| Date: | Wed, 2 Nov 2005 15:02:32 -0500 |
|---|
| Reply-To: | Richard Ristow <wrristow@mindspring.com> |
|---|
| Sender: | "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU> |
|---|
| From: | Richard Ristow <wrristow@mindspring.com> |
|---|
| Subject: | Re: brain teaser |
|---|
|
| In-Reply-To: | <20051102173356.19096BA86E3@smtpgate.email.arizona.edu> |
|---|
| Content-Type: | text/plain; charset="us-ascii"; format=flowed |
|---|
At 12:33 PM 11/2/2005, Jean Campbell wrote:
>I'm not sure if this is an SPSS question..but I've lost patience.
It's certainly a data-transformation problem that can be solved in
SPSS, of a type that very commonly is.
It shouldn't be hard, either; a modest exercise in AGGREGATE. But,
>I am trying to determine the % of subjects who experience an adverse
>event (symptom) by their treatment group (1,2, or 3).
Here's the problem
>Each subject could have reported multiple adverse events, so the data
>looks
>like this:
>
>Subj ID AE Txt group
>1 2 1
[...]
It looks like you have a record for each AE experienced by each
subject; and you have the treatment group for each subject. But what
record do you have of subjects who have no adverse events? Without
that, I don't see how you can calculate thing like "% of subjects who
experience an adverse event", except by counting the number of subjects
with an AE and dividing by hand.
| 
