Power analysis simulation for a planned missing data design

TITLE:

section 1.13 power analysis (r = .30);

MONTECARLO:

! variable names;

names are q1-q8;

! sample size;

nobservations = 300;

! number of artificial data sets;

nreps = 5000;

! random number seed;

seed= 56798;

! specify missing data patterns (1 = missing);

patmiss = 

q1(0) q2(0) q3(0) q4(0) q5(0) q6(0) q7(1) q8(1)|

q1(0) q2(0) q3(0) q4(0) q5(1) q6(1) q7(0) q8(0)|

q1(0) q2(0) q3(1) q4(1) q5(0) q6(0) q7(0) q8(0);

! missing data pattern proportions;

patprobs = .333 | .333 | .334;

MODEL POPULATION:

! population model and parameter values;

[q1 - q8*0];

q1 - q8*1;

q1 - q8 with q1 - q8*.30;

ANALYSIS:

estimator = ml;

MODEL: 

! analysis model;

[q1 - q8*0];

q1 - q8*1;

q1 - q8 with q1 - q8*.30;

Questions or suggestions? Email Craig Enders