Read by Grade Three Evaluation


Literacy knowledge and skills developed in the early elementary grades predict long-term literacy achievement, on-time graduation, and later-life outcomes. In recognizing the critical importance of establishing early literacy skills, there has been substantial instructional and policy attention given to improving rates of reading proficiency in the elementary grades. In 2016, based on the recommendation of the Michigan Third Grade Reading Workgroup, the Michigan legislature passed the Read by Grade Three Law (RBG3), which requires schools to identify learners who are struggling with reading and writing and to provide additional help.

The Read by Grade Three Law relies on an inherent Theory of Change, which suggests that state support and early intervention – combined with the threat of retention – improves student (and teacher) effort. For students for whom early intervention is insufficient to enable their meeting third-grade proficiency standards, retention will allow them extra time and instruction to read on grade level.

Who is on the Project Team?

Brian Jacob, University of Michigan
Susan Dynarski, University of Michigan
Katharine Strunk, Michigan State University
Venessa Keesler, Michigan Department of Education
Jasmina Camo, University of Michigan

How is the Study Funded?

This study is funded by the United States Department of Education, Institute of Education Sciences grant number R305H1900004 to Michigan State University and the University of Michigan. The opinions expressed are those of the authors and do not represent views of the Institute or the U.S. Department of Education.

The Partners

The project is being completed in partnership with the Michigan Department of Education, Center for Educational Performance and Information, Education Policy Innovation Collaborative at Michigan State University, and Michigan DataHub.

What are the Research Objectives?

A key goal for the RBG3 law is to understand how the policy codified within this legislation modifies existing practices in Michigan schools and how any changes relate to the efficacy of the reform model. This policy includes a highly prescriptive set of requirements for districts, intermediate school districts (ISDs), the Michigan Department of Education (MDE), and the Center for Educational Performance and Information (CEPI). We group these requirements into two main categories:

  1. Improved literacy instruction.
  2. Monitoring, remediation, and retention.

The legislation codifies perceived best practices in reading instruction, identifying five evidence-based instructional components of effective literacy instruction: phonemic awareness, phonics, fluency, vocabulary, and comprehension. The legislation also mandates that districts provide teachers with professional development aligned to these “best practices.” These instructional practices are intended to improve outcomes for all Michigan students.

Districts must administer valid, reliable reading assessments to all K-3 students at least three times annually. MDE has released a list of 27 acceptable assessments. Students identified as having a reading deficiency must receive an Individualized Reading Improvement Plan (IRIP) that includes supplemental reading instruction during school hours and a “Read at Home” plan for parents, guardians, or caretakers. This requirement reflects the understanding, codified in the legislation, that there is an important role for home caregivers to play in improving literacy outcomes for children

The RBG3 evaluation addresses the following research questions:

  1. Does the Read by Grade Three law improve the achievement and attainment of Michigan’s students? Is there heterogeneity in this effect across populations and places?
  2. Do students subject to remediation and retention increase their achievement and attainment? Is there heterogeneity in this effect across populations and places?
  3. How is the Read by Grade Three Law being implemented in Michigan? Does implementation vary across populations and places and if so, why?
  4. Do differences in implementation (RQ3) explain any heterogeneity in effects (RQ1 & RQ2)?
  5. Is this policy an efficient use of resources?


content here

Additional Resources

content here