Multiple imputation confirmatory factor analysis (imputation phase with categorical variables)

TITLE:

section 8.16 data analysis example 3: confirmatory factor analysis (imputation phase);

DATA:

file = eatingattitudes.dat;

VARIABLE:

! id = participant id;

! eat1-eat21 = eating attitudes test items;

! bmi = body mass index;

! wsb = western standards of beauty;

! anx = anxiety;

names = id eat1 eat2 eat10 eat11 eat12 eat14 eat24 eat3 eat18 eat21 bmi wsb anx;

! select variables for analysis;

usevariables = eat1 - anx;

! specify missing value code;

missing are all (-99);

ANALYSIS:

! specify saturated imputation model;

type = basic;

! random number seed for mcmc algorithm;

bseed = 98124;

DATA IMPUTATION:

! incomplete variables to be imputed;

! (c) denotes a binary or an ordinal variable;

impute = eat1 (c) eat2 (c) eat10 (c) eat12 (c) eat24 (c) eat18 (c) eat21 (c) 
   bmi wsb anx;

! number of imputed data sets;

ndatasets = 100;

! filename prefix for imputed data sets;

save = eatimp*.dat;

! number of between-imputation iterations;

thin = 100;

OUTPUT:

! tech8 gives the potential scale reduction factor convergence diagnostic;

tech8;

Questions or suggestions? Email Craig Enders