Handling extremely small samples

• For extremely small samples,  2 - test even with
  Yate’s correction is NOT recommended


• Fisher’s exact test should be used when there are
  small expected frequencies
  
  
  • Involves calculating the exact probability of a table
    as extreme or more extreme than the one
    observed, given that the null hypothesis is correct
  
  
  
  


• When to use fisher’s exact test (rule of thumb)
  
  
  • When the overall sample size < 20
  
  
  • Overall sample size is between 20 and 40 and the
    smallest of the four expected value < 5