An official website of the United States government
Module 8. Quantitative Analytic Techniques, Part 2
- Thomas Belin PhD; University of California, Los Angeles
- Edward J. Miech, PhD; Regenstrief Institute, Indianapolis, IN
Training Materials
View Presentation (PPTX, 2.5 MB)
Module Summary
- Challenges to internal and external validity
- Foundations for statistical inference and the role of randomization
- Strategies for enhancing precision of inferences
- Statistical issues arising from clustering/nesting of sub-units within units
- Configurational comparative methods
Reading Materials
- Berkel, C., Mauricio, A.M., Schoenfelder, E., & Sandler, I. N. Putting the pieces together: An integrated model of program implementation
. Prevention Science. 2011; 12, 23-33. - Belin, T.R., Jones, A., Tang, L., Chung, B., Stockdale, S.E., Jones, F. Pulido, E., Ong., M.K., Gilmore, J., Miranda, J., Dixon, E., Jones, L., Wells, K.B. Maintaining internal validity in community partnered participatory research: Experience from the community partners in care study. Ethnicity and Disease. 2018;28;S2.
- Cleary, P. D., Gross, C. P., Zaslavsky, A. M., & Taplin, S. H. Multilevel interventions: study design and analysis issues. Journal of the National Cancer Institute Monographs, 2012; (44),49-55.
Web Resources
- Center for Multilevel Modeling
- Configurational Comparative Methods (CCM), Edward J. Miech. Regenstrief pioneering development and uptake of Configurational Comparative Methods
- David Kenny mediation website
- FDA guidance on co-primary endpoints
- Applied Multilevel Analysis (PDF)
- Longitudinal mediation models
- Mediation should be done even when there is no overall treatment effect
- Missing data, London School of Hygiene & Tropical Medicine
- Analysis of partially observed clustered data using generalized estimating equations and multiple imputation
- Weighted Methods for Analyzing Missing Data with the GEE Procedure (in SAS) (PDF)
Self-reflection Questions
- What type of quantitative research question are you trying to answer?
- What are you measuring, manipulating and/or controlling?
- What are you using for your dependent, independent, categorical, and continuous variables and how do they apply to the analytic methods?