Date:         Tue, 16 Nov 1999 03:45:04 +0100
Reply-To:     "J.A. Muller" <muller@CORNNET.NL>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "J.A. Muller" <muller@CORNNET.NL>
Organization: WorldOnline News server
Subject:      Re: Proc Report

Leonora,

I came accross this problem some time ago.

The reason the across defenition rsults in columns with a Cn name is quite
simple. Since you can use multiple accross variable defenitions proc report
would have to be able to combine different character string values into one.
Since the frequency  (of the class variable) is not known beforehand the
stability could not be certain.

Proc report was designed for printing to paper, and the OUT= option was
added later (and can be very handy).

If you compare the across to, for exapmle proc transpose or ms-access's
PIVOT you'l find that they have their own limitations.

hope this clarifies the problem.

Regards,
Joost

leonora dela cruz <lcdelac@pacbell.net> wrote in message
news:382FAB29.40EE@pacbell.net...

> .....
> What could have caused the across variable in the resulting dataset to
> come out as C1(label of first value), C2(label of first percent), C3(
> label of second value), C4(label of second percent) instead of the
> formatted labels.
>
> Any opinion or reply will be greatly appreciated.
>
> Thanks.