Random Assignment: Yet another, but this time more empirically based, assessment of how superior it is for describing causal connections
Tuesday, June 11, 2013
Noon – 1:30pm
Marvin Center, 800 21st Street, N.W., Continental Ballroom (third floor), Washington, DC 20052
The George Washington University
Thomas D. Cook , Joan and Sarepta Harrison Chair of Ethics and Justice,
Professor of Sociology, Psychology, and Education and Social Policy,
There is no doubt that a central task of evaluation is to assess effectiveness and that random assignment is the most widely advocated design tool for describing the causal nature of links between a program and its consequences. The question is whether random assignment is so much more effective than its design-centered alternatives that its limitations with respect to external validity and knowledge of causal mechanisms should be ignored. We assess the viability of the major design-centered alternatives in terms of their ability to reproduce the same causal results as experiments. By this empirical criterion, we conclude that experiments are to be preferred -- but only very slightly so and thus only if their limitations regarding external validity and knowledge of causal mechanisms can be tolerated. Relative to other accounts for or against random assignment, the present talk muddies assessments of its standing and implies that cause-testing policy officials have to be more selective about when they require random assignment versus its design-centered alternatives.
Thomas Cook is interested in social science research methodology, program evaluation, whole school reform, and contextual factors that influence adolescent development, particularly for urban minorities. Cook has written or edited 10 books and published numerous articles and book chapters. He received the Myrdal Prize for Science from the Evaluation Research Society in 1982, the Donald Campbell Prize for Innovative Methodology from the Policy Sciences Organization in 1988, the Distinguished Scientist Award of Division 5 of the American Psychological Association in 1997, and the Sells Award for Lifetime Achievement, Society of Multivariate Experimental Psychology in 2008, and the Rossi Award from the Association for Public Policy Analysis and Management in 2012. He was chair of the board of the Russell Sage Foundation from 2006 to 2008. Cook was elected to the American Academy of Arts and Sciences in 2000 and was inducted as the Margaret Mead Fellow of the American Academy of Political and Social Science in 2003. He was part of the congressionally appointed committee evaluating Title I (No Child Left Behind) from 2006 to 2008.