LECTURE 6 repeated measures anova Flashcards
how is f ratio calculated in repeated measures one way anova
0 individual diff therefore
f = sys diff + 0 + exp error (diff between sample means)
/ indiv diff + exp error (diff expected by chance)
if no effect - f around 1
how does repeated measures differe to independent
subjects experience both conditions therefore no differences between subjects
BUT - order effects could occur so must counterbalance
benefits of repeated measures anova
fewer participants
removes individual differences variability
more sensitive than independent measures
increased statistical power
within condition variability =
subject score - collumn mean
between condition variability =
collumn mean - overall mean
variance of participant ability =
mean of first subject ability - all means (how bad one person is relative to everyone else )
how do you work out between condition df
no of coniditons (k) - 1
how do you work out between subjects df
no of subjects (n) - 1
how do you work out error df
(k-1) x (n-1)
what does mulchys test of sphericity test
homogeniety of variance and covariance
no significant effect means assume equal variance across IV groups
what box look at when repeated measures anova
within subjects NOT BETWEEN
sphericity assumed if maulchys not sig
greenhouse geisser if maulchys sig
Need for one way within subj (repeated measures) ANOVA
1x IV
3+ levels
Pps in all conditions
