WTF Statistics

Statistical help for non-statisticians

  • Finding the Right Statistical Test
  • Statistics References
  • Finding the Right Statistical Test
  • Statistics References
Home / Statistics References / Core Statistical Concepts / What is post-hoc testing

What is post-hoc testing

Post-Hoc Testing

Post-hoc testing is used when a statistical test of three or more samples is statistically significant, and you wish to determine which individual samples are different from one another.  The main test tells you only that one or more of the samples is different from one or more of the other samples.  Post-hoc testing is needed to determine which individual samples differ.

WTF Contents

  • Home
  • Finding the Right Statistical Test
    • Single Variable Analysis
      • Testing for Normality
    • Comparisons of the Same Measure Between Groups
    • Relationships Between Different Variables
      • Correlations
      • Interrater Reliability
    • Variable Reporting
    • Statistical Test Reporting
  • Statistics References
    • Variables & Distributions
      • Statistical Data Types
      • Independent and Dependent Variables
      • Paired Data and Repeated Measures
      • Normal vs. Non-Normal, Parametric vs. Non-Parametric
      • Arithmetic, Geometric, and Harmonic Means
      • Understanding Medians and Quartiles
    • Core Statistical Concepts
      • What does a p-value from a test mean?
      • What is post-hoc testing
      • Confidence Intervals
      • Power Analysis
      • Statistical versus Clinical Significance
  • ROC Analysis
  • About WTF Statistics
(c) WTF Statistics 2018