Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I'm trying to find which type of statistical test I need for the study below (inputting into SPSS). I was thinking it could be repeated measures ANOVA or numerous paired t-tests??? Thoughts please...

Title: A study into the effects of acute creatine supplement on cognitive function of sleep deprived subjects.

  • 1 Group (n=11)
  • 4 weeks of testing
  • 3 cognition tests
  • 2 supplements: Creatine & Placebo
  • 2 conditions: Sleep deprived and non-sleep deprived

  • Week 1: Non-sleep deprived & Creatine (3 cognitive tests)

  • Week 2: Non-sleep deprived & Placebo (3 cognitive tests)
  • Week 3: Sleep deprived & Creatine (3 cognitive tests)
  • Week 4: Sleep deprived & Placebo (3 cognitive tests)
share|improve this question
Were there any time lapses between weeks? Or are the 4 weeks consecutive? – Penguin_Knight Feb 27 '14 at 16:33
@Penguin_Knight the 4 tests were done over 2 months, and it wasn't necessary for certain time lapses between the weeks (no effecting dominators) – Harry Feb 27 '14 at 16:41

A within-subjects ANOVA is preferred over paired t-tests both because it protects against family-wise alpha inflation and allows you to analyze the potential interactive effects of your two factors.

However ,your sample size is also quite small. I would look carefully to ensure that the ANOVA assumptions are met. If not, you might consider a Friedman Test.

T-tests (parametric) or Wilcoxon signed-rank tests (non-parametric) would have to follow a significant omnibus test to determine precisely where the differences lie.

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.