Bayesian statistics for evaluation and policy analysis
Guest Editor: David Rindskopf, CUNY Graduate Center
Bayesian statistics has seen large gains in popularity and could be of immense use in evaluation and policy analysis for (at least) two reasons: (i) Bayesian computation makes it relatively easy to make inferences about complicated quantities, and (ii) Bayesian interpretation makes the presentation of results more useful than frequentist statistics (e.g. the interpretation of a Bayesian 95% credible interval as a 95 percent chance the parameter is in the interval; and more importantly, calculation of other useful probabilities, such as the probability that an effect is bigger than a specific value), and makes it relatively simple to do prediction and decision analysis. A special issue of Evaluation Research will be devoted to uses of Bayesian methods, for which we are requesting proposals for papers. We welcome both theoretical and applied papers as long as there is direct utility for evaluation or policy analysis.
To be considered, interested authors should prepare to a 2-page prospectus or abstract outlining the proposed manuscript. Please email that abstract, no later than April 15, 2017, to David Rindskopf (email@example.com ), copying Jacob Klerman (Jacob_Klerman@abtassoc.com). If invited to proceed with a full manuscript, submissions of full manuscripts must be received no later than July 15, 2017.
Submissions will be peer-reviewed.