| Date: | Wed, 20 Jul 2005 02:52:18 -0400 |
|---|
| Reply-To: | Richard Ristow <wrristow@mindspring.com> |
|---|
| Sender: | "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU> |
|---|
| From: | Richard Ristow <wrristow@mindspring.com> |
|---|
| Subject: | Re: Removing empty string cells and replacing with next variable
cell |
|---|
|
|
| In-Reply-To: | <6.2.1.2.2.20050719115719.01e4b3e0@mail.qut.edu.au> |
|---|
| Content-Type: | text/plain; charset="us-ascii"; format=flowed |
|---|
At 10:06 PM 7/18/2005, Kirsten McKenzie wrote:
>I have five character string data in the format:
> D01 D02 D03 D04
>1 A0001 S0002 G0059
>2 G0009 S0350
>3 S0789 S9900
>4 S9913
>I want to remove all empty string cells and move the next variable
>cell into the blank cell so it will now appear:
> D01 D02 D03 D04
>1 A0001 S0002 G0059
>2 G0009 S0350
>3 S0789 S9900
>4 S9913
>I also want to delete all empty string variables at the end of the
>dataset once all blank cells have been replaced.
As to dropping unused variables at the end of the list, I second Simon
Freidin's advice to check Raynald Levesque's Web site. (Or, of course,
decide it isn't necessary to drop them.) I believe Simon's solution to
moving values to close up empty cells has two drawbacks:
A.) It doesn't work if there are two or more empty cells contiguous
B.) It leaves variable X in the final file.
Below is SPSS draft output for Simon's solution, and an alternative
that I believe solves these problems. I think Jan Spousta has posted
with this logic, though I can't find the posting at the moment.
GET FILE=TESTDATA.
LIST.
List
Notes
|---------------------|--------------------|
|Output Created |20 Jul 05 02:47:20 |
|---------------------|--------------------|
ID D01 D02 D03 D04
1 A0001 S0002 G0059
2 G0009 S0350
3 S0789 S9900
4 S9913
5 T1201 U5342
Number of cases read: 5 Number of cases listed: 5
* Simon Freidin solution .
/* moving data across */
vector d=d01 to d04.
loop x=1 to 4 if x<4.
- do if d(x)=''.
. compute d(x)=d(x+1).
. compute d(x+1)=''.
. compute x=x+1.
- end if.
end loop.
LIST.
List
Notes
|---------------------|--------------------|
|Output Created |20 Jul 05 02:47:20 |
|---------------------|--------------------|
ID D01 D02 D03 D04 X
1 A0001 S0002 G0059 4.00
2 G0009 S0350 4.00
3 S0789 S9900 4.00
4 S9913 4.00
5 T1201 U5342 4.00
Number of cases read: 5 Number of cases listed: 5
* Jan Spousta/Ristow solution .
GET FILE=TESTDATA.
vector d=d01 to d04.
COMPUTE #DEST = 1.
LOOP #SRCE = 1 TO 4.
- DO IF D(#SRCE) NE ''.
. COMPUTE D(#DEST)=D(#SRCE).
. COMPUTE #DEST = #DEST + 1.
- END IF.
END LOOP.
LOOP #ZAP = #DEST TO 4.
. COMPUTE D(#ZAP) = ''.
END LOOP.
LIST.
List
Notes
|---------------------|--------------------|
|Output Created |20 Jul 05 02:47:22 |
|---------------------|--------------------|
ID D01 D02 D03 D04
1 A0001 S0002 G0059
2 G0009 S0350
3 S0789 S9900
4 S9913
5 T1201 U5342
Number of cases read: 5 Number of cases listed: 5
| 
