1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
*(Q) I have nine record types called 1,2,3,4,5,6,7,8, and 9.

* Record type 1 is really 2 record types lets say 0 and 1.

* I need to split off the first 40 variables if the record is a background
only record and call it record type 0.

* In summary I need to add a new record type 0 for each record type 1.

*(A) Posted to SPSS newsgroup by rlevesque@videotron.ca on 2002/01/11.

* Create a data file for illustration purposes.

DATA LIST LIST /id rectype var1 var2.
BEGIN DATA
1 1 2 3 
2 7 5 2 
3 4 8 2 
4 9 2 5 
5 6 3 3 
6 3 8 5 
7 1 9 6 
8 5 0 2 
END DATA.
LIST.

* Create a few more dummy variables.

VECTOR v(40F8).
* Fill the variables with some data.
LOOP #cnt=1 TO 40.
COMPUTE v(#cnt)=TRUNC(UNIFORM(10)).
END LOOP.

********************.
* START the job.
********************.

* Create a new record with rectype=0 for each case having rectype =1.

LOOP #cnt=1 TO 1+ (rectype=1).
+	DO IF #cnt=2.
+	COMPUTE rectype=0.
*	Get rid of unneeded variables when rectype=0.
+	LOOP #cnt1=21 TO 40.
+	COMPUTE v(#cnt1)=$SYSMIS.
+	END LOOP.
+	END IF.
+	XSAVE OUTFILE='c:\\temp\\temp.sav' /KEEP=ALL.
END LOOP.
EXECUTE.
GET FILE='c:\\temp\\temp.sav'.