Epidemiologic Methods
Faculty
- Darin Erickson
- Richard Maclehose
- Alvaro Alonso
- Sue Duval
- Michael Oakes
- Peter Hannan
Suggested Courses
- PubH 6370: Social Epidemiology
- PubH 6363: Group Randomized Trials
- PubH 7401: Fundamentals of Biostatistical Inference
- PubH 8300: Advanced Epidemiologic Methods: Concepts
- PubH 8300: Advanced Epidemiologic Methods: Applications
- PubH 6803: Conducting a Systematic Literature Review
You might also like:
- EpiCH Social Epi Work Group
- Statistical Modeling Blog, Columbia University
- Gary King, Institute for Quantitative Sciences, Harvard
- Methodology Center, Penn State
- Judea Pearl, Cognitive Systems Lab, UCLA
Questions?
Contact: Michael Oakes
Valid and meaningful results through new and sophisticated design and analysis
All research depends on sound study design and analytic techniques to produce valid and meaningful results.
Epidemiology is an arena of research that provides a particularly stringent focus on the application of appropriate study designs and analysis. Thus, as research questions in epidemiology and related fields become ever more complex, new study designs and more sophisticated techniques are required. The Epidemiology faculty at the University of Minnesota is instrumental in developing, implementing and clarifying a wide variety of methodologic designs and analytic techniques .
Areas of Research Concentration:
Bayesian Methods:
Bayesian methods have become increasingly attractive in epidemiologic research as a general tool to explicitly incorporate prior knowledge and fit more complicated regression models.
Faculty working in this area: Richard MacLehose, Peter Hannan
Causal Inference:
There has been a quiet revolution in how epidemiologists address causality. The counterfactual theory of causation has provided a unified way of conceiving of, implementing and analyzing epidemiologic studies.
Faculty working in this area: Michael Oakes, Alvaro Alonso,
Richard MacLehose
Latent Variable Modeling:
Many of the constructs in our studies are not directly measurable. Latent variables are one way to combine multiple, incomplete measures of these constructs into usable variables in our models.
Faculty working in this area: Darin Erickson, Peter Hannan
Longitudinal Data Analysis:
Longitudinal or repeated measures data are ubiquitous in epidemiologic research. Numerous techniques are used to analyze these data, including covariance pattern models, generalized estimating equations, random coefficient/growth curve modeling, survival analysis, and time series modeling, to name a few.
All of our faculty work on one or more of these specific areas.
Meta Analysis:
It is widely recognized that observational and interventional studies often provide the best available evidence, however individual studies may lack power to provide either definitive or hypothesis-generating knowledge. Thus, statistical techniques for combining data from several studies with similar hypotheses have become increasingly popular in epidemiology. The application of systematic review methodology and the tools of meta-analysis are now applied as key evidence-based methods in all areas of epidemiology.
Faculty working in this area: Sue Duval
Social Epidemiology:
Michael Oakes is advancing our methodological understanding of how social systems affect the health of populations. Contributions include the measurement of socioeconomic status, identification of neighborhood effects, fitting multilevel models, design and analysis of group-randomized trials, simulation methods, and matched-sampling designs.