Follow Up Tests

Question 1

3x independent t tests

Answer 1

Alpha (cut off) =

Question 2

Family-wise error rate

Answer 2

If we divide 15% by 3 get back to 5% chance of error

Divide the alpha .05 by the number of t tests we need to conduct

.05 divided by 3 = .016 -> bonferroni correction

If p value is

Question 3

Post hoc test

Answer 3

T tests used to follow up significant anova

Establish whether there are significant differences between the levels of a factor

Independent t tests are used for an independent anova, paired samples t tests used for repeated measures anova

Question 4

Post hoc test reporting

Answer 4

Bonferroni corrected post hoc test revealed that ….. resulted in …. (M= , SD= ), compared to ….. (M= , SD= ; t(…) = …, p= )

Second factor also …..

Differences / no differences (t< 1 / t >1)

Summarising sentence to tell reader is what results mean

Report results of anova and then results of post hoc tests