Calculating percents for a dimension

[Laura Leach]

Hello all,

I have question about calculating dimension percents for a group of
questions, across cases with missing values.

Below is a sample dataset for Dimension 1, comprised for 4 questions.  1=
Unfavorable, 2 = Neutral, and 3 = Favorable.  Is there a custom table that
can summarize these 4 variables into a single value, ignoring the missing
values?


q1 q2 q3 q4
. . 1 1
1 . . .
3 3 2 3
3 3 3 1
2 2 2 1
3 1 3 3
3 3 3 3
3 2 1 3
1 3 3 1
3 3 3 3


The way we typically calculate these percents is through Excel formulas.
We past this data into excel, we then sum the frequency (count in SPSS) all
the 1s 2s and 3s and divide each by the total.  In this example, the
percents would be as follows:

           Unfav    Neu       Fav      Total
counts     9      5        21 35
percents    0.257    0.14      0.6

Is there any way to do this in SPSS?  Thanks so much in advance if you have
a solution to this time-consuming problem!!

Laura Leach

[Justin Black]

Hi Laura,

Try this:

*** Aggregates cases to calculate percentages for each response option .

COMPUTE alldata = 1 .
AGGREGATE
  /OUTFILE='c:\agg.sav'
  /BREAK=alldata
  /q1_u= PIN(q1 1 1)
  /q2_u= PIN(q2 1 1)
  /q3_u= PIN(q3 1 1)
  /q4_u=  PIN(q4 1 1)
  /q1_n= PIN(q1 2 2)
  /q2_n= PIN(q2 2 2)
  /q3_n= PIN(q3 2 2)
  /q4_n=  PIN(q4 2 2)
  /q1_f= PIN(q1 3 3)
  /q2_f= PIN(q2 3 3)
  /q3_f= PIN(q3 3 3)
  /q4_f=  PIN(q4 3 3).

*** Computes the dimension scores .

GET
  FILE='C:\agg.sav'.

COMPUTE DIM1_FAV = MEAN(q1_f to q4_f) .
COMPUTE DIM1_NEU = MEAN(q1_n to q4_n) .
COMPUTE DIM1_UNF = MEAN(q1_u to q4_u) .
EXE .

Hope that's what you were looking for.

Best,

--Justin Black


On 8/22/06, Laura Leach <[hidden email]> wrote:

>
> Hello all,
>
> I have question about calculating dimension percents for a group of
> questions, across cases with missing values.
>
> Below is a sample dataset for Dimension 1, comprised for 4 questions.  1=
> Unfavorable, 2 = Neutral, and 3 = Favorable.  Is there a custom table that
> can summarize these 4 variables into a single value, ignoring the missing
> values?
>
>
> q1 q2 q3 q4
> . . 1 1
> 1 . . .
> 3 3 2 3
> 3 3 3 1
> 2 2 2 1
> 3 1 3 3
> 3 3 3 3
> 3 2 1 3
> 1 3 3 1
> 3 3 3 3
>
>
> The way we typically calculate these percents is through Excel formulas.
> We past this data into excel, we then sum the frequency (count in SPSS)
> all
> the 1s 2s and 3s and divide each by the total.  In this example, the
> percents would be as follows:
>
>            Unfav    Neu       Fav      Total
> counts     9      5        21 35
> percents    0.257    0.14      0.6
>
> Is there any way to do this in SPSS?  Thanks so much in advance if you
> have
> a solution to this time-consuming problem!!
>
> Laura Leach
>

[Edward Boadi]

This should work for you.

GET FILE = 'Type file name here' .

Value Labels
/q1 to q4
1  "Unfavorable"
2  "Neutral"
3  "Favorable".

VARSTOCASES  /MAKE Dimension FROM Q1 TO Q4 .

FREQ  Dimension.


********Out put ***********
Dimension
                         Frequency      Percent Valid Percent   Cumulative Percent
Valid   Unfavorable              9             25.7             25.7              25.7
        Neutral          5             14.3             14.3              40.0
        Favorable                21            60.0             60.0              100.0
        Total                  35              100.0      100.0


Edward.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]]On Behalf Of
Laura Leach
Sent: Tuesday, August 22, 2006 1:49 PM
To: [hidden email]
Subject: Calculating percents for a dimension


Hello all,

I have question about calculating dimension percents for a group of
questions, across cases with missing values.

Below is a sample dataset for Dimension 1, comprised for 4 questions.  1=
Unfavorable, 2 = Neutral, and 3 = Favorable.  Is there a custom table that
can summarize these 4 variables into a single value, ignoring the missing
values?


q1 q2 q3 q4
. . 1 1
1 . . .
3 3 2 3
3 3 3 1
2 2 2 1
3 1 3 3
3 3 3 3
3 2 1 3
1 3 3 1
3 3 3 3


The way we typically calculate these percents is through Excel formulas.
We past this data into excel, we then sum the frequency (count in SPSS) all
the 1s 2s and 3s and divide each by the total.  In this example, the
percents would be as follows:

           Unfav    Neu       Fav      Total
counts     9      5        21 35
percents    0.257    0.14      0.6

Is there any way to do this in SPSS?  Thanks so much in advance if you have
a solution to this time-consuming problem!!

Laura Leach

[Beadle, ViAnn]

If you think of these 4 variables as a multiple response set, then you can tabulate counts and percents (using responses as the base) in the old MULT RESPONSE procedure, the old TABLES procedure, or Custom Tables. People often overlook what the reporting procedures can do and opt for file transformation approaches.

________________________________

From: SPSSX(r) Discussion on behalf of Edward Boadi
Sent: Tue 8/22/2006 3:20 PM
To: [hidden email]
Subject: Re: Calculating percents for a dimension



This should work for you.

GET FILE = 'Type file name here' .

Value Labels
/q1 to q4
1  "Unfavorable"
2  "Neutral"
3  "Favorable".

VARSTOCASES  /MAKE Dimension FROM Q1 TO Q4 .

FREQ  Dimension.


********Out put ***********
Dimension
                         Frequency      Percent Valid Percent   Cumulative Percent
Valid   Unfavorable              9             25.7             25.7              25.7
        Neutral          5             14.3             14.3              40.0
        Favorable                21            60.0             60.0              100.0
        Total                  35              100.0      100.0


Edward.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]]On Behalf Of
Laura Leach
Sent: Tuesday, August 22, 2006 1:49 PM
To: [hidden email]
Subject: Calculating percents for a dimension


Hello all,

I have question about calculating dimension percents for a group of
questions, across cases with missing values.

Below is a sample dataset for Dimension 1, comprised for 4 questions.  1=
Unfavorable, 2 = Neutral, and 3 = Favorable.  Is there a custom table that
can summarize these 4 variables into a single value, ignoring the missing
values?


q1 q2 q3 q4
. . 1 1
1 . . .
3 3 2 3
3 3 3 1
2 2 2 1
3 1 3 3
3 3 3 3
3 2 1 3
1 3 3 1
3 3 3 3


The way we typically calculate these percents is through Excel formulas.
We past this data into excel, we then sum the frequency (count in SPSS) all
the 1s 2s and 3s and divide each by the total.  In this example, the
percents would be as follows:

           Unfav    Neu       Fav      Total
counts     9      5        21 35
percents    0.257    0.14      0.6

Is there any way to do this in SPSS?  Thanks so much in advance if you have
a solution to this time-consuming problem!!

Laura Leach

[Laura Leach]

WOW!  Thank you SOO much!  I tried it on the real dataset and it worked to
the 10th decimal!!!!!  I wish I had asked months ago, before I did
hundreds of reports by hand to bypass SPSS!  Thank you thank you THANK
YOU!!!!!

Laura

On Tue, 22 Aug 2006 16:20:53 -0400, Edward Boadi <[hidden email]> wrote:

>This should work for you.
>
>GET FILE = 'Type file name here' .
>
>Value Labels
>/q1 to q4
>1  "Unfavorable"
>2  "Neutral"
>3  "Favorable".
>
>VARSTOCASES  /MAKE Dimension FROM Q1 TO Q4 .
>
>FREQ  Dimension.
>
>
>********Out put ***********
>Dimension
>                         Frequency      Percent Valid Percent

Cumulative Percent
>Valid   Unfavorable              9             25.7
25.7              25.7
>        Neutral          5             14.3             14.3
40.0
>        Favorable                21            60.0
60.0              100.0
>        Total                  35              100.0      100.0
>
>
>Edward.
>