1 Way Repeated Measures ANOVA Flashcards
repeated measures design
several idependent variables or predictors have been measures, but you used the same group for all conditions
sphericity
assumption that the relationship between pairs of experimental conditions is similiar
compound symmetry
when both the variances across conditions are equal, and the covariances between pairs of conditions are equal
how many conditions do you need for sphericity to be an issue
3
how do we assess sphercitiy
Mauchly’s test
does Mauchely’s test depend on n?
YES. big samples small deviations from sphericity can be significant
what is the effect of sphercitiy on F ratio
creates a loss of power, and fucks up the distribution
When sphericity is fucked up, what post hoc test can you still use?
Bonferronis
Greenhouse Geisser estimate symbol
funny E with hat
Greenhouse Geisser estimate should be used
when greathouse geisser estimate is below.75
is Greenhouse Geisser considered more or less conservative
more
what is Huynh Feldt estimate
less conservative way of adjusting df when sphercity is fucked
when would you use Huynh Feldt
when greenhouse geisser estimate is greater than .75
What is the issue with LSD post hoc
no adjustment for multiple comparisions