Assumptions Flashcards

1
Q

Multilevel Modelling

A
  • all of regression

- also, random effects are normally distributed with mean 0

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Path Analysis

A
  • multivariate normality
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

CFA

A
  • multivariate normality
  • large sample size
  • continuous variables
  • identified models
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

EFA

A
  • latent underlying factors
  • common factors standardised
  • common factors uncorrelated
  • specific factors uncorrelated
  • common and specific uncorrelated with one another
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Factor Scores

A
  • equal weights for each item
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Regression

A
  • error is normally distributed, mean of 0 and independent of each other
  • homoscedasticity
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
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
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

MANOVA

A
  • multivariate normality
  • correlated DVs
  • homogenity of covariance matrices
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Logistic Regression

A
  • binary outcomes are mutually exclusive
  • independent observations
  • independent variables: continuous or categorical
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
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
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
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
How well did you know this?
1
Not at all
2
3
4
5
Perfectly