MDRC, which summarizes the evidence on different types of tutoring). The project involves a multiyear student-level randomized controlled trial (RCT) with over 15,000 students to rigorously assess the efficacy of the high-dosage tutoring model and the less resource-intensive models developed by the study team and partner localities. The PLI project is led by the University of Chicago Education Lab in collaboration with MDRC. For more information on the PLI project, see this page or check out these recent results and blog posts from the PLI team.

Reading Partners Connects Evaluation: During the pandemic, Reading Partners—a nonprofit that provides evidence-based volunteer tutoring in reading in Title I elementary schools—developed and launched an online tutoring platform, Reading Partners Connects. Reading Partners is continuing to pursue this model because it allows them to extend their services to areas where it’s difficult to recruit volunteers, including meeting higher demand in schools that already have in-person tutoring and expanding tutoring to new places, like rural areas, where finding in-person volunteers can be particularly hard. The program provides 45 minutes of one-on-one tutoring conducted by volunteers two times per week using video conferencing. Reading Partners Connects includes high-quality lesson content and eBooks for students and tutors and incorporates elements of social and emotional learning into the curriculum. MDRC and the University of Michigan will evaluate the implementation (including costs) and impact of Reading Partners Connects in several Title I elementary schools, using a student-level RCT. The study team will also collect lessons learned from the scale-up of Reading Partners Connects in schools. Reading Partners and MDRC are currently recruiting school and district participants. To learn more about Reading Partners Connects and the evaluation, listen to this recent episode of MDRC’s Evidence First podcast or check out this project description.

ReSolve Math Study: The ReSolve Math Study is a national study contracted by the U.S. Department of Education’s Institute of Education Sciences and being led by MDRC in partnership with Digital Promise, RAND, Westat, and Public Strategies. The study will employ a student-level RCT to determine how best to use digital math products to accelerate elementary school students’ math learning. The study will involve two years of implementation of high-quality adaptive digital math products with students as a complement to core teacher instruction. The study will compare two different instructional approaches the products can use to help students catch up:

• Broad Foundational Skill-Building—a sequential approach, which views all below-grade content as foundational and systematically reteaches any content where there are knowledge gaps.
• Focused Just-In-Time Skill-Building—an approach that reteaches only below-grade content deemed most essential to understanding the current grade-level topic being addressed in class.

By comparing these two different instructional approaches, the study will generate much-needed evidence on a debate in mathematics education: whether students’ mathematics proficiency—in both the short and long term—is better served by instruction that systematically supports building mastery in unfinished learning (Broad Foundational Skill-Building) or by instruction that targets unfinished learning in just the prerequisite concepts thought to be essential to succeed in grade-level content (Focused Just-In-Time Skill-Building). The study will also explore costs of the products and the support needed for their use. The ReSolve Math study team is currently recruiting school and district partners. See this page for more information about the project.

Taken together, these three studies will improve our understanding of whether and how supplemental academic supports can effectively accelerate learning and of which models work best for different types of students. Each of these evaluations will be conducted over the next few years, and the study teams will publish findings as they become available.

For more information about MDRC’s work on tailored individualized instruction, please reach out to Barbara Condliffe (barbara.condliffe@mdrc.org) or William Corrin (william.corrin@mdrc.org).