The William T. Grant Foundation offers several online resources, including tools and consulting services, to help grantees design more effective studies. (Relevant links to our website, www.wtgrantfdn.org, are provided.)

One of the most useful of these online resources, Optimal Design, was recently updated using feedback from researchers and users. Optimal Design is a software package, developed by Stephen Raudenbush and his colleagues, which helps researchers plan adequately powered randomized trials. This includes determining sample size, statistical power, and optimal allocation of resources. Version 2.0 of the software includes several important updates. These include a new user interface that organizes designs into three groups: person randomized trials, cluster randomized trials with person-level outcomes, and cluster randomized trials with group-level outcomes. There are also new graphical options related to minimum effect size and a new option for binary outcomes. Raudenbush and his team created a comprehensive guide for using and understanding Optimal Design v.2.0.

In addition to the new software, the Foundation offers a consultation service for the design of group-randomized studies. This service is offered in collaboration with Dr. Raudenbush at the University of Chicago and his colleagues at the University of Michigan. This same team and colleagues at MDRC led by Howard Bloom have published widely on study design and other topics related to studying youth settings and relevant papers and journal articles are also available on the Foundation’s website.

“An important part of our approach is to provide researchers with tools and other services to help improve their work. This new generation of the Optimal Design software is a significant step forward, and it will lead to better studies,” said Robert C. Granger, president of the Foundation.