Profile

M. H. Clark is an associate lecturer, statistical consultant, and program evaluator at UCF. She has a Ph.D. in Experimental Psychology with a specialization in Research Design and Statistics from the University of Memphis. Her specific areas of expertise are in causal inference, selection bias in non-randomized experiments, and propensity score methods. She has experience working as a statistical and research consultant in behavioral research and program evaluation since 2000. She teaches graduate courses in statistics and research methods and works as a coordinator for the Computing and Statistical Technology Laboratory in Education (CASTLE), which provides statistical instruction and research consultation to faculty and students. She is part of the UCF Court Health Services and Policy Workgroup, which investigates substance use disorder treatment in the court system. She periodically works as an analyst for the Program Evaluation and Educational Research Group, which provides program evaluations and research services to academic programs. She is a member of the American Evaluation Association and frequently presents statistical papers and workshops on propensity scores at their annual conventions. Some of her studies have been published in the Journal of the American Statistical Association, Evaluation Review, and Learning and Individual Differences.

• Causal Inference
• Experimental Designs
• Applied Statistics

Ph.D. in Experimental Psychology, Research Design and Statistics
University of Memphis

• Propensity score methods
• Academic performance among college students
• Social impacts of substance abuse

Substance Use Policy Evaluation and Research (SUPER)

• EDF 6401: Statistics for Educational Data, spring and summer, Introduction to statistics
• EDF 6481: Fundamentals of Graduate Research in Education, fall and spring, Introduction to research methods
• EDF 7403: Quantitative Foundations of Educational Research, spring, Intermediate statistics
• EDF 7476: Advanced Research Methods, summer, Designs and analyses for improving causal inference