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 (May 2002)Back to main SPSSX-L pageJoin or leave SPSSX-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Thu, 2 May 2002 03:23:21 -0400
Reply-To:     Frank Thomas <>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Frank Thomas <>
Subject:      Re: PPS Sampling
Comments: To: Seumas Rogan <>
Content-Type: text/plain; charset=iso-8859-1


if you have so few cases, why don't you use the full sample instead of making a complicated sampling WITHIN A CITY ??

Good luck

Frank Thomas

>Hi everyone - > >I have a questions regarding sampling. My questions are theoretical >and practical, i.e., I need advice regarding statistical/sampling issues >and SPSS syntax to do what i need to do once i know what i want to do... > >I have visited Raynald's website (THANKS! Raynald); he has several scripts >that address what I want to do, but not exactly. I'm having trouble >extending what Raynald has to what I would like to do. > >I want to draw a sample of households within census blockgroups in my >city. I have the 2000 census data for these blockgroups. I want the sample >to be representative of the city population (or at least representative of >the population captured by the 2000 census). Since the census block groups >are not of equal size, i believe that probability proportional to size >sampling is the correct sampling approach. yes? > >The census block group is my PSU. I want to stratify the PSUs by region >within the city (10 classes), median income, percent Hispanic, median age >of housing (all continuous). Further, I want to 'oversample' certain PSU's >with minority populations, and within certain regions, to make sure that I >get sufficient sample size in my analyses. I understand that oversampling >will require the use of sampling weights in my analyses. > >Let's say that I want to draw 50 PSUs, and then randomly sample 25 >households within each PSU. Any advice on how to draw this sample in SPSS? >Any advice on good 'how to' texts on sampling in general? > >Thank you for any help/advice! > >SR > > >

-- Dr. Frank Thomas FTR 34 rue Jean-Baptiste Pigalle 75009 Paris France

__________________________________________________________________ Your favorite stores, helpful shopping tools and great gift ideas. Experience the convenience of buying online with Shop@Netscape!

Get your own FREE, personal Netscape Mail account today at

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