| Date: | Mon, 23 Apr 2007 17:23:50 -0400 |
|---|
| Reply-To: | Peter Flom <peterflomconsulting@mindspring.com> |
|---|
| Sender: | "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU> |
|---|
| From: | Peter Flom <peterflomconsulting@MINDSPRING.COM> |
|---|
| Subject: | Re: Chi-square analysis with zero proportions |
|---|
|
| Content-Type: | text/plain; charset=UTF-8 |
|---|
abc <bc_abc2@YAHOO.COM> wrote
>I would like to see if there are significant difference in proportions
>between treatments but, for some groups, my proportions=0. How do I
>go about making these comparisons? Is this possible?
>
>For example, the data =
>
>Group A=0.92
>Group B=0.55
>Group C=0
>
>Can the proportion be transformed to be>0 or can the comparison not be
>made? Thanks.
It would be nice to have a name.....This one sounds kind of like a homework problem (I'm not saying it IS a HW problem, but it sounds like one)
You have not given us enough information to be able to do much, except to say "Yes, it is possible".
That is, it is possible to compare proportions when one is 0. But since you haven't told us what you want to compare to what, or the sample size, or where the data came from, or any other context, I don't think it's possible to say more than that.
Peter
|
