Over the last twenty years, education researchers have increasingly conducted randomised experiments with the goal of informing the decisions of educators and policymakers. Such experiments have generally employed broad, consequential, standardised outcome measures in the hope that this would allow decisionmakers to compare effectiveness of…

This paper discusses, compares, and contrasts 4 different models for bringing evidence from randomised controlled trials (RCTs) into practice and into practical reasoning. I look at what questions the models can and cannot answer, what role they accord to RCT evidence, and what their possible attraction for practitioners might be. The models are…

Demands for scientific knowledge of what works in educational policy and practice has driven interest in quantitative investigations of educational outcomes, and randomized controlled trials (RCTs) have proliferated under these conditions. In educational settings, even when individuals are randomized, both experimental and control students are…

Cluster randomized experiments are ubiquitous in modern education research. Although a variety of modeling approaches are used to analyze these data, perhaps the most common methodology is a normal mixed effects model where some effects, such as the treatment effect, are regarded as fixed, and others, such as the effect of group random assignment…

Design-based methods have recently been developed as a way to analyze data from impact evaluations of interventions, programs, and policies. The impact estimators are derived using the building blocks of experimental designs with minimal assumptions, and have good statistical properties. The methods apply to randomized controlled trials (RCTs) and…

A new design-based theory has recently been developed to estimate impacts for randomized controlled trials (RCTs) and basic quasi-experimental designs (QEDs) for a wide range of designs used in social policy research (Imbens & Rubin, 2015; Schochet, 2016). These methods use the potential outcomes framework and known features of study designs…

Design-based methods have recently been developed as a way to analyze data from impact evaluations of interventions, programs, and policies (Imbens and Rubin, 2015; Schochet, 2015, 2016). The estimators are derived using the building blocks of experimental designs with minimal assumptions, and are unbiased and normally distributed in large samples…

Attrition occurs when members of the initial research sample are not part of the final analysis sample, such as due to missing data or leaving the study. Both the overall sample attrition and the differences in attrition between the groups can affect the statistical equivalence of the sample and create potential for bias. The WWC has given careful…

Meeting the What Works Clearinghouse (WWC) attrition standard (or one of the attrition standards based on the WWC standard) is now an important consideration for researchers conducting studies that could potentially be reviewed by the WWC (or other evidence reviews). Understanding the basis of this standard is valuable for anyone seeking to meet…

Attrition occurs when members of the initial research sample are not part of the final analysis sample, such as due to missing data or leaving the study. Both the overall sample attrition and the differences in attrition between the groups can affect the statistical equivalence of the sample and create potential for bias. The WWC has given careful…

Purpose: The past decade has seen considerable debate about how to best evaluate the efficacy of educational improvement initiatives, and members of the educational leadership research community have entered the debate with great energy. Throughout this debate, the use of randomized experiments has been a particularly contentious subject. This…

In randomized control trials (RCTs) of educational interventions, there is a growing literature on impact estimation methods to adjust for missing student outcome data using such methods as multiple imputation, the construction of nonresponse weights, casewise deletion, and maximum likelihood methods (see, for example, Allison, 2002; Graham, 2009;…