Analysing 3 or more conditions One Flashcards
what is ANOVA
analysis of variance
extention of the t test
allows us to test if 3 or more population means are the same, without a reduction in power
test statistic is F (as it was invented by Fisher)
compares the amount of variance explained by our experiment with the variance that is unexplained
what are the three versions of ANOVA and when would you use them
ANOVA - analysis of variance
(one way/one factor anova)
MANOVA - multivariate analysis of variance
(extension of anova for multiple DVS)
ANCOVA - analysis of covariance
(extension of anova for continous variables such as correlations)
what does ANOVA assume
scores were sampled randomly and are independent
normal distribution
roughly equal number of participants in the groups
roughly equal variance for each condition
what is explained and unexplained variance in between groups designs for ANOVA
explained - variance between groups
unexplained - variance within a group
talk about anova degrees of freedom
theres two values - one for each aspect of the variance
we report both, such as F(3,32)=15.3, p=0.05
how do you do a single IV anova in SPSS
analyze
compare means
one way anova
how do you do a general ANOVA in SPSS
analyze general linear model univariate (put IV in 'fixed factor' box) select post-hoc test (bonferroni)
when do you run a post hoc test for ANOVA
if main test found a significant result
what are other terms for the unexplained variance
residual or error
whats SS
sum of squares
whats MS
mean square
whats it called when theres one IV
one-way test, then two way etc
whats it called when theres 1 DV
univariate
whats it called when theres multiple DVs
multivariate
how do you calculate the f ratio
compute variance betwen groups (effect)
divide by the variance within groups (noise)
