BookmarkSubscribeRSS Feed
UnGiffted
Calcite | Level 5

Hello,

 

I was wondering if there was a quick workaround to a problem I'm having with my data in SAS 9.4. Because of the way my data is structured, I'm not sure how to get the results I want. I'm using SAS to analyze the effects that things like age, sex, and region have on some metrics I collected from muscle attachment sites (such as porosity, bone growth, etc). I have been working with another graduate student in a statistics consulting service on my campus, but I'm not able to go back and visit him, so I'm having some trouble. Here is the code:

 

proc glm data=data_full plots = diagnostics;
class Site Age Pre_Post Sex Region Enthesis;
model metric = Age Pre_Post Sex Region Enthesis/ ss3 solution;
lsmeans age/pdiff adjust = tukey;
run;

 

Here is a sample of my data and the organization of the columns:

 

1 - ID number (all of the 96's are the same individual, all of the 172's are the same individual)

2 - different muscle attachment

3-18 - various metrics, such as bone growth, bone porosity, etc. 

19 - site

20 - age

21 - time period

22 - sex

23 - region 

 

96 SUBSCAPULARIS 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 STENJEVEC YA PREOTTO F CONTINENTAL
96 SUPRASPINATUS 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 STENJEVEC YA PREOTTO F CONTINENTAL
96 INFRASPINATUS 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 STENJEVEC YA PREOTTO F CONTINENTAL
96 TERESMINOR 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 STENJEVEC YA PREOTTO F CONTINENTAL

172 SUBSCAPULARIS 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 STENJEVEC MA PREOTTO M CONTINENTAL
172 SUPRASPINATUS 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 STENJEVEC MA PREOTTO M CONTINENTAL
172 INFRASPINATUS 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 STENJEVEC MA PREOTTO M CONTINENTAL
172 TERESMINOR 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 STENJEVEC MA PREOTTO M CONTINENTAL

 

What I'm getting with my code:

- for bone porosity, the triceps is different than the supraspinatus, and the triceps has a higher score.

 

What I want:

- summary statistics for each muscle attachment (what the mean value is for each of the metrics at each enthesis)

- a comparison of sex vs muscle attachment score (I'm kind of getting this now, but I'm getting "for bone porosity, males are higher than females" when what I want is, "for bone porosity, males are higher than females at this muscle attachment")

 

I hope this all makes sense. Thanks in advance!

 

1 REPLY 1
ballardw
Super User

If SITE is your "muscle attachment" location variable then add that to the MODEL statement. If a different variable make sure the appropriate variable is on both the CLASS and Model statements.