Date:         Sat, 10 Feb 2007 10:30:39 -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: cost analysis
Comments: To: "Bereti, Diana" <DBereti@mtsinai.on.ca>
In-Reply-To:  <35F00980C17ED8118F31000BCDEF3459052C2CC3@MSHMAIL10.mountsi
              nai.hospital>
Content-Type: text/plain; charset="us-ascii"; format=flowed

At 02:53 PM 2/8/2007, Bereti, Diana wrote:

>Please advise what would be the best way to handle cost analysis
>(patient in
>house visits versa telehealth) in SPSS?
>We have a very small database of 16 patients.

I see you haven't had an answer. It's not surprising; you gave far too
little information to work with.

With 16 patients, you've not got much chance with inferential
statistical methods. MAYBE with a very simple model, if there are
marked differences between groups. Even then, you'd probably need to
mistrust population variances estimated from a group that small, so
your results would inevitably be in question.

You don't even tell us what you measured.

May I recommend finding a statistical methodologist, a statistician
familiar with study design, in your area? You may need to go back to
the beginning. Good advice in how to do so will help you a lot.