Cleaner syntax for weighting responses

[parisec]

 

Hi all,

 

I've got a questionnaire that needs some weights applied to the scores before summing the totals.

 

There are 10 questions.

 

If a person responds with a 1, multiply the response x 1

If a person responds with a 2, multiply the response x 2

If a person responds with a 3, multiply the response x 3

i created 10 new variables:

 

numeric Tot1 to Tot10.

 

Now, I'm in the middle of a bunch of "if then" statments to do the above,  but i gotta think there is a cleaner and less error ridden way to accomplish this than the mess of syntax i'm creating right now.

 

I also suspect this is not the first time this has been asked.

 

If anyone can direct me to a previous thread that gives this syntax, i'd appreciate it. I searched through nabble but i don't think i'm using the correct search terms.

 

TIA

 

Carol

 

 

 

[Rick Oliver-3]

weight cases by varname.


Rick Oliver
Senior Information Developer
IBM Business Analytics (SPSS)
E-mail: [hidden email]



From:        "Parise, Carol A." <[hidden email]>
To:        [hidden email],
Date:        09/13/2013 01:03 PM
Subject:        Cleaner syntax for weighting responses
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

 
Hi all,
 
I've got a questionnaire that needs some weights applied to the scores before summing the totals.
 
There are 10 questions.
 
If a person responds with a 1, multiply the response x 1
If a person responds with a 2, multiply the response x 2
If a person responds with a 3, multiply the response x 3
i created 10 new variables:
 
numeric Tot1 to Tot10.
 
Now, I'm in the middle of a bunch of "if then" statments to do the above,  but i gotta think there is a cleaner and less error ridden way to accomplish this than the mess of syntax i'm creating right now.
 
I also suspect this is not the first time this has been asked.
 
If anyone can direct me to a previous thread that gives this syntax, i'd appreciate it. I searched through nabble but i don't think i'm using the correct search terms.
 
TIA
 
Carol
 
 
 

[Art Kendall]

can you provide a small set of data say with 2 variables and 6 cases that shows (where wants shows what value you want as results)
 id var1 var2 want1 want2
On 9/13/2013 2:04 PM, parisec [via SPSSX Discussion] wrote:

 

Hi all,

 

I've got a questionnaire that needs some weights applied to the scores before summing the totals.

 

There are 10 questions.

 

If a person responds with a 1, multiply the response x 1

If a person responds with a 2, multiply the response x 2

If a person responds with a 3, multiply the response x 3

i created 10 new variables:

 

numeric Tot1 to Tot10.

 

Now, I'm in the middle of a bunch of "if then" statments to do the above,  but i gotta think there is a cleaner and less error ridden way to accomplish this than the mess of syntax i'm creating right now.

 

I also suspect this is not the first time this has been asked.

 

If anyone can direct me to a previous thread that gives this syntax, i'd appreciate it. I searched through nabble but i don't think i'm using the correct search terms.

 

TIA

 

Carol

 

 

 

---

If you reply to this email, your message will be added to the discussion below:

http://spssx-discussion.1045642.n5.nabble.com/Cleaner-syntax-for-weighting-responses-tp5721977.html

To start a new topic under SPSSX Discussion, email [hidden email]
To unsubscribe from SPSSX Discussion, click here.
NAML

[parisec]

Sure...I have Q1 to Q4;  Tq1 to Tq4 is the goal.


ID      Q1      Q2      Q3      Q4      Tq1     Tq2     Tq3     Tq4
1       1       2       1       3       1       4       1       9
2       2       1       2       1       4       1       4       1
3       3       2       0       2       9       4       0       4
4       1       3       3       0       1       9       9       0


numeric Tq1 to Tq4.
        if (Q1=1 or Q2=1 or Q3=1 or Q4=1) ToQ1=Q1*1
        if (Q1=2 or Q2=2 or Q3=2 or Q4=2) ToQ1=Q1*2
        if (Q1=3 or Q2=3 or Q3=3 or Q4=3) ToQ1=Q1*3
....follows same pattern for each question.




-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Art Kendall
Sent: Friday, September 13, 2013 11:17 AM
To: [hidden email]
Subject: Re: Cleaner syntax for weighting responses

can you provide a small set of data say with 2 variables and 6 cases that shows (where wants shows what value you want as results)  id var1 var2 want1 want2

Art Kendall
Social Research Consultants
On 9/13/2013 2:04 PM, parisec [via SPSSX Discussion] wrote:




        Hi all,

        I've got a questionnaire that needs some weights applied to the scores before summing the totals.

        There are 10 questions.

        If a person responds with a 1, multiply the response x 1
        If a person responds with a 2, multiply the response x 2
        If a person responds with a 3, multiply the response x 3

        i created 10 new variables:

        numeric Tot1 to Tot10.

        Now, I'm in the middle of a bunch of "if then" statments to do the above,  but i gotta think there is a cleaner and less error ridden way to accomplish this than the mess of syntax i'm creating right now.

        I also suspect this is not the first time this has been asked.

        If anyone can direct me to a previous thread that gives this syntax, i'd appreciate it. I searched through nabble but i don't think i'm using the correct search terms.

        TIA

        Carol





________________________________

        If you reply to this email, your message will be added to the discussion below:
        http://spssx-discussion.1045642.n5.nabble.com/Cleaner-syntax-for-weighting-responses-tp5721977.html
        To start a new topic under SPSSX Discussion, email [hidden email]
        To unsubscribe from SPSSX Discussion, click here.
        NAML <http://spssx-discussion.1045642.n5.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml>


Art Kendall
Social Research Consultants

________________________________

View this message in context: Re: Cleaner syntax for weighting responses <http://spssx-discussion.1045642.n5.nabble.com/Cleaner-syntax-for-weighting-responses-tp5721977p5721979.html>
Sent from the SPSSX Discussion mailing list archive <http://spssx-discussion.1045642.n5.nabble.com/>  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]

DO REPEAT v=v1 TO v10 /t=t1 TO t10.
+  COMPUTE t=v**2.
+  COMPUTE total=SUM(total,t).
END REPEAT.

or if you don't require the t variables?

DO REPEAT v=v1 TO v10 .
+  COMPUTE total=SUM(total,v**2).
END REPEAT.

parisec wrote

Sure...I have Q1 to Q4;  Tq1 to Tq4 is the goal.


ID      Q1      Q2      Q3      Q4      Tq1     Tq2     Tq3     Tq4
1       1       2       1       3       1       4       1       9
2       2       1       2       1       4       1       4       1
3       3       2       0       2       9       4       0       4
4       1       3       3       0       1       9       9       0


numeric Tq1 to Tq4.
        if (Q1=1 or Q2=1 or Q3=1 or Q4=1) ToQ1=Q1*1
        if (Q1=2 or Q2=2 or Q3=2 or Q4=2) ToQ1=Q1*2
        if (Q1=3 or Q2=3 or Q3=3 or Q4=3) ToQ1=Q1*3
....follows same pattern for each question.




-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Art Kendall
Sent: Friday, September 13, 2013 11:17 AM
To: [hidden email]
Subject: Re: Cleaner syntax for weighting responses

can you provide a small set of data say with 2 variables and 6 cases that shows (where wants shows what value you want as results)  id var1 var2 want1 want2

Art Kendall
Social Research Consultants
On 9/13/2013 2:04 PM, parisec [via SPSSX Discussion] wrote:




        Hi all,

        I've got a questionnaire that needs some weights applied to the scores before summing the totals.

        There are 10 questions.

        If a person responds with a 1, multiply the response x 1
        If a person responds with a 2, multiply the response x 2
        If a person responds with a 3, multiply the response x 3

        i created 10 new variables:

        numeric Tot1 to Tot10.

        Now, I'm in the middle of a bunch of "if then" statments to do the above,  but i gotta think there is a cleaner and less error ridden way to accomplish this than the mess of syntax i'm creating right now.

        I also suspect this is not the first time this has been asked.

        If anyone can direct me to a previous thread that gives this syntax, i'd appreciate it. I searched through nabble but i don't think i'm using the correct search terms.

        TIA

        Carol





________________________________

        If you reply to this email, your message will be added to the discussion below:
        http://spssx-discussion.1045642.n5.nabble.com/Cleaner-syntax-for-weighting-responses-tp5721977.html
        To start a new topic under SPSSX Discussion, email [hidden email]
        To unsubscribe from SPSSX Discussion, click here.
        NAML <http://spssx-discussion.1045642.n5.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml>


Art Kendall
Social Research Consultants

________________________________

View this message in context: Re: Cleaner syntax for weighting responses <http://spssx-discussion.1045642.n5.nabble.com/Cleaner-syntax-for-weighting-responses-tp5721977p5721979.html>
Sent from the SPSSX Discussion mailing list archive <http://spssx-discussion.1045642.n5.nabble.com/>  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

[Richard Ristow]

At 02:28 PM 9/13/2013, Parise, Carol A. wrote:

>I have Q1 to Q4;  Tq1 to Tq4 is the goal.
>
>
>ID      Q1      Q2      Q3      Q4      Tq1     Tq2     Tq3     Tq4
>1       1      2       1       3       1       4       1       9
>2       2      1       2       1       4       1       4       1
>3       3      2       0       2       9       4       0       4
>4       1       3       3      0       1       9       9      0
>
>
>numeric Tq1 to Tq4.
>         if (Q1=1 or Q2=1 or Q3=1 or Q4=1) ToQ1=Q1*1
>         if (Q1=2 or Q2=2 or Q3=2 or Q4=2) ToQ1=Q1*2
>         if (Q1=3 or Q2=3 or Q3=3 or Q4=3) ToQ1=Q1*3
>....follows same pattern for each question.

This would give Tq1=6 rather than 4 in case 1. Which is correct?

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

[parisec]

ugh.

Tq1 should actually be 1 for Q1 case 1 .


The rules are:

Multiply any question with a response of 1 by 1
Multiply any question with a response of 2 by 2
Multiply any question with a response of 3 by 3

this is why i let spss do the math.



-----Original Message-----
From: Richard Ristow [mailto:[hidden email]]
Sent: Friday, September 13, 2013 11:57 AM
To: Parise, Carol A.; [hidden email]
Subject: Re: Cleaner syntax for weighting responses

At 02:28 PM 9/13/2013, Parise, Carol A. wrote:

>I have Q1 to Q4;  Tq1 to Tq4 is the goal.
>
>
>ID      Q1      Q2      Q3      Q4      Tq1     Tq2     Tq3     Tq4
>1       1      2       1       3       1       4       1       9
>2       2      1       2       1       4       1       4       1
>3       3      2       0       2       9       4       0       4
>4       1       3       3      0       1       9       9      0
>
>
>numeric Tq1 to Tq4.
>         if (Q1=1 or Q2=1 or Q3=1 or Q4=1) ToQ1=Q1*1
>         if (Q1=2 or Q2=2 or Q3=2 or Q4=2) ToQ1=Q1*2
>         if (Q1=3 or Q2=3 or Q3=3 or Q4=3) ToQ1=Q1*3 ....follows same
>pattern for each question.

This would give Tq1=6 rather than 4 in case 1. Which is correct?

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

[Richard Ristow]

Bother!

At 02:28 PM 9/13/2013, Parise, Carol A. wrote:

>I have Q1 to Q4;  Tq1 to Tq4 is the goal.
>
>
>ID      Q1      Q2      Q3      Q4      Tq1     Tq2     Tq3     Tq4
>1       1      2       1       3       1       4       1       9
>2       2      1       2       1       4       1       4       1
>3       3      2       0       2       9       4       0       4
>4       1       3       3      0       1       9       9      0
>
>
>numeric Tq1 to Tq4.
>         if (Q1=1 or Q2=1 or Q3=1 or Q4=1) ToQ1=Q1*1
>         if (Q1=2 or Q2=2 or Q3=2 or Q4=2) ToQ1=Q1*2
>         if (Q1=3 or Q2=3 or Q3=3 or Q4=3) ToQ1=Q1*3
>....follows same pattern for each question.

This would give Tq1=3 rather than 1 in case 1; and 3 rather than 1 in
case 4. Which is correct?

(I should have taken that nap.)

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

[Albert-Jan Roskam]

----- Original Message -----

> From: David Marso <[hidden email]>
> To: [hidden email]
> Cc:
> Sent: Friday, September 13, 2013 8:46 PM
> Subject: Re: [SPSSX-L] Cleaner syntax for weighting responses
>
> DO REPEAT v=v1 TO v10 /t=t1 TO t10.
> +  COMPUTE t=v**2.
> +  COMPUTE total=SUM(total,t).
> END REPEAT.

But where did "total" come from? Or did that variable exist already? How about this (untested):

vector #total = Tq1 to Tq4.
loop #i = 1 to 4.
+do repeat #question = q1 to q4 / #n = 1 to 4.
+  if ( any(#n, q1 to q4) ) #total(#i) = #question * #n.
+end repeat.
end loop.

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

[Art Kendall]

I sent this before but it did not  appear so I am sending it again.
It produces exactly the request results for the cases given.

if i understand correctly you just want to square each variable.
open a new instance of SPSS.
copy and paste the syntax below into a syntax window, run it.
data list list/
ID    Q1    Q2    Q3    Q4    Tq1    Tq2    Tq3    Tq4(9f1).
begin data
1    1    2    1    3    1    4    1    9
2    2    1    2    1    4    1    4    1
3    3    2    0    2    9    4    0    4
4    1    3    3    0    1    9    9    0
end data.
do repeat myvar= q1 to q4
 /result= r1 to r4.
compute result = myvar**2.
end repeat.
formats r1 to r4(f2).
list.

On 9/13/2013 2:04 PM, parisec [via SPSSX Discussion] wrote:

 

Hi all,

 

I've got a questionnaire that needs some weights applied to the scores before summing the totals.

 

There are 10 questions.

 

If a person responds with a 1, multiply the response x 1

If a person responds with a 2, multiply the response x 2

If a person responds with a 3, multiply the response x 3

i created 10 new variables:

 

numeric Tot1 to Tot10.

 

Now, I'm in the middle of a bunch of "if then" statments to do the above,  but i gotta think there is a cleaner and less error ridden way to accomplish this than the mess of syntax i'm creating right now.

 

I also suspect this is not the first time this has been asked.

 

If anyone can direct me to a previous thread that gives this syntax, i'd appreciate it. I searched through nabble but i don't think i'm using the correct search terms.

 

TIA

 

Carol

 

 

 

---

If you reply to this email, your message will be added to the discussion below:

http://spssx-discussion.1045642.n5.nabble.com/Cleaner-syntax-for-weighting-responses-tp5721977.html

To start a new topic under SPSSX Discussion, email [hidden email]
To unsubscribe from SPSSX Discussion, click here.
NAML

[parisec]

I've been off trying out everyones suggestions.

1) Yes. now that i look at it, doi! the variables just need to be squared.
2) The below worked as did David Marso's

I haven't tried Albert-Jan's yet but am going to give it a shot.

You have all given me solutions that not only work here, but that i think i can apply to some other projects i've got going.

Thanks a bunch for taking your time to help.

Carol



-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Art Kendall
Sent: Friday, September 13, 2013 12:50 PM
To: [hidden email]
Subject: Re: Cleaner syntax for weighting responses

I sent this before but it did not  appear so I am sending it again.
It produces exactly the request results for the cases given.

if i understand correctly you just want to square each variable.
open a new instance of SPSS.
copy and paste the syntax below into a syntax window, run it.
data list list/
ID    Q1    Q2    Q3    Q4    Tq1    Tq2    Tq3    Tq4(9f1).
begin data
1    1    2    1    3    1    4    1    9
2    2    1    2    1    4    1    4    1
3    3    2    0    2    9    4    0    4
4    1    3    3    0    1    9    9    0
end data.
do repeat myvar= q1 to q4
 /result= r1 to r4.
compute result = myvar**2.
end repeat.
formats r1 to r4(f2).
list.


Art Kendall
Social Research Consultants
On 9/13/2013 2:04 PM, parisec [via SPSSX Discussion] wrote:




        Hi all,

        I've got a questionnaire that needs some weights applied to the scores before summing the totals.

        There are 10 questions.

        If a person responds with a 1, multiply the response x 1
        If a person responds with a 2, multiply the response x 2
        If a person responds with a 3, multiply the response x 3

        i created 10 new variables:

        numeric Tot1 to Tot10.

        Now, I'm in the middle of a bunch of "if then" statments to do the above,  but i gotta think there is a cleaner and less error ridden way to accomplish this than the mess of syntax i'm creating right now.

        I also suspect this is not the first time this has been asked.

        If anyone can direct me to a previous thread that gives this syntax, i'd appreciate it. I searched through nabble but i don't think i'm using the correct search terms.

        TIA

        Carol





________________________________

        If you reply to this email, your message will be added to the discussion below:
        http://spssx-discussion.1045642.n5.nabble.com/Cleaner-syntax-for-weighting-responses-tp5721977.html
        To start a new topic under SPSSX Discussion, email [hidden email]
        To unsubscribe from SPSSX Discussion, click here.
        NAML <http://spssx-discussion.1045642.n5.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml>


Art Kendall
Social Research Consultants

________________________________

View this message in context: Re: Cleaner syntax for weighting responses <http://spssx-discussion.1045642.n5.nabble.com/Cleaner-syntax-for-weighting-responses-tp5721977p5721988.html>
Sent from the SPSSX Discussion mailing list archive <http://spssx-discussion.1045642.n5.nabble.com/>  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]

Albert-Jan:
total does *NOT* exist prior to:
COMPUTE total = SUM(total,t).
It doesn't need to.  
Neat trick IMO to bypass unnecessary initialization code ;-)

--

Albert-Jan Roskam wrote

----- Original Message -----

> From: David Marso <[hidden email]>
> To: [hidden email]
> Cc:
> Sent: Friday, September 13, 2013 8:46 PM
> Subject: Re: [SPSSX-L] Cleaner syntax for weighting responses
>
> DO REPEAT v=v1 TO v10 /t=t1 TO t10.
> +  COMPUTE t=v**2.
> +  COMPUTE total=SUM(total,t).
> END REPEAT.

But where did "total" come from? Or did that variable exist already? How about this (untested):

vector #total = Tq1 to Tq4.
loop #i = 1 to 4.
+do repeat #question = q1 to q4 / #n = 1 to 4.
+  if ( any(#n, q1 to q4) ) #total(#i) = #question * #n.
+end repeat.
end loop.

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

[Jon K Peck]

A great way to write unreadable, unmaintainable code.


Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
phone: [phone]




From:        David Marso <[hidden email]>
To:        [hidden email],
Date:        09/14/2013 09:33 AM
Subject:        Re: [SPSSX-L] Cleaner syntax for weighting responses
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

Albert-Jan:
total does *NOT* exist prior to:
COMPUTE total = SUM(total,t).
It doesn't need to.
Neat trick IMO to bypass unnecessary initialization code ;

[Albert-Jan Roskam]

----- Original Message -----

> From: David Marso <[hidden email]>
> To: [hidden email]
> Cc:
> Sent: Saturday, September 14, 2013 5:32 PM
> Subject: Re: [SPSSX-L] Cleaner syntax for weighting responses
>
> Albert-Jan:
> total does *NOT* exist prior to:
> COMPUTE total = SUM(total,t).
> It doesn't need to.
> Neat trick IMO to bypass unnecessary initialization code ;-)

Hi David,

ehh? So you can take the sum of a variable (and something else) before it even springs into life? *)

Btw, my code was wrong although it did seem to be a generalization of what the OP mentioned, but maybe I shiohould have read the entire thread.

*) confusing: I am now going to drink a beer without actually opening the bottle **)
**) on the second thought I *will* open it! --cheers!

Albert-Jan

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

Both readable and maintainable Jon!
Scratching an itch?
--

Jon K Peck wrote

A great way to write unreadable, unmaintainable code.


Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
phone: [phone]




From:   David Marso <[hidden email]>
To:     [hidden email],
Date:   09/14/2013 09:33 AM
Subject:        Re: [SPSSX-L] Cleaner syntax for weighting responses
Sent by:        "SPSSX(r) Discussion" <[hidden email]>



Albert-Jan:
total does *NOT* exist prior to:
COMPUTE total = SUM(total,t).
It doesn't need to.
Neat trick IMO to bypass unnecessary initialization code ;