Combining variables and calculating frequency

[Alexia Maddox]

Hello to All

I am new to this list and still a learner driver on SPSS.

I would like any advice on how to combine ten variables and do a basic frequency count

In this case, the respondent was able to provide up to ten web addresses  these have all been coded as ten seperate variables with the same 48 response options for each.

I would like to end up with a single variable that can provide the frequency count.

I realize that I will need to split the data file, and maybe transform each of the ten variables 48 times ( ending up and then do something like 480) variables, but I can't really figure out how to do all this.

I would appreciate any advice or suggested resources on how to go about this. I am also happy to provide further clarity where needed

Thank you for your time

Alexia Maddox
Phd candidate
Deakin University

=====================
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

[Kylie]

Hi Alexia,

It sounds like your variables can be defined as a multiple response set.
Look at Analyse > Multiple Response > Define Sets to create a set made up of
your 10 variables. Then, you can run the Multiple Response > Frequencies
procedure to get the frequency counts across all of the variables in the
set; there is no need to restructure the variables.

Hope this helps.

Cheers,
Kylie.


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Alexia Maddox
Sent: Monday, 21 March 2011 9:51 AM
To: [hidden email]
Subject: Combining variables and calculating frequency

Hello to All

I am new to this list and still a learner driver on SPSS.

I would like any advice on how to combine ten variables and do a basic
frequency count

In this case, the respondent was able to provide up to ten web addresses
these have all been coded as ten seperate variables with the same 48
response options for each.

I would like to end up with a single variable that can provide the frequency
count.

I realize that I will need to split the data file, and maybe transform each
of the ten variables 48 times ( ending up and then do something like 480)
variables, but I can't really figure out how to do all this.

I would appreciate any advice or suggested resources on how to go about
this. I am also happy to provide further clarity where needed

Thank you for your time

Alexia Maddox
Phd candidate
Deakin University

=====================
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

=====================
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]

Look up VARSTOCASES in the manual.  Or possibly MULTRESPONSE will suffice.

[Alexia Maddox]

Thank you Kylie. This does indeed seem to be a very good solution. I have noticed that this new variable can only be used to provide frequency data and cross tabs. It may be possible that I would want to use this new variable in a cluster analysis. Is this possible through this technique ?

Many thanks for your very helpful response

Alexia

On 21/03/2011, at 10:28 AM, "Kylie Lange" <[hidden email]> wrote:

> Hi Alexia,
>
> It sounds like your variables can be defined as a multiple response set.
> Look at Analyse > Multiple Response > Define Sets to create a set made up of
> your 10 variables. Then, you can run the Multiple Response > Frequencies
> procedure to get the frequency counts across all of the variables in the
> set; there is no need to restructure the variables.
>
> Hope this helps.
>
> Cheers,
> Kylie.
>
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
> Alexia Maddox
> Sent: Monday, 21 March 2011 9:51 AM
> To: [hidden email]
> Subject: Combining variables and calculating frequency
>
> Hello to All
>
> I am new to this list and still a learner driver on SPSS.
>
> I would like any advice on how to combine ten variables and do a basic
> frequency count
>
> In this case, the respondent was able to provide up to ten web addresses
> these have all been coded as ten seperate variables with the same 48
> response options for each.
>
> I would like to end up with a single variable that can provide the frequency
> count.
>
> I realize that I will need to split the data file, and maybe transform each
> of the ten variables 48 times ( ending up and then do something like 480)
> variables, but I can't really figure out how to do all this.
>
> I would appreciate any advice or suggested resources on how to go about
> this. I am also happy to provide further clarity where needed
>
> Thank you for your time
>
> Alexia Maddox
> Phd candidate
> Deakin University
>
> =====================
> 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
>

=====================
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

[John F Hall]

There's a set of tutorials on multiple response on my website.

See section 3.3  Multiple response  on
http://surveyresearch.weebly.com/block-3-analysing-two-variables-and-sometim
es-three.html


John Hall
[hidden email]
www.surveyresearch.weebly.com




-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Alexia Maddox
Sent: 21 March 2011 06:11
To: [hidden email]
Subject: Re: Combining variables and calculating frequency

Thank you Kylie. This does indeed seem to be a very good solution. I have
noticed that this new variable can only be used to provide frequency data
and cross tabs. It may be possible that I would want to use this new
variable in a cluster analysis. Is this possible through this technique ?

Many thanks for your very helpful response

Alexia

On 21/03/2011, at 10:28 AM, "Kylie Lange" <[hidden email]>
wrote:

> Hi Alexia,
>
> It sounds like your variables can be defined as a multiple response set.
> Look at Analyse > Multiple Response > Define Sets to create a set made up
of

> your 10 variables. Then, you can run the Multiple Response > Frequencies
> procedure to get the frequency counts across all of the variables in the
> set; there is no need to restructure the variables.
>
> Hope this helps.
>
> Cheers,
> Kylie.
>
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
> Alexia Maddox
> Sent: Monday, 21 March 2011 9:51 AM
> To: [hidden email]
> Subject: Combining variables and calculating frequency
>
> Hello to All
>
> I am new to this list and still a learner driver on SPSS.
>
> I would like any advice on how to combine ten variables and do a basic
> frequency count
>
> In this case, the respondent was able to provide up to ten web addresses
> these have all been coded as ten seperate variables with the same 48
> response options for each.
>
> I would like to end up with a single variable that can provide the

frequency
> count.
>
> I realize that I will need to split the data file, and maybe transform
each

> of the ten variables 48 times ( ending up and then do something like 480)
> variables, but I can't really figure out how to do all this.
>
> I would appreciate any advice or suggested resources on how to go about
> this. I am also happy to provide further clarity where needed
>
> Thank you for your time
>
> Alexia Maddox
> Phd candidate
> Deakin University
>
> =====================
> 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
>

=====================
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

=====================
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]

"It may be possible that I would want to use this new variable in a cluster analysis."
Please elaborate on this.  As you are aware a MR "variable" is really sort of a fictitious 'metadata' construct and is basically useless except for tabs and freqs.  
Quoting from your first post:
"> In this case, the respondent was able to provide up to ten web addresses
> these have all been coded as ten seperate variables with the same 48
> response options for each."

Are your data numerical codes or are they strings?
If numeric see VECTOR command and LOOP.
If string see AUTORECODE with GROUPED option.
Are these fields contiguous, are the numeric codes contiguous?
Assuming 10 numeric contiguous fields (X1 TO X10) coded from 1 to 48.

NUMERIC MDSITE01 TO MDSITE48.
RECODE MDSITE01 TO MDSITE48 (ELSE=0).
VECTOR MDSite=MDSITE01 TO MDSITE48.
VECTOR MRFields =X1 TO X10.
LOOP #=1 TO 10.
COMPUTE MDSITE(MRFIELDS(#))=1.
END LOOP.
----