Question about combining seperate answer columns in dataset into one variable

[Brenda Zollitsch]

Hi there, SPSS folks,

 

I have just been provided a dataset from surveymonkey for a project.

 

The database breaks the answers down into different columns (new variables) for each answer option. Fake example.  What is your interest level? Answers: low, medium, high.

Instead of providing me with a variable (interest level) with 1=low, 2=medium and 3 = high, the data set has one column for low (with either 0 or 1) as the data, same for medium and same for high. 

 

My question is how to combine these multiple columns into one variable with different answer options: variable =interest level with 1=low, 2=medium and 3=high.

 

I have done lots of recoding of data from one variable, but have never had to combine answers from multiple variables into one.

 

Can you point me in the right direction?

 

This is a quick project, and I am not great with syntax, so if there is anyway you can provide this via point and click tools within SPSS that would be my preference.

 

Thanks!

 

Brenda

[Maguin, Eugene]

I'll bet there are a number of ways to do this. One example:

Compute interest=9.

Do repeat x=low medium high/y=1 2 3.

+  if (x eq 1) interest=y.

End repeat.

Missing values interest(9).

 

That works if you have a few items. But if you have many, ALL with the same response format.

 

Numeric x1 a2 a3 a4 t6  t7 ww y78(f1.0).

Vector new=x1 to y78.

Loop #i=1 to >number of new variables<.

Compute new(#i)=9.

Do repeat x=low medium high/y=1 2 3.

+  if (x eq 1) new(#i)=y.

End repeat.

End loop.

Missing values x1 to y78(9).

 

If you have varied response formats, .e., 1-3, 1-6, etc, I think you have no simple choice but to chunk the code by response format.

 

Gene Maguin

 

 

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Brenda Zollitsch
Sent: Wednesday, April 11, 2012 1:21 PM
To: [hidden email]
Subject: Question about combining seperate answer columns in dataset into one variable

 

Hi there, SPSS folks,

 

I have just been provided a dataset from surveymonkey for a project.

 

The database breaks the answers down into different columns (new variables) for each answer option. Fake example.  What is your interest level? Answers: low, medium, high.

Instead of providing me with a variable (interest level) with 1=low, 2=medium and 3 = high, the data set has one column for low (with either 0 or 1) as the data, same for medium and same for high. 

 

My question is how to combine these multiple columns into one variable with different answer options: variable =interest level with 1=low, 2=medium and 3=high.

 

I have done lots of recoding of data from one variable, but have never had to combine answers from multiple variables into one.

 

Can you point me in the right direction?

 

This is a quick project, and I am not great with syntax, so if there is anyway you can provide this via point and click tools within SPSS that would be my preference.

 

Thanks!

 

Brenda

[John F Hall]

Brenda

 

Assuming that your three columns are mutually exclusive:

 

File > New > Syntax

 

Type in the following using your own col names (start in col 1 and don’t forget the full stops!)

 

compute rank = v1*100+v2*10+v3.

recode rank (100=1)(10=2).

freq rank.

 

Run > All

 

Get back if you need more help.

 

 

John F Hall (Mr)

 

Email:     [hidden email]

Website: www.surveyresearch.weebly.com

Skype:   surveyresearcher1

Phone:    (+33) (0) 2.33.45.91.47

 

 

 

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Brenda Zollitsch
Sent: 11 April 2012 19:21
To: [hidden email]
Subject: Question about combining seperate answer columns in dataset into one variable

 

Hi there, SPSS folks,

 

I have just been provided a dataset from surveymonkey for a project.

 

The database breaks the answers down into different columns (new variables) for each answer option. Fake example.  What is your interest level? Answers: low, medium, high.

Instead of providing me with a variable (interest level) with 1=low, 2=medium and 3 = high, the data set has one column for low (with either 0 or 1) as the data, same for medium and same for high. 

 

My question is how to combine these multiple columns into one variable with different answer options: variable =interest level with 1=low, 2=medium and 3=high.

 

I have done lots of recoding of data from one variable, but have never had to combine answers from multiple variables into one.

 

Can you point me in the right direction?

 

This is a quick project, and I am not great with syntax, so if there is anyway you can provide this via point and click tools within SPSS that would be my preference.

 

Thanks!

 

Brenda

[Jim Marks]

Made up example for 3 interest variables-- (interest_1 interest_2 interest_3 coded 1 = selected).

Compute interest = -9.

IF interest_1 = 1 interest = 1.
IF interest_2 = 1 interest = 2.
IF interest_3 = 1 interest = 3.

CROSSTABS interest BY interest_1 interest_2 interest_3

Cases with interest = -9 were not coded into the single variable. Cross tabs will show you if any cases are miscoded.



Jim Marks
Sr Market Research Manager
National Market Research
Kaiser Foundation Health Plan of the Mid-Atlantic States, Inc.
2101 E. Jefferson St.
Rockville, MD 20852
Phone: [phone]
Cell Phone: [phone]

NOTICE TO RECIPIENT:  If you are not the intended recipient of this e-mail, you are prohibited from sharing, copying, or otherwise using or disclosing its contents.  If you have received this e-mail in error, please notify the sender immediately by reply e-mail and permanently delete this e-mail and any attachments without reading, forwarding or saving them.  Thank you.




From:        Brenda Zollitsch <[hidden email]>
To:        [hidden email]
Date:        04/11/2012 01:22 PM
Subject:        Question about combining seperate answer columns in dataset into              one variable
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

Hi there, SPSS folks,
 
I have just been provided a dataset from surveymonkey for a project.
 
The database breaks the answers down into different columns (new variables) for each answer option. Fake example.  What is your interest level? Answers: low, medium, high.
Instead of providing me with a variable (interest level) with 1=low, 2=medium and 3 = high, the data set has one column for low (with either 0 or 1) as the data, same for medium and same for high. 
 
My question is how to combine these multiple columns into one variable with different answer options: variable =interest level with 1=low, 2=medium and 3=high.
 
I have done lots of recoding of data from one variable, but have never had to combine answers from multiple variables into one.
 
Can you point me in the right direction?
 
This is a quick project, and I am not great with syntax, so if there is anyway you can provide this via point and click tools within SPSS that would be my preference.
 
Thanks!
 
Brenda

[David Marso]

"I am not great with syntax, so if there is anyway you can provide this via point and click tools ..."
So Sorry, that COMPUTE dialog is a complete pain in the ass!  
As a general rule I will *NOT* walk anyone through a point and click session without a billing address!
Time to open a syntax window.
COMPUTE newvar=SUM(v1*1,v2*2,v3*3).
---
HTH
David
--

Brenda Zollitsch wrote

Hi there, SPSS folks,

I have just been provided a dataset from surveymonkey for a project.

The database breaks the answers down into different columns (new variables)
for each answer option. Fake example.  What is your interest level?
Answers: low, medium, high.
Instead of providing me with a variable (interest level) with 1=low,
2=medium and 3 = high, the data set has one column for low (with either 0
or 1) as the data, same for medium and same for high.

My question is how to combine these multiple columns into one variable with
different answer options: variable =interest level with 1=low, 2=medium and
3=high.

I have done lots of recoding of data from one variable, but have never had
to combine answers from multiple variables into one.

Can you point me in the right direction?

This is a quick project, and I am not great with syntax, so if there is
anyway you can provide this via point and click tools within SPSS that
would be my preference.

Thanks!

Brenda

[Rick Oliver-3]

Nice solution.

Rick Oliver
Senior Information Developer
IBM Business Analytics (SPSS)
E-mail: [hidden email]
Phone: [phone] | T/L: 206-4922



From:        David Marso <[hidden email]>
To:        [hidden email]
Date:        04/11/2012 01:07 PM
Subject:        Re: Question about combining seperate answer columns in dataset              into one variable
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

"I am not great with syntax, so if there is anyway you can provide this via
point and click tools ..."
So Sorry, that COMPUTE dialog is a complete pain in the ass!
As a general rule I will *NOT* walk anyone through a point and click session
without a billing address!
Time to open a syntax window.
COMPUTE newvar=SUM(v1*1,v2*2,v3*3).
---
HTH
David
--

Brenda Zollitsch wrote
>
> Hi there, SPSS folks,
>
> I have just been provided a dataset from surveymonkey for a project.
>
> The database breaks the answers down into different columns (new
> variables)
> for each answer option. Fake example.  What is your interest level?
> Answers: low, medium, high.
> Instead of providing me with a variable (interest level) with 1=low,
> 2=medium and 3 = high, the data set has one column for low (with either 0
> or 1) as the data, same for medium and same for high.
>
> My question is how to combine these multiple columns into one variable
> with
> different answer options: variable =interest level with 1=low, 2=medium
> and
> 3=high.
>
> I have done lots of recoding of data from one variable, but have never had
> to combine answers from multiple variables into one.
>
> Can you point me in the right direction?
>
> This is a quick project, and I am not great with syntax, so if there is
> anyway you can provide this via point and click tools within SPSS that
> would be my preference.
>
> Thanks!
>
> Brenda
>


--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Question-about-combining-seperate-answer-columns-in-dataset-into-one-variable-tp5633564p5633635.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[David Marso]

Here is how I would deal with 'many' of such abominations
(what a stupid way for a piece of SW to output a variable or do they consider
everything open game for mult response and not believe in parsimony?).
--
DATA LIST /W1 TO W6 1-6 X1 TO X3 8-10  Y1 TO Y4 12-15  Z1 TO Z2 17-18.
BEGIN DATA
000001 100 0100 10
100000 010 1000 01
000100 001 0010 10
END DATA.
** Note the potentially brain numbing double use of Var to represent
** both the VECTOR and the accumulated value ;-).
** Also might ask, where in the hell is the initialization?  
** We *don't* need any ;-)))).
VECTOR W=W1 TO W6 / X= X1 TO X3 /Y=Y1 TO Y4 /Z= Z1 TO Z2 .
DO REPEAT Var=W X Y Z /MAXCAT=6 3 4 2.
+  LOOP #=1 TO MAXCAT.
+    COMPUTE Var=SUM(Var,Var(#)*#).
+  END LOOP IF V > 0.
END REPEAT.
LIST W X Y Z.

Maguin, Eugene wrote

I'll bet there are a number of ways to do this. One example:
Compute interest=9.
Do repeat x=low medium high/y=1 2 3.
+  if (x eq 1) interest=y.
End repeat.
Missing values interest(9).

That works if you have a few items. But if you have many, ALL with the same response format.

Numeric x1 a2 a3 a4 t6  t7 ww y78(f1.0).
Vector new=x1 to y78.
Loop #i=1 to >number of new variables<.
Compute new(#i)=9.
Do repeat x=low medium high/y=1 2 3.
+  if (x eq 1) new(#i)=y.
End repeat.
End loop.
Missing values x1 to y78(9).

If you have varied response formats, .e., 1-3, 1-6, etc, I think you have no simple choice but to chunk the code by response format.

Gene Maguin


From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Brenda Zollitsch
Sent: Wednesday, April 11, 2012 1:21 PM
To: [hidden email]
Subject: Question about combining seperate answer columns in dataset into one variable

Hi there, SPSS folks,

I have just been provided a dataset from surveymonkey for a project.

The database breaks the answers down into different columns (new variables) for each answer option. Fake example.  What is your interest level? Answers: low, medium, high.
Instead of providing me with a variable (interest level) with 1=low, 2=medium and 3 = high, the data set has one column for low (with either 0 or 1) as the data, same for medium and same for high.

My question is how to combine these multiple columns into one variable with different answer options: variable =interest level with 1=low, 2=medium and 3=high.

I have done lots of recoding of data from one variable, but have never had to combine answers from multiple variables into one.

Can you point me in the right direction?

This is a quick project, and I am not great with syntax, so if there is anyway you can provide this via point and click tools within SPSS that would be my preference.

Thanks!

Brenda

[Bruce Weaver]

David's example gave me error messages until I changed

+  END LOOP IF V > 0.

to

+  END LOOP IF Var > 0.

;-)

Also, if the sets of indicators can be read in as string variables, then you could also just use the INDEX function, like this:

DATA LIST LIST /Wstring (a6) Xstring (A3) Ystring (a4) Zstring (A2).
BEGIN DATA
000001 100 0100 10
100000 010 1000 01
000100 001 0010 10
END DATA.
DO REPEAT New = W X Y Z / Old = Wstring to Zstring .
- COMPUTE New = index(Old,'1').
END REPEAT.
FORMATS W to Z (f1).
LIST.

OUTPUT:

Wstring Xstring Ystring Zstring W X Y Z

000001  100     0100    10      6 1 2 1
100000  010     1000    01      1 2 1 2
000100  001     0010    10      4 3 3 1

Number of cases read:  3    Number of cases listed:  3

David Marso wrote

Here is how I would deal with 'many' of such abominations
(what a stupid way for a piece of SW to output a variable or do they consider
everything open game for mult response and not believe in parsimony?).
--
DATA LIST /W1 TO W6 1-6 X1 TO X3 8-10  Y1 TO Y4 12-15  Z1 TO Z2 17-18.
BEGIN DATA
000001 100 0100 10
100000 010 1000 01
000100 001 0010 10
END DATA.
** Note the potentially brain numbing double use of Var to represent
** both the VECTOR and the accumulated value ;-).
** Also might ask, where in the hell is the initialization?  
** We *don't* need any ;-)))).
VECTOR W=W1 TO W6 / X= X1 TO X3 /Y=Y1 TO Y4 /Z= Z1 TO Z2 .
DO REPEAT Var=W X Y Z /MAXCAT=6 3 4 2.
+  LOOP #=1 TO MAXCAT.
+    COMPUTE Var=SUM(Var,Var(#)*#).
+  END LOOP IF V > 0.
END REPEAT.
LIST W X Y Z.

Maguin, Eugene wrote

I'll bet there are a number of ways to do this. One example:
Compute interest=9.
Do repeat x=low medium high/y=1 2 3.
+  if (x eq 1) interest=y.
End repeat.
Missing values interest(9).

That works if you have a few items. But if you have many, ALL with the same response format.

Numeric x1 a2 a3 a4 t6  t7 ww y78(f1.0).
Vector new=x1 to y78.
Loop #i=1 to >number of new variables<.
Compute new(#i)=9.
Do repeat x=low medium high/y=1 2 3.
+  if (x eq 1) new(#i)=y.
End repeat.
End loop.
Missing values x1 to y78(9).

If you have varied response formats, .e., 1-3, 1-6, etc, I think you have no simple choice but to chunk the code by response format.

Gene Maguin


From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Brenda Zollitsch
Sent: Wednesday, April 11, 2012 1:21 PM
To: [hidden email]
Subject: Question about combining seperate answer columns in dataset into one variable

Hi there, SPSS folks,

I have just been provided a dataset from surveymonkey for a project.

The database breaks the answers down into different columns (new variables) for each answer option. Fake example.  What is your interest level? Answers: low, medium, high.
Instead of providing me with a variable (interest level) with 1=low, 2=medium and 3 = high, the data set has one column for low (with either 0 or 1) as the data, same for medium and same for high.

My question is how to combine these multiple columns into one variable with different answer options: variable =interest level with 1=low, 2=medium and 3=high.

I have done lots of recoding of data from one variable, but have never had to combine answers from multiple variables into one.

Can you point me in the right direction?

This is a quick project, and I am not great with syntax, so if there is anyway you can provide this via point and click tools within SPSS that would be my preference.

Thanks!

Brenda

[David Marso]

Yeah, my mistake ;-(
I changed V to Var in all but one place after pasting the syntax.
OTOH:  It might have been deliberate to keep you all on your toes ;-)
---

Bruce Weaver wrote

David's example gave me error messages until I changed

+  END LOOP IF V > 0.

to

+  END LOOP IF Var > 0.

;-)

Also, if the sets of indicators can be read in as string variables, then you could also just use the INDEX function, like this:

DATA LIST LIST /Wstring (a6) Xstring (A3) Ystring (a4) Zstring (A2).
BEGIN DATA
000001 100 0100 10
100000 010 1000 01
000100 001 0010 10
END DATA.
DO REPEAT New = W X Y Z / Old = Wstring to Zstring .
- COMPUTE New = index(Old,'1').
END REPEAT.
FORMATS W to Z (f1).
LIST.

OUTPUT:

Wstring Xstring Ystring Zstring W X Y Z

000001  100     0100    10      6 1 2 1
100000  010     1000    01      1 2 1 2
000100  001     0010    10      4 3 3 1

Number of cases read:  3    Number of cases listed:  3

David Marso wrote

Here is how I would deal with 'many' of such abominations
(what a stupid way for a piece of SW to output a variable or do they consider
everything open game for mult response and not believe in parsimony?).
--
DATA LIST /W1 TO W6 1-6 X1 TO X3 8-10  Y1 TO Y4 12-15  Z1 TO Z2 17-18.
BEGIN DATA
000001 100 0100 10
100000 010 1000 01
000100 001 0010 10
END DATA.
** Note the potentially brain numbing double use of Var to represent
** both the VECTOR and the accumulated value ;-).
** Also might ask, where in the hell is the initialization?  
** We *don't* need any ;-)))).
VECTOR W=W1 TO W6 / X= X1 TO X3 /Y=Y1 TO Y4 /Z= Z1 TO Z2 .
DO REPEAT Var=W X Y Z /MAXCAT=6 3 4 2.
+  LOOP #=1 TO MAXCAT.
+    COMPUTE Var=SUM(Var,Var(#)*#).
+  END LOOP IF V > 0.
END REPEAT.
LIST W X Y Z.

Maguin, Eugene wrote

I'll bet there are a number of ways to do this. One example:
Compute interest=9.
Do repeat x=low medium high/y=1 2 3.
+  if (x eq 1) interest=y.
End repeat.
Missing values interest(9).

That works if you have a few items. But if you have many, ALL with the same response format.

Numeric x1 a2 a3 a4 t6  t7 ww y78(f1.0).
Vector new=x1 to y78.
Loop #i=1 to >number of new variables<.
Compute new(#i)=9.
Do repeat x=low medium high/y=1 2 3.
+  if (x eq 1) new(#i)=y.
End repeat.
End loop.
Missing values x1 to y78(9).

If you have varied response formats, .e., 1-3, 1-6, etc, I think you have no simple choice but to chunk the code by response format.

Gene Maguin


From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Brenda Zollitsch
Sent: Wednesday, April 11, 2012 1:21 PM
To: [hidden email]
Subject: Question about combining seperate answer columns in dataset into one variable

Hi there, SPSS folks,

I have just been provided a dataset from surveymonkey for a project.

The database breaks the answers down into different columns (new variables) for each answer option. Fake example.  What is your interest level? Answers: low, medium, high.
Instead of providing me with a variable (interest level) with 1=low, 2=medium and 3 = high, the data set has one column for low (with either 0 or 1) as the data, same for medium and same for high.

My question is how to combine these multiple columns into one variable with different answer options: variable =interest level with 1=low, 2=medium and 3=high.

I have done lots of recoding of data from one variable, but have never had to combine answers from multiple variables into one.

Can you point me in the right direction?

This is a quick project, and I am not great with syntax, so if there is anyway you can provide this via point and click tools within SPSS that would be my preference.

Thanks!

Brenda

[Bruce Weaver]

I assumed the latter.  ;-)

David Marso wrote

Yeah, my mistake ;-(
I changed V to Var in all but one place after pasting the syntax.
OTOH:  It might have been deliberate to keep you all on your toes ;-)
---

Bruce Weaver wrote

David's example gave me error messages until I changed

+  END LOOP IF V > 0.

to

+  END LOOP IF Var > 0.

;-)

Also, if the sets of indicators can be read in as string variables, then you could also just use the INDEX function, like this:

DATA LIST LIST /Wstring (a6) Xstring (A3) Ystring (a4) Zstring (A2).
BEGIN DATA
000001 100 0100 10
100000 010 1000 01
000100 001 0010 10
END DATA.
DO REPEAT New = W X Y Z / Old = Wstring to Zstring .
- COMPUTE New = index(Old,'1').
END REPEAT.
FORMATS W to Z (f1).
LIST.

OUTPUT:

Wstring Xstring Ystring Zstring W X Y Z

000001  100     0100    10      6 1 2 1
100000  010     1000    01      1 2 1 2
000100  001     0010    10      4 3 3 1

Number of cases read:  3    Number of cases listed:  3

David Marso wrote

Here is how I would deal with 'many' of such abominations
(what a stupid way for a piece of SW to output a variable or do they consider
everything open game for mult response and not believe in parsimony?).
--
DATA LIST /W1 TO W6 1-6 X1 TO X3 8-10  Y1 TO Y4 12-15  Z1 TO Z2 17-18.
BEGIN DATA
000001 100 0100 10
100000 010 1000 01
000100 001 0010 10
END DATA.
** Note the potentially brain numbing double use of Var to represent
** both the VECTOR and the accumulated value ;-).
** Also might ask, where in the hell is the initialization?  
** We *don't* need any ;-)))).
VECTOR W=W1 TO W6 / X= X1 TO X3 /Y=Y1 TO Y4 /Z= Z1 TO Z2 .
DO REPEAT Var=W X Y Z /MAXCAT=6 3 4 2.
+  LOOP #=1 TO MAXCAT.
+    COMPUTE Var=SUM(Var,Var(#)*#).
+  END LOOP IF V > 0.
END REPEAT.
LIST W X Y Z.

Maguin, Eugene wrote

I'll bet there are a number of ways to do this. One example:
Compute interest=9.
Do repeat x=low medium high/y=1 2 3.
+  if (x eq 1) interest=y.
End repeat.
Missing values interest(9).

That works if you have a few items. But if you have many, ALL with the same response format.

Numeric x1 a2 a3 a4 t6  t7 ww y78(f1.0).
Vector new=x1 to y78.
Loop #i=1 to >number of new variables<.
Compute new(#i)=9.
Do repeat x=low medium high/y=1 2 3.
+  if (x eq 1) new(#i)=y.
End repeat.
End loop.
Missing values x1 to y78(9).

If you have varied response formats, .e., 1-3, 1-6, etc, I think you have no simple choice but to chunk the code by response format.

Gene Maguin


From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Brenda Zollitsch
Sent: Wednesday, April 11, 2012 1:21 PM
To: [hidden email]
Subject: Question about combining seperate answer columns in dataset into one variable

Hi there, SPSS folks,

I have just been provided a dataset from surveymonkey for a project.

The database breaks the answers down into different columns (new variables) for each answer option. Fake example.  What is your interest level? Answers: low, medium, high.
Instead of providing me with a variable (interest level) with 1=low, 2=medium and 3 = high, the data set has one column for low (with either 0 or 1) as the data, same for medium and same for high.

My question is how to combine these multiple columns into one variable with different answer options: variable =interest level with 1=low, 2=medium and 3=high.

I have done lots of recoding of data from one variable, but have never had to combine answers from multiple variables into one.

Can you point me in the right direction?

This is a quick project, and I am not great with syntax, so if there is anyway you can provide this via point and click tools within SPSS that would be my preference.

Thanks!

Brenda

CONTENTS DELETED

The author has deleted this message.