04-28-2018 05:40 AM
Hi everyone, I am trying to conduct a confirmatory factor analysis of a 7-factor structure (identified through exploratory factor analysis) consisting of 39 items that were measured on a 5-point likert scale. Since my 39 variables are ordinal, it does not meet the multivariate normality assumption. Therefore, I am attempting to use the robust maximum likelihood using the method=mlsb using proc calis. Is this possible with SAS 9.4? If not, what can I do to use this method with SAS 9.4?
Thank you in advance!
04-29-2018 12:48 AM
my code is:
proc calis data=ifsns_25_3c modification method=mlm; factor extendedbf ---> snpre_74 snpre_67 snpre_73 snpre_76 snpre_75 snpre_68 snpre_77 snpre_70, mottocomply ---> snpre_6 snpre_1 snpre_4 snpre_5 snpre_14 snpre_8 snpre_3, exbf ---> snpre_38 snpre_45 snpre_44 snpre_43 snpre_39, public ---> snpre_83 snpre_84 snpre_80 snpre_82 snpre_85; pvar extendedBF mottocomply exbf public = 4*1.; by fa; run;
and I get the error message: "Assuming the symbol ML was misspelled as mlm." but I meant mlm or mlsb for the Satorra-Bentler method. Can anyone help with this? Thank you!
Need further help from the community? Please ask a new question.