vew all observation in proc univariate

[podarum]

Hi, i'm sure this is such a simpel task, and I'm sure I've done before, amny years back.  But for the life of me, I can't seem to remember how I can view all the observations after a proc univariate..

Ex.

A   5

A   6

A   7

B  10

B  12

B  14
after getting median in proc univarite I want to see

A   5   6

A   6   6

A   7   6

B  10  12

B  12  12

B  14  12

rather than

A    6

B  12

thank you.

[Linlin]

you can merge the two datasets:

data one;

input var$ num;

cards;

A   5

A   6

A   7

B  10

B  12

B  14

;

data two;

input var $ m;

cards;

A 6

B 12

;

data three;

merge one two;

by var;

run;

proc print;run;

[podarum]

Yeah, that's what I ended up doing.. But for some reason I thought there was something in the procedure that I could add to do that...  Thanks

[Tom]

You can do it with PROC SQL as it will automatically remerge the aggregate function results when you also select the detailed variables.

data have ;

  input group $ val @@;

cards;

A 5 A 6 A 7 B 10 B 12 B 14

run;

proc sql noprint ;

create table want as

   select group,val,mean(val) as mean_val

   from have

   group by group

;

quit;

NOTE: The query requires remerging summary statistics back with the original data.

NOTE: Table WORK.WANT created, with 6 rows and 3 columns.

A 5 6

A 7 6

A 6 6

B 14 12

B 12 12

B 10 12

[Haikuo]

Tom,

OP was asking for Median, but somehow median() is not support for group computation by proc SQL.

Haikuo