| Date: | Tue, 28 Nov 2000 16:30:40 -0000 |
|---|
| Reply-To: | Shail Dobson <dobson_shail@HOTMAIL.COM> |
|---|
| Sender: | "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU> |
|---|
| From: | Shail Dobson <dobson_shail@HOTMAIL.COM> |
|---|
| Subject: | Re: Correlation Analysis |
|---|
|
| Content-Type: | text/plain; format=flowed |
|---|
Talino,
In my opinion, if your data are interval in nature, even though measures are
in three levels, I still use the Pearson’s. You may also consider recoding
the data values either interval or ordinal. If changing to Ordinal-level,
then you will use Spearman. Either way, I am sure it will give you an idea
how these variables are related.
Shail
>From: Talino Bruno <tbruno@AXION.NET>
>Reply-To: Talino Bruno <tbruno@AXION.NET>
>To: SPSSX-L@LISTSERV.UGA.EDU
>Date: Mon, 27 Nov 2000 20:10:35 -0800
>
>Could you suggest a way for me to do a correlation between variables with 3
>levels each. I don't think I can do the Spearman as that is for continuous
>variables. Any suggestions?
>
>Take care,
>Talino
_____________________________________________________________________________________
Get more from the Web. FREE MSN Explorer download : http://explorer.msn.com
|
