Within subject T tests

Question 1

What does a within subjects T test tell us?

Answer 1

Is the difference between 2 means significant?

How likely are we to obtain the observed difference between conditions if the null hypothesis is true? E.g If the distributions overlap perfectly
Is the variance between conditions larger than the variance within conditions?
The t statistic represents this ratio t = variance explained by our IV / unexplained variance

Question 2

Assumptions of a within subjects t test

Answer 2

Data is interval or ratio (continuous)
N of at least 12
Homogeneity of Variance - Differences are normally distributed

Question 3

How to test normality of difference?

Answer 3

Shapiro-Wilk should be non-significant (p>0.05)

Question 4

Non parametric alternative to a paired sample t test

Answer 4

Wilcoxon Signed ranks test for related samples