top of page

STATISTICAL TUTORIAL CHANNEL RESEARCH

Find out which is the best statistical test for your research!

ANSWER:

Post Hoc Tests

  • TUKEY:  Most used test because it is very conservative with respect to Type I error

 

  • GAMES-HOWELL:  Used when you don't assume equality of variances

​

  • SCHEFFÉ:  The most conservative of all tests. May excessively increase the chance of committing a Type II error

​

  • BONFERRONI:  Suggests that the p-value of each test should equal its alpha divided by the number of tests performed

​ ​

  • HOLM:  Sorts from lowest p (raw) value to highest. Multiply the p-value in order. For the largest jth of the p-values, the fit is whichever is greater

See the step-by-step guide on how to run and interpret the ANOVA Test in JAMOVI software.

bottom of page