midterm 2 Flashcards
compare independent samples t test, one way anova, and two way anova
t test:
- two means, one factor
one way anova:
- more than two means, one factor
two way anova:
- more than two means, two factors
what are the three null hypotheses for a two way anova?
main effect of factor A
main effect of factor B
interaction between factor A and B
what do we call the two independent variables in two way anova?
factors
what are the assumptions of two way anova
- homogeneity of variance- variance of popln distributions is equal for each group’
- subjects serve in only one of the treatment conditions (independent samples or between subjects design)
- independence of observations
- populationa distributions is normal within each group
what is a main effect
effect of one factor when the other factor is ignored - difference among marginal means (the mean of all the scores in a column or a row) for a factor
what is the interaction effect
the extent to which the effect of one factor depends on the level of the other factor - interaction occurs when the effects of one factor on the DV change at different levels of the other factor (indicates that main effects along do not fully describe the outcome
how to identify interaction effects and main effects on a graph
- lines cross
- difference in marginal means
within subjects design
- experimental design in which the DV is measured several times within the same subject
- also called a repeated measures design
simplest example is a before and after treatment design - measure subjects before a treatment and after
one way repeated measures design
subjects go through multiple levels of a factor
what are the two possible research questions for one way repeated measures designs
within subject effect:
- are there differences in the mean scores of the DV across groups/conditions? each subject measured at each time point
H0: µ1=µ2 etc
between subject effect
- are there differences across subjects
- the variability of subjects
H0: Vs (variance between subjects)=0
what are we more interested in in one way repeated measures ANOVA, the within subjects effect or the between subject effect
we are more interested in the within subjects effect
the between subjects effect is irrelevent because it has nothing to do with our treatment (IV) - all it tells us is that differences exist naturally between subjects
how do we partition variation in one way repeated measures anova?
SST (total sums of squares) is broken into:
SS(S): variation between row means (subject means) - Xr
SS(A): variation between group means (IV) - Xc
SS (AxS): variation between cell means - SS(AxS)= SS(T)-SS(A)-SS(S)
SS(A) and SS(AxS) come from SS(W)
what are the assumptions in one way repeated measures ANOVA?
Normality
- the distributions of observations on the dependent variable is normal within each level of the factor
compound symmetry:
- homogeneity of variance: population variance observations is equal at each level of the factor - gets replaced with sphericity
- homogeneity of covariance: the population covariance between any pair of repeated measurements is equal (homogenous covariance)
variance vs covariance
variance is the spread of a data set around its mean value, while a covariance refers to the measure of the directional relationship between two random variables - higher covariance means a stronger relationship
compound symmetry
all variances are equal and all covariances are equal