Random Assignment Overview

The use of randomized field trials (RFTs) in education research is a topic of continuing controversy. In the push to transform education into an "evidence-based" field, current federal policy initiatives rank RFTs at the top of a hierarchy of methods for conducting research in education and for justifying the use of federal funds as grounded in rigorous research. Scientific Research in Education (NRC, 2002) also highlights RFTs as a powerful method for establishing causal relationships in education, while situating its use within the broader context of scientific inquiry and outlining its limitations.

The practical experience of researchers and practitioners who have directly participated in RFTs in social settings has generated important lessons that can inform these debates and the appropriate use of RFTs in the future. Convening researchers and practitioners who have this experience can help to identify common obstacles, share successful solutions, air ongoing concerns, and engage in dialogue about the implications of the push to promote more RFTs for education research, policy, and practice.

With these issues in mind, the workshop will bring together researchers and practitioners to promote understanding, and the appropriate use, of random assignment in education research.

Broadly stated, the dual audiences are federal education research policymakers and educational practitioners. By federal research policymakers, we mean officials in Congress and the Administration who influence education research priorities and funding. By practitioners, we mean the diverse set of educators and administrators who are influential in (e.g., in agreeing to participate in such studies) and influenced by (e.g., in having to justify federal program dollar expenditures as “scientifically based”) the current trend to promote random assignment methods in education settings.

Within practitioner groups in particular, we will need to leverage existing networks (e.g., via professional associations) to maximize our efforts both in terms of getting participants and disseminating products. Also, the audience varies by question/workshop session.

1) Under what conditions and for what kinds of questions is random assignment best suited? What kinds of questions are not well suited for random assignment experimentation?

2) How can this tool be used in conjunction with other methods to advance understanding of causal connections? What balance of investments between random assignment experiments and other types of studies seems most likely to be fruitful?

3) Within the broader context of federal statutes, regulations, and other issuances governing ethical access to human research participants, what are the ethical considerations for random assignment experimentation in education, and how can they be overcome?

4) What resources are necessary to implement random assignment effectively in educational settings? (e.g., time, money, partnering with schools or other entities)

5) What can be learned from looking at other fields about the strengths and weaknesses of different structures for supporting large-scale random assignment experiments?

A workshop summary will be generated to provide a synopsis of themes and issues (but not include recommendations). Finally, the discussions will feed into the committee's final consensus report on how to promote improvements in education research more broadly.