Date:         Mon, 19 Mar 2012 17:56:47 +0000
Reply-To:     Joan Casellas <jcasellasvega@HOTMAIL.CO.UK>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Joan Casellas <jcasellasvega@HOTMAIL.CO.UK>
Subject:      FW: random sample of cases by groups
In-Reply-To:  <000901cd05f9$0603f290$120bd7b0$@casellas@eu-survey.com>
Content-Type: multipart/alternative;








Hi Everyone,
 
I was wondering if someone could help me . I’m trying to select a random sample base on three different variables (at the same time). So far I’ve got a syntax that does it based on one variable (see syntax below). The variables and the proportion I would need to use are described next:
 
 
 




 

total


1


Count

Column N %


Gender

Female

1000

100.0%


Total

1000

100.0%


Age

18-24

400

40.0%


25-34

290

29.0%


35-44

180

18.0%


45-54

100

10.0%


55-64

20

2.0%


65+

10

1.0%


Total

1000

100.0%


Region

East Anglia

-



East Midlands & West Midlands & East Anglia 

100

10.0%


Northern & Yorkshire/Humberside

150

15.0%


Northern Ireland

-



Northwest

120

12.0%


Scotland

140

14.0%


Southeast

400

40.0%


Southwest

60

6.0%


Wales

30

3.0%


West Midlands

-



Yorkshire/Humberside

-



Total

1000

100.0%


Social Grade

A

60

6.0%


B

270

27.0%


C1

380

38.0%


C2

170

17.0%


D

100

10.0%


E

20

2.0%



Total

1000

100.0%
 
 
Any suggestions will be really appreciate it.
 
Thanks in advance!!!!
 
Joan
 
 
COMPUTE SCRAMBLE=UNIFORM(1).
SORT CASES BY AGE SCRAMBLE.
IF $CASENUM=1 OR (LAG(age) NE age) Counter=1.
IF MISSING(Counter) Counter=LAG(Counter)+1.
COMPUTE Keeper=Age.
RECODE Keeper (1=48)(2=100)(3=150)(4=125)(5=86)(6=15).
 
SELECT IF (Counter LE Keeper).
FREQ Age.
                                          

[text/html]