Methods Inquiries, 2013-2022

This brief has two main goals:

• Describe the features of a well-designed and implemented subgroup analysis that uses a multiple regression framework.
• Provide an overview of recent methodological developments and alternative approaches to conducting subgroup analyses.

The brief builds on a 2009 meeting of experts Visit disclaimer page convened by the Administration for Children and Families’ Office of Planning, Research, and Evaluation and a corresponding 2012 publication in a special issue of Prevention Science Visit disclaimer page (MacKinnon, Supplee, Kelly, & Barofsky, 2012).

“Open science” represents a broad movement to make all phases of research—from design to dissemination—more transparent and accessible. The scientific community and Federal agencies that support research have a growing interest in open science methods. In part this interest stems from highly publicized news stories and journal articles that cast doubt on research credibility...

“Open science” represents a broad movement to make all phases of research—from design to dissemination—more transparent and accessible. The scientific community and Federal agencies that support research have a growing interest in open science methods in response to highly publicized news stories and journal articles that cast doubt on research credibility...

Social service program stakeholders need timely evidence to inform ongoing program decisions. Rapid learning methods, defined here as a set of approaches designed to quickly and/or iteratively test program improvements and evaluate program implementation or impact, can help inform such decisions. However, stakeholders may be unsure which rapid learning methods are most appropriate for a program’s specific challenges and how to best apply the methods...

Social service program stakeholders need timely evidence to inform ongoing program decisions. Rapid learning methods, defined here as a set of approaches designed to quickly and/or iteratively test program improvements and evaluate program implementation or impact, can help inform such decisions. However, stakeholders may be unsure which rapid learning methods are most appropriate for a program’s specific challenges and how to best apply the methods. Additionally, they may be unsure how to cultivate a culture of continuous, iterative learning.

For nearly 100 years, the null hypothesis significance testing (NHST) framework has been used to determine which findings are meaningful (Fisher 1925; Neyman and Pearson 1933). Under this framework, findings deemed meaningful are called “statistically significant.”  But the meaning of statistical significance is often...

Bayesian methods allow researchers to describe findings in probabilistic terms. They also allow for incorporating prior knowledge and considering uncertainty in parameters.

Bayesian methods may also help with transparency...

Federally funded systematic reviews of research evidence play a central role in efforts to base policy decisions on evidence. Historically, evidence reviews have reserved the highest ratings of quality for studies that employ experimental designs, namely randomized control trials (RCTs). However, RCTs are not appropriate for evaluating all intervention programs. To develop an evidence base for those programs, evaluators may need to use non-experimental study designs.

There is a growing understanding that there are some inherent limitations in using p-values to guide decisions about programs and policies. Bayesian methods are emerging as the primary alternative to p-values and offer a number of advantages...

In this video roundtable, government experts and experienced researchers discuss the opportunities and challenges presented when using administrative data for social policy research. Topics include: tips for planning administrative data research; working with (federal and state) data custodians; negotiating data...