Date:         Tue, 27 Jan 2009 13:29:31 -0500
Reply-To:     Paul St Louis <pstloui@DOT.STATE.TX.US>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Paul St Louis <pstloui@DOT.STATE.TX.US>
Subject:      Re: Proc Transpose

Another good solution. Thanks to all. You guys are awesome.

On Tue, 27 Jan 2009 09:33:34 -0800, Jack Hamilton <jfh@STANFORDALUMNI.ORG>
wrote:

>If you're willing to live with some restrictions on the input and
>output, you can do it with a single PROC REPORT:
>
>=====
>data test;
>   input @1 cont_id $9. @14  prsnl_t $10. @28 nbr_prsn 2.;
>datalines;
>000902050    SKILLED       20
>000902050    SKILLED       12
>000902050    SUPERVISOR    10
>000902050    SUPERVISOR     1
>000902050    UNSKILLED     49
>000902050    UNSKILLED     30
>;;;;
>
>proc report data=test nowindows missing
>       out=summary (drop=_break_
>                    rename=(_c2_=skilled _c3_=supervisor
>                    _c4_=unskilled));
>   column cont_id prsnl_t , nbr_prsn nbr_prsn=total;
>   define cont_id  / group;
>   define prsnl_t  / across;
>   define nbr_prsn / sum;
>   define total    / sum;
>run;
>=====
>
>Obs     cont_id     skilled    supervisor    unskilled    total
>
> 1     000902050       32          11            79        122
>
>=====
>
>If you don't know in advance what the values of PRSNL_T will be, this
>approach won't work (or would have to be modified).