Category 3: Standards for Data Integrity and Rigorous Analyses
Learning Objectives
Cognitive: Compare the content of different types of secondary data to select data appropriate to study goals; Describe the characteristics of a well-validated exposure or outcome variable, including scales; Explain the key threats to validity of exposure and outcome variables in secondary data sets; State the purpose of control of confounding in secondary data analyses; Define internal and external validity of a study using secondary data.
Attitudinal: Appreciate the challenge of assessing the causal effects of exposures on outcomes from observational or experimental studies; Commit to protecting data from breaches of confidentiality and privacy; Recognize the importance of capturing the diverse sources of uncertainty in estimates of causal effects through sensitivity and other analyses.
Skills: Draw a causal graph that represents the major observed variables, unobserved variables, and assumptions that constitute the proposed analysis; Select one or more data sets that meet the project’s needs; Develop an analytic plan to address the specific aims of the project; Choose validated exposure and outcome variables, when available; Prepare a satisfactory description of a plan for control of confounding; Prepare a plan to address potential bias and variance in effect estimates caused by unobserved variables—for example, using instrumental variables or sensitivity analyses; Formulate a thoughtful plan for subgroup analyses and/or sensitivity analyses as needed to address specific aims; Apply tools to assess a proposed study’s internal validity at the time of study design; Conduct analyses that are reproducible by others.
Learning Modules
This category contains the curriculum Introduction and nine main modules. Learn about the instructors for this curriculum.
- Introductory Lecture: Prepared and presented by Jodi Segal, MD, MPH
- Learning Modules: Prepared by Jodi Segal, MD, MPH, and Scott Zeger, PhD; Presented by Jodi Segal, MD, MPH
Module 1: Introduction
Watch | Listen | Handout
Module 2: Objectives and Objectives Illustration
Watch | Listen | Handout
Module 3: Initial Considerations regarding Data Integrity and Rigorous Analyses
Watch | Listen | Handout
Module 4: Thinking about Causality
Watch | Listen | Handout
Module 5a: Primary Data
Watch | Listen | Handout
Module 5b: Secondary Data
Watch | Listen | Handout
Module 6a: Imperfect Exposure Variables
Watch | Listen | Handout
Module 6b: Imperfect Outcome Variables
Watch | Listen | Handout
Module 6c: Attention to Scales and Instruments
Watch | Listen | Handout
Module 7a: Planning Analyses of Observational Data
Watch | Listen | Handout
Module 7b: Describing a Data Analysis Plan
Watch | Listen | Handout
Module 8a: Matching and Restriction
Watch | Listen | Handout
Module 8b: Standardization, Stratification, and Regression
Watch | Listen | Handout
Module 8c: Introduction to Propensity Scores
Watch | Listen | Handout
Module 9: Summary
Watch | Listen | Handout
Interview with Nancy Kass, ScD, Phoebe R. Berman Professor of Bioethics and Public Health in the Department of Health Policy and Management, Johns Hopkins Bloomberg School of Public Health, and Deputy Director for Public Health in the Berman Institute of Bioethics, on why the application of good methods is an ethical question
Listen
Category 3 Self-Assessment
View
Back to the Methodology Standards Academic Curriculum main page
Posted: February 2016
What's Happening at PCORI?
The Patient-Centered Outcomes Research Institute sends weekly emails about opportunities to apply for funding, newly funded research studies and engagement projects, results of our funded research, webinars, and other new information posted on our site.