Date: Thu, 2 Feb 2006 23:12:31 -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: Random ditribution
In-Reply-To: <5263A45F57547546AB4F295AA26A38D321C37C@eqaoexc01.eqao.com>
Content-Type: text/plain; charset=us-ascii; format=flowed;
x-avg-checked=avg-ok-6B7C419A
At 05:04 PM 2/2/2006, Ebby Madera asked (off-list):
>Thank you very much for the explaination and syntax. I am in the
>process of applying it.
I'd written,
>>Here's a second inefficiency, and it might matter with a long enough
>>list of schools (though I'll guess that 4,000 is not a long enough
>>list.)
>>
>>Logic below has two full-file sorts, and not easy ones - records are
>>moved far from their initial positions, and back again.
>>
>>SORT CASES BY CURRNUMB.
>>MATCH FILES
>> /FILE=*
>> /TABLE=CURRICULA
>> /BY CURRNUMB.
>>SORT CASES BY SCHLNUMB CURRNUMB.
and she asked,
> I will appreciate if you sent me the look-up-table logic.
Here goes. For the process of assigning curricula randomly to schools,
see my posting in this thread, Wed, 1 Feb 2006 16:13:27 -0500.
As drawbacks, this method leaves one initial useless record in the
output, plus the variables CURR01 TO CURR16.
The 'LIST' commands are included for illustration, of course. In a
large file, they'd slow processing noticeably; each causes a full pass
through the data, as EXECUTE does.
* The list of curricula and their names .
DATASET ACTIVATE Curricula.
LIST.
List
|-----------------------------|---------------------------|
|Output Created |02-FEB-2006 22:58:09 |
|-----------------------------|---------------------------|
[Curricula]
CURRNUMB CURRNAME
01 Alpha
02 Beta
03 Gamma
04 Delta
05 Epsilon
06 Zeta
07 Eta
08 Theta
09 Iota
10 Kappa
11 Lambda
12 Mu
13 Nu
14 Xi
15 Omicron
16 Pi
Number of cases read: 16 Number of cases listed: 16
* The schools, with curriculum numbers attached .
GET FILE=Result.
DATASET NAME Result.
LIST.
List
|-----------------------------|---------------------------|
|Output Created |02-FEB-2006 22:58:09 |
|-----------------------------|---------------------------|
[Result] C:\Documents and Settings\All Users\Documents
\Temporary\SPSS
\2006-02-01 Madera - Random ditribution.SAV
SCHLNUMB SCHLNAME CURRNUMB
01 Alpine 02
01 Alpine 03
01 Alpine 04
01 Alpine 12
02 Bonaparte 07
02 Bonaparte 12
02 Bonaparte 15
02 Bonaparte 16
03 Celsius 05
03 Celsius 08
03 Celsius 10
03 Celsius 13
04 Davenport 08
04 Davenport 09
04 Davenport 14
04 Davenport 15
05 Equality 05
05 Equality 06
05 Equality 13
05 Equality 15
Number of cases read: 20 Number of cases listed: 20
* First, "flatten" the list of curricula, to one .
* record and 16 variables. This is the lookup .
* table. .
DATASET ACTIVATE Curricula.
DATASET COPY Flattened.
Dataset Copy
|----------------------------|---------------------------|
|Output Created |02-FEB-2006 22:58:09 |
|----------------------------|---------------------------|
DATASET ACTIVATE Flattened.
STRING CURR01 TO CURR16 (A10).
VECTOR CURRS=CURR01 TO CURR16.
COMPUTE CURRS(CURRNUMB) = CURRNAME.
COMPUTE NOBREAK=1.
AGGREGATE OUTFILE=*
/BREAK = NOBREAK
/ CURR01 TO CURR16
= MAX(CURR01 TO CURR16).
LIST CURR01 TO CURR16.
List
|-----------------------------|---------------------------|
|Output Created |02-FEB-2006 22:58:10 |
|-----------------------------|---------------------------|
The variables are listed in the following order:
LINE 1: CURR01 CURR02 CURR03 CURR04 CURR05 CURR06
LINE 2: CURR07 CURR08 CURR09 CURR10 CURR11 CURR12
LINE 3: CURR13 CURR14 CURR15 CURR16
CURR01:
Alpha Beta Gamma Delta Epsilon Zeta
CURR07:
Eta Theta Iota Kappa Lambda Mu
CURR13: Nu Xi Omicron Pi
Number of cases read: 1 Number of cases listed: 1
* Add the lookup table to at the beginning of .
* records with schools and curricula .
ADD FILES
/FILE=*
/FILE=Result
/DROP=NOBREAK
/KEEP=SCHLNUMB SCHLNAME CURRNUMB ALL.
LIST /CASES=3.
List
|-----------------------------|---------------------------|
|Output Created |02-FEB-2006 22:58:11 |
|-----------------------------|---------------------------|
The variables are listed in the following order:
LINE 1: SCHLNUMB SCHLNAME CURRNUMB CURR01 CURR02 CURR03 CURR04
LINE 2: CURR05 CURR06 CURR07 CURR08 CURR09 CURR10
LINE 3: CURR11 CURR12 CURR13 CURR14 CURR15 CURR16
SCHLNUMB: . .
Alpha Beta Gamma Delta
CURR05:
Epsilon Zeta Eta Theta Iota Kappa
CURR11:
Lambda Mu Nu Xi Omicron Pi
SCHLNUMB: 01 Alpine 02
CURR05:
CURR11:
SCHLNUMB: 01 Alpine 03
CURR05:
CURR11:
Number of cases read: 3 Number of cases listed: 3
* 'Spread' the lookup table over all records, .
* and use it to look up curriculum names .
STRING CURRNAME(A10).
VECTOR CURRS = CURR01 TO CURR16.
DO IF $CASENUM > 1.
. DO REPEAT CURRIC = CURR01 TO CURR16.
. COMPUTE CURRIC = LAG(CURRIC).
. END REPEAT.
END IF.
IF NOT MISSING(CURRNUMB)
CURRNAME = CURRS(CURRNUMB).
. /*-- SELECT IF NOT MISSING(SCHLNUMB).
LIST /CASES=6
/VARIABLES=SCHLNUMB SCHLNAME CURRNUMB CURRNAME.
List
|-----------------------------|---------------------------|
|Output Created |02-FEB-2006 22:58:11 |
|-----------------------------|---------------------------|
SCHLNUMB SCHLNAME CURRNUMB CURRNAME
. .
01 Alpine 02 Beta
01 Alpine 03 Gamma
01 Alpine 04 Delta
01 Alpine 12 Mu
02 Bonaparte 07 Eta
Number of cases read: 6 Number of cases listed: 6
--
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.1.375 / Virus Database: 267.14.23/243 - Release Date: 1/27/2006
