Mathematica is conducting this study to provide new information on preschool special education programs and inform policymakers and educators about promising practices to improve outcomes for preschool children with disabilities.
- Design and analysis of data
- Random assignment
- Regression discontinuity
- Propensity score matching
- Strengthening and Disseminating Research
- Labor: Strengthening and Disseminating Research
- Human Services
John Deke has experience in the design and analysis of data from rigorous evaluations based on diverse designs, including random assignment, regression discontinuity, and propensity score matching.
Some recent examples of Deke’s work include a clustered randomized controlled trial (RCT) evaluating mandatory random student drug testing, a clustered RCT evaluating four supplemental reading comprehension curricula, and an evaluation of supplemental educational services based on a regression discontinuity design (RDD). Deke has also contributed to the development of evidence standards for the What Works Clearinghouse (WWC), including attrition standard for RCTs and standards for RDDs. He currently leads the WWC Statistical and Technical Analysis Team responsible for refining WWC evidence standards. Deke has directed two methodological studies to provide analytic and technical support to the Regional Education Laboratories. The first examined the statistical precision gains from adjusting for different types of pre-test measures in clustered RCTs in education. The second study examined the potential for control-group contamination in RCTs that rely on within-school, as opposed to between-school, random assignment.
Deke holds a Ph.D. in economics from the University of Illinois at Urbana-Champaign.
Evaluation of Preschool Special Education Practices
Expanding the Use and Understanding of Evidence-Based Teen Pregnancy Prevention Programs
In recent years, researchers have identified dozens of teen pregnancy prevention programs with demonstrated evidence of success in reducing adolescent sexual risk behaviors. The federal government has invested millions of dollars to disseminate knowledge about these programs and to implement and evaluate...
Feasibility & Design Study: Expanding the Use & Understanding of Evidence-Based Teen Pregnancy Prevention Programs
Much of the current research evidence on teen pregnancy prevention dates to the late 1980s and early 1990s. To address gaps in research, Mathematica and its partners are launching a new set of evaluation activities that will address targeted research questions with practical relevance.
The What Works Clearinghouse: Improving Practice, Research, and Policy
In 2002, the U.S. Department of Education's (ED) Institute of Education Sciences (IES) established the What Works Clearinghouse (WWC) to collect, review, and report on studies of education interventions. The growing focus on evidence-based decision making increased demand for this type of information. Since...