Replicate the bootstrap
eigenvalue estimates for the subsample of Marsh SDQ data presented
in Results for the first four eigenvalues using ULS extraction. Select the subsample of 300 using a seed value of 39302.
(See Chapter 6 for the syntax.) Then update the syntax provided in
the text to conduct an EFA for the data and estimate bootstrapped
CIs for the eigenvalues. Use a seed value of 1 for the bootstrap resampling
from the subsample. When you set the seed values, your results should
match those presented in the table exactly. If you have trouble with
this syntax, check your code against that provided on the book website.
Using the same subsample
of 300 records, replicate the bootstrapped results for the factor
loadings presented in Pattern matrix loadings using ULS extraction and oblimin rotation. Try estimating
the CI before running the %checkFactorOrder macro on the data and
then estimate them after running the macro. Notice the difference
in the estimates.
Using the GDS data set,
conduct an EFA and estimate bootstrapped 95% CI around the initial
eigenvalues. Use 2000 resamples of the data. What do these results
suggest about the factor structure of the GDS data?