topic Pooling ICCs after/within MIANALYZE in Statistical Procedures

Pooling ICCs after/within MIANALYZE

halladje — Sun, 17 Nov 2019 20:31:04 GMT

Hello there,

I have done multilevel multiple imputation and I am running a 3-level model using GLIMMIX. I have ran and pooled my results using PROC MIANALYZE but this only prints fixed effects and I am wondering if there is a way to print random effects i.e. the random intercepts for level 2 and level 3 so I can calculate the associated pooled ICC?

Thanks in advance,

Jillian

Re: Pooling ICCs after/within MIANALYZE

SAS_Rob — Tue, 19 Nov 2019 13:03:55 GMT

You can also us MIANALYZE to combine the random effects as well. You would have to do something like the example below.

data hlm3;
seedval = 9273448;
do _imputation_=1 to 3;
do school = 1 to 5;
call rannor(seedval,usi);
call rannor(seedval,ust);
do class = 1 to 3;
call rannor(seedval,uci);
call rannor(seedval,uct);
do pupil = 1 to 10;
call rannor(seedval,upi);
call rannor(seedval,upt);
do time = 1 to 3;
call rannor(seedval,e);
y = 1 + usi + uci + upi +
time*(1 + ust + uct + upt) + e;
output;
end;
end;
end;
end;end;
run;

proc mixed data=hlm3;
by _imputation_;
class school;
model y = time / s;
random int time / subject=school solution;;
ods output solutionr=rparms(rename=(stderrpred=stderr));
run;
/*Skip this step and remove the BY statement if you do not have*/
/*the SUBJECT= option on the RANDOM statement*/
proc sort data=rparms;
by school _imputation_;
run;
proc mianalyze parms=rparms;
by school;
modeleffects intercept time;
run;

Re: Pooling ICCs after/within MIANALYZE

halladje — Tue, 19 Nov 2019 16:19:43 GMT

Thanks for the suggestion! I tried using your suggested code and the rparms dataset was not created.

This is the code I used - let me know your thoughts:

proc glimmix data=fullblimp noclprint method=rspl noclprint ;
class x_idschool x_class_id ;
model ov=
/CL dist=normal link=identity solution cl;
by _imputation_;
random intercept/subject=x_idschool type=VC;
random intercept/subject=x_class_ID (x_idschool) Type=VC;
ods output solutionr=rparms(rename=(stderrpred=stderr));
run;
quit;

proc sort data=rparms; by _imputation_ x_idschool x_class_ID;run;
proc mianalyze parms=rparms;
by x_idschool x_class_ID;
modeleffects intercept;
run;

Re: Pooling ICCs after/within MIANALYZE

SAS_Rob — Tue, 19 Nov 2019 16:28:34 GMT

You'll need the SOLUTION option on the RANDOM statement

Re: Pooling ICCs after/within MIANALYZE

halladje — Wed, 20 Nov 2019 02:25:14 GMT

Still running into errors - let me know if you see an issue with the code.

Thanks so much!

input:

**ICCS*;
proc glimmix data=fullblimp noclprint method=rspl noclprint ;
class x_idschool x_class_id ;
model ov=
/CL dist=normal link=identity solution cl;
by _imputation_;
random intercept/subject=x_idschool type=VC solution;
random intercept/subject=x_class_ID (x_idschool) Type=VC solution;
ods output solutionr=rparms(rename=(stderrpred=stderr));
run;
quit;

proc sort data=rparms; by _imputation_ x_idschool x_class_ID;run;
proc mianalyze parms=rparms;
by x_idschool x_class_ID;
modeleffects intercept;
run;

output:

proc sort data=rparms; by _imputation_ x_idschool x_class_ID;run;
ERROR: Variable X_IDSCHOOL not found.
ERROR: Variable X_CLASS_ID not found.

proc mianalyze parms=rparms;
40 by x_idschool x_class_ID;
ERROR: Variable X_IDSCHOOL not found.
ERROR: Variable X_CLASS_ID not found.
41 modeleffects intercept;
42 run;

Re: Pooling ICCs after/within MIANALYZE

halladje — Wed, 08 Jan 2020 17:07:25 GMT

Hi there,

I am still experiencing issues with pooling ICCs. Please let me know if you have any further suggestions.

Thanks so much,

Jillian