BIOS601 Resources: Bias Reduction

• Comparative Studies, in general [expt'l / non-expt'l / quasi-expt'l]

Textbooks

• Planning of Experiments - Cox
• Planning and Analysis of Observational Studies - Cochran
• Textbook: Statistical Methods for Comparative Studies: Techniques for Bias Reduction
  - Anderson et al.     [HIGHLY RECOMMENDED by JH]
• A Short Textbook of Medical Statistics - Bradford Hill

Chapters, Articles

• The Method of Expected Number of Deaths, 1786-1886-1986 Keiding1987
• Standardization and control for confounding in observational studies -
  a historical perspective Keiding&Clayton2012
• Rates & Proportions: Standardization (Van Belle et al.)
• Appropriate uses of multivariate analysis (Hanley 1983).
• Glossary from Theoretical Epidemiology (Miettinen)
• Miettinen's Dictionary of Statistics and Epidemiology

Other...

• "I don't get it"
• Video Program 11. The Question of Causation: This program defines the concepts of common response and confounding, explains the use of two-way tables of percents to calculate marginal distribution, uses a segmented bar to show how to visually compare sets of conditional distributions, and presents a case of Simpson's Paradox. The relationship between smoking and lung cancer provides a clear example.
• Confounding -- in pictures and numbers
• Admissions to Berkeley (incl. Mantel-Haenszel calculations)
• Mac vs IBM, Simpson's Paradox, Other confounding examples
• Outcomes of Pregnancy during Residency for women and wives of their male classmates
• Blood Pressure and Altitude
• Neurodevelopment of children exposed in utero to antidepressant drugs
• Confounding: Reducing it by Regression
• Sexual activity reduces the lifespan of male fruitflies
• Breast milk and subsequent intelligence quotient in children born preterm
• Exposure to Scientific Theories Affects Women's Math Performance
• Weekend versus weekday admission and mortality from myocardial infarction.
• Link to Resources on Comparative Studies (general)

Datasets, R code

• Fruitfly Data and programs [ SAS ] [ Stata ] ] [ full data for R] [cf article under "Other" above ]
  No. partners: 0 1 or 8 ; active: 0 (no) 1 (yes) 9 (not applicable, living alone)
   
  
• (MEASLES, MUMPS, AND RUBELLA) VACCINATION AND AUTISM
  
  See `Data and articles with data' in this link