LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (March 2004, week 3)Back to main SAS-L pageJoin or leave SAS-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:   Tue, 16 Mar 2004 09:43:07 -0500
Reply-To:   Peter Flom <flom@NDRI.ORG>
Sender:   "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:   Peter Flom <flom@NDRI.ORG>
Subject:   Re: Sas/Stat Question?
Comments:   To: michael.ross@ASTRAZENECA.COM
Content-Type:   text/plain; charset=US-ASCII

Before you can figure out what PROC you want in SAS STAT, you have to figure out what statistic you want in general. And before we can help you do that, you need to tell us what your variables are. Below, I take some guesses, but guesses have a way of being wrong.

Assuming that the first column is an ID number, you have nested data. This can be dealt with, but it's complex. The second and third column seem fairly clear to be date and sex. The fourth column looks like a time variable, but either you are doing your experiments in the middle of the night, or you have to recode to make AM and PM clear.

It's not so important how many YEARS of data you have, but rather how many people you have data on (unless you think hand dominance will change over the years; this could be, but probably not for the most recent 20 years - attempts to "force" left handers to be right handers started to fade quite a while back.

Next, from your statement that 'they want statistics on hand dominance male vs. female for different activities" it seems like the model you want is something like

hand pref = fx(sex + activity)

If I am correct that you have nested data, this woiuld require PROC NLMIXED; this is a very powerful PROC, and should be used more often, IMHO, but it's not simple to use. The fact that you have other covariates available is good, but do you have any reason to suspect that they are related to hand dominance? (maybe age would be.....). As to your hypothesis that hand dominance might vary by day of the week.....well, it can be investigated, but it seems far fetched to me.

Finally, as to 'p-values, std dev. and any other calculations you can see that might be useful......' ..... well, p-value of WHAT exactly? Std dev. of what?

In short, you have a complex set of problems. If you are not statsticially sophisticated, I suggest you hire a consultant.

HTH

Peter

Peter L. Flom, PhD Assistant Director, Statistics and Data Analysis Core Center for Drug Use and HIV Research National Development and Research Institutes 71 W. 23rd St www.peterflom.com New York, NY 10010 (212) 845-4485 (voice) (917) 438-0894 (fax)

>>> "Ross, Michael D" <michael.ross@ASTRAZENECA.COM> 3/16/2004 9:19:52 AM >>> Hi All, I'm very new to Sas/Stat. I'm trying to figure out how to calculate p-value, std deviation and any other calculations that you can see that would be useful for the data below: I have about 20 years worth of data where they want statistics on hand dominance Men vs. Women during specific activities (so the hand can change pending upon the activity). I also have variables like weight, height, DOB.

Also, I'd like to find out maybe who felt more dominant on a specific day of the week.

1 Male 01JAN2000 0900 Left Handed Frizbee 1 Male 02JAN2000 0700 Left Handed Basketball 1 Male 03JAN2000 0900 RightHanded Baseball 2 Male 04JAN2000 0400 RightHanded Tennis 2 Male 05JAN2000 0100 Left Handed Basketball 2 Male 06JAN2000 0200 RightHanded Football 3 Female16JAN2000 0200 RightHanded Baseball 3 Female26JAN2000 0200 Left Handed Baseball 3 Female06JAN2000 0200 RightHanded Basketball

Thanks for all your help! Mike


Back to: Top of message | Previous page | Main SAS-L page