You are previewing Statistics Done Wrong.
O'Reilly logo
Statistics Done Wrong

Book Description

Statistics Done Wrong describes how researchers often go wrong and teaches you the best practices for avoiding their mistakes.

Table of Contents

  1. Statistics Done Wrong: The Woefully Complete Guide
  2. About the Author
  3. Preface
    1. Acknowledgments
  4. Introduction
  5. 1. An Introduction to Statistical Significance
    1. The Power of p Values
      1. Psychic Statistics
      2. Neyman-Pearson Testing
    2. Have Confidence in Intervals
  6. 2. Statistical Power and Underpowered Statistics
    1. The Power Curve
    2. The Perils of Being Underpowered
      1. Wherefore Poor Power?
      2. Wrong Turns on Red
    3. Confidence Intervals and Empowerment
    4. Truth Inflation
      1. Little Extremes
  7. 3. Pseudoreplication: Choose Your Data Wisely
    1. Pseudoreplication in Action
    2. Accounting for Pseudoreplication
    3. Batch Biology
    4. Synchronized Pseudoreplication
  8. 4. The P Value and the Base Rate Fallacy
    1. The Base Rate Fallacy
      1. A Quick Quiz
      2. The Base Rate Fallacy in Medical Testing
      3. How to Lie with Smoking Statistics
      4. Taking Up Arms Against the Base Rate Fallacy
    2. If At First You Don’t Succeed, Try, Try Again
    3. Red Herrings in Brain Imaging
    4. Controlling the False Discovery Rate
  9. 5. Bad Judges of Significance
    1. Insignificant Differences in Significance
    2. Ogling for Significance
  10. 6. Double-Dipping in the Data
    1. Circular Analysis
    2. Regression to the Mean
    3. Stopping Rules
  11. 7. Continuity Errors
    1. Needless Dichotomization
    2. Statistical Brownout
    3. Confounded Confounding
  12. 8. Model Abuse
    1. Fitting Data to Watermelons
    2. Correlation and Causation
    3. Simpson’s Paradox
  13. 9. Researcher Freedom: Good Vibrations?
    1. A Little Freedom Is a Dangerous Thing
    2. Avoiding Bias
  14. 10. Everybody Makes Mistakes
    1. Irreproducible Genetics
    2. Making Reproducibility Easy
    3. Experiment, Rinse, Repeat
  15. 11. Hiding the Data
    1. Captive Data
      1. Obstacles to Sharing
      2. Data Decay
    2. Just Leave Out the Details
      1. Known Unknowns
      2. Outcome Reporting Bias
    3. Science in a Filing Cabinet
      1. Unpublished Clinical Trials
      2. Spotting Reporting Bias
      3. Forced Disclosure
  16. 12. What Can Be Done?
    1. Statistical Education
    2. Scientific Publishing
    3. Your Job
  17. A. Notes
    1. Introduction
    2. Chapter 1
    3. Chapter 2
    4. Chapter 3
    5. Chapter 4
    6. Chapter 5
    7. Chapter 6
    8. Chapter 7
    9. Chapter 8
    10. Chapter 9
    11. Chapter 10
    12. Chapter 11
    13. Chapter 12
  18. B. Colophon
  19. C. Updates
  20. Index
  21. Copyright