topic Re: Intraclass correlation for two variables in a data set without a data step in SAS Programming

Intraclass correlation for two variables in a data set without a data step

sms1891 — Wed, 26 Oct 2022 19:55:41 GMT

Dear experts,

I need to find the intraclass correlation coefficient (ICC) between two different variable (X vs Y) in a data set. I have a list of 10 subjects and measurements were taken at time1 (variable X) and at time2 (variable Y). Similarly I have to compare X1 vs Y1, X2 vs Y2,....X34 vs Y34. I would really appreciate your help with this. I do not want to do a data step for each of these comparisons 34 times. Is there a better way to handle this?

Data variables/ columns in the data set:

ID X Y X1 Y1 X2 Y2 X3 Y3....X34 Y34

Thanks a bunch in advance!

Re: Intraclass correlation for two variables in a data set without a data step

Ksharp — Thu, 27 Oct 2022 12:03:58 GMT

https://support.sas.com/kb/25/031.html
https://support.sas.com/kb/30/333.html#errinvar

Re: Intraclass correlation for two variables in a data set without a data step

sms1891 — Thu, 27 Oct 2022 15:06:59 GMT

Thanks for the reply and the link Ksharp, but I am still confused. Do you have any examples which fit my case scenario and output interpretation?

Re: Intraclass correlation for two variables in a data set without a data step

Ksharp — Fri, 28 Oct 2022 11:34:15 GMT

Ksharp:
/*According to the example from URL,
You need to change your data structure .
Did you read it ?
*/

data have;
           input id x1 y1 x2 y2 x3 y3;
           output;
     datalines;
1  5.9     5.8     5.5     5.6     5.4     5.5
2   5.9     5.7     5.7     5.8     5.5     5.2
3   5.9     5.8     5.7     5.6     5.4     5.2
4     5.9     5.7     5.8     5.7     5.4     5.3
5 5.8     5.7     5.3     5.4     5.4     5.1
6  5.8     5.7     5.6     5.6     5.4     5.4
7   5.9     5.8     5.5     5.3     5.4     5.2
8 5.9     5.8     5.7     5.6     5.4     5.2
9      5.9     5.8     5.3     5.7     5.5     5.3
;
proc transpose data=have out=temp;
by id;
var x: y:;
run;
data want;
 set temp;
 by=compress(_name_,,'kd');
 if _name_=:'x' then time=1;
 if _name_=:'y' then time=2;
run;
proc sort data=want;
by by id time;
run;

proc nested data=want;
by by;
  class id;
  var col1;
  run;