Assumptions Flashcards
1
Q
Multilevel Modelling
A
- all of regression
- also, random effects are normally distributed with mean 0
2
Q
Path Analysis
A
- multivariate normality
3
Q
CFA
A
- multivariate normality
- large sample size
- continuous variables
- identified models
4
Q
EFA
A
- latent underlying factors
- common factors standardised
- common factors uncorrelated
- specific factors uncorrelated
- common and specific uncorrelated with one another
5
Q
Factor Scores
A
- equal weights for each item
6
Q
Regression
A
- error is normally distributed, mean of 0 and independent of each other
- homoscedasticity
7
Q
ANCOVA
A
- normality
- homogeneity of variance
- linearity b/w covariates
- linearity b/w covariates and DV
- homogeneity of regression
- independence of covariate and factor
- equal sample sizes
- no multivariate outliers
- not too high multiple covariance
8
Q
MANOVA
A
- multivariate normality
- correlated DVs
- homogenity of covariance matrices
9
Q
Logistic Regression
A
- binary outcomes are mutually exclusive
- independent observations
- independent variables: continuous or categorical
10
Q
Loglinear Regression
A
- each case in one and only one cell
- 5x as many cases as variables
- all more than 1, less than 20% less than 5
- standardised residuals normally distributed, no pattern when plotted against observed values
11
Q
MDS
A
- rship b/w proximity data and derived from distances if smooth
- want smooth transformation plot
- degeneracy: points of representation are located in a few tight clusters
