Reading...
Play button
Play button
Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
18 Cards in this Set
- Front
- Back
|
Discriminant Analysis
|
which IV has the most impact on the DV
|
|
|
Factor Analysis
|
underlying explanatory structure
|
|
|
ANOVA
|
1 IV, 1 IV with more than 1 level or categories, 1+ IV
but only 1 DV |
|
|
The DV in ANOVA
|
must be ordinal or nominal
|
|
|
ANOVA assumptions
|
homogeneity of variance in the population
normal distribution in population (random selection) independent observations in the sample |
|
|
ANOVA statistic result
|
F > 1 reject null
F = 1 retain null (between-group and within group are same and therefore due to chance) critical value of F (a table) |
|
|
Why do you do 2-way ANOVA?`
|
to see if there's a MAIN OR INTERACTION effect
to get the EFFECT SIZE |
|
|
ANCOVA
|
partials out the effect of a co-variate (thought to influence the DV) when it runs the analysis
|
|
|
ANCOVA Assumptions
|
reliability of covariate
linear relationship (correl) of DV & IV homogeneity of regression for covariate (same for each group of IV's) |
|
|
the DVs in MANOVA
|
interval or ratio
|
|
|
MANOVA & Type I
|
MANOVA decreases Type I error because you compare all the DV's at the same time
|
|
|
MANCOVA covariate
|
related to DV, not the IV
continuous (interval or ratio) |
|
|
What do you do if you get an interaction effect?
|
Stop the analysis.
|
|
|
MANOVA assumptions
|
random samples
homescedasticity linear relationship between DV's for "Franken-DV" (linear combination of DV's that maximize group differences) |
|
|
Multiple Regression stat
|
R2
how much variability in DV can be explained by a linear combination of IV's aka coefficient of determination |
|
|
Multiple regression threat
|
multicollinearity:
the predictor IV' are highly correlated to each other |
|
|
Multiple Regression Assumptions
|
homoscedasticity of variance (the same across IV's)
linear relationship IV and DV prediction errors are normal dist, don't correlate w/IV, independent observations IV's are FIXED |
|
|
Coefficient of determination
|
R2
|
