SPSSX Discussion

grouping variables

Classic

List

Threaded

7 messages Options

Nico Peruzzi

grouping variables

Hi listers,

I have the following data structure from a survey:

v1 v2 v3 v4 v5 s1 s2 s3 s4 s5
1 3
4 2
3 5 1 4
1 2 3 4 5 4 5 3 4 4

The "v" variables stand for "vendor". The respondent could identify from 0
to 5 vendors, which are numbered 1-5.

The "s" variables are satisfaction. If the respondent identified a vendor,
the rate the vendor. So, in case 1, only vendor #1 was identified and was
rated 3. In case 3, the respondent identified vendors 3 and 5, and rated
them a 1 and a 4, respectively. Case 4 named all teh vendors and gave them
all ratings.

I would like create a simple percentage table, and get the mean score for
each of vendors 1-5. I'm v13 with Tables module.

Any ideas would be appreciated.

Thanks, Nico

--
Nico Peruzzi, Ph.D.

Edward Boadi

Re: grouping variables

Nico, I guess the scores are categorical , if the answer is yes , why
do you want to compute the mean score for each vendor ?

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]]On Behalf Of
Nico Peruzzi
Sent: Wednesday, August 09, 2006 3:16 PM
To: [hidden email]
Subject: grouping variables

Hi listers,

I have the following data structure from a survey:

v1 v2 v3 v4 v5 s1 s2 s3 s4 s5
1 3
4 2
3 5 1 4
1 2 3 4 5 4 5 3 4 4

The "v" variables stand for "vendor". The respondent could identify from 0
to 5 vendors, which are numbered 1-5.

The "s" variables are satisfaction. If the respondent identified a vendor,
the rate the vendor. So, in case 1, only vendor #1 was identified and was
rated 3. In case 3, the respondent identified vendors 3 and 5, and rated
them a 1 and a 4, respectively. Case 4 named all teh vendors and gave them
all ratings.

I would like create a simple percentage table, and get the mean score for
each of vendors 1-5. I'm v13 with Tables module.

Any ideas would be appreciated.

Thanks, Nico

--
Nico Peruzzi, Ph.D.

Nico Peruzzi

Re: grouping variables

The s variables are Likert scales.

Thanks, Nico

On 8/9/06, Edward Boadi <[hidden email]> wrote:

>
> Nico, I guess the scores are categorical , if the answer is yes , why
> do you want to compute the mean score for each vendor ?
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]]On Behalf Of
> Nico Peruzzi
> Sent: Wednesday, August 09, 2006 3:16 PM
> To: [hidden email]
> Subject: grouping variables
>
>
> Hi listers,
>
> I have the following data structure from a survey:
>
> v1 v2 v3 v4 v5 s1 s2 s3 s4 s5
> 1 3
> 4 2
> 3 5 1 4
> 1 2 3 4 5 4 5 3 4 4
>
> The "v" variables stand for "vendor". The respondent could identify from
> 0
> to 5 vendors, which are numbered 1-5.
>
> The "s" variables are satisfaction. If the respondent identified a
> vendor,
> the rate the vendor. So, in case 1, only vendor #1 was identified and was
> rated 3. In case 3, the respondent identified vendors 3 and 5, and rated
> them a 1 and a 4, respectively. Case 4 named all teh vendors and gave
> them
> all ratings.
>
> I would like create a simple percentage table, and get the mean score for
> each of vendors 1-5. I'm v13 with Tables module.
>
> Any ideas would be appreciated.
>
> Thanks, Nico
>
> --
> Nico Peruzzi, Ph.D.
>

--
Nico Peruzzi, Ph.D.

Edward Boadi

Re: grouping variables

In reply to this post by Nico Peruzzi

Hope this helps :

DATA LIST FREE /v1 v2 v3 v4 v5 s1 s2 s3 s4 s5 .<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />

BEGIN DATA

1 0 0 0 0 3 0 0 0 0

4 0 0 0 0 2 0 0 0 0

3 5 0 0 0 1 4 0 0 0

1 2 3 4 5 4 5 3 4 4

END DATA.

RECODE v1 to v5 (0=sysmis).

RECODE s1 to s5 (0=sysmis).

exe.

* Custom Tables.

CTABLES

/VLABELS VARIABLES=s1 s2 s3 s4 s5 v1 v2 v3 v4 v5

DISPLAY=DEFAULT

/TABLE v1 [C] + v2 [C] + v3 [C] + v4 [C] + v5 [C] BY s1 [VALIDN,COLPCT.VALIDN,MEAN] + s2 [VALIDN,COLPCT.VALIDN,MEAN]

+ s3 [VALIDN,COLPCT.VALIDN,MEAN] + s4[VALIDN,COLPCT.VALIDN,MEAN] + s5 [VALIDN,COLPCT.VALIDN,MEAN]

/CATEGORIES VARIABLES=v1 v2 v3 v4 v5 ORDER=A KEY=VALUE EMPTY=EXCLUDE.

Regards.

Edward

[Edward Boadi]
-----Original Message-----
From: Nico Peruzzi [mailto:[hidden email]]
Sent: Wednesday, August 09, 2006 3:36 PM
To: Edward Boadi
Cc: [hidden email]
Subject: Re: grouping variables

The s variables are Likert scales.

Thanks, Nico

On 8/9/06, Edward Boadi < [hidden email]> wrote:

Nico, I guess the scores are categorical , if the answer is yes , why
do you want to compute the mean score for each vendor ?

-----Original Message-----
From: SPSSX(r) Discussion [mailto: [hidden email]]On Behalf Of
Nico Peruzzi
Sent: Wednesday, August 09, 2006 3:16 PM
To: [hidden email]
Subject: grouping variables

Hi listers,

I have the following data structure from a survey:

v1 v2 v3 v4 v5 s1 s2 s3 s4 s5
1 3
4 2
3 5 1 4
1 2 3 4 5 4 5 3 4 4

The "v" variables stand for "vendor". The respondent could identify from 0
to 5 vendors, which are numbered 1-5.

The "s" variables are satisfaction. If the respondent identified a vendor,
the rate the vendor. So, in case 1, only vendor #1 was identified and was
rated 3. In case 3, the respondent identified vendors 3 and 5, and rated
them a 1 and a 4, respectively. Case 4 named all teh vendors and gave them
all ratings.

I would like create a simple percentage table, and get the mean score for
each of vendors 1-5. I'm v13 with Tables module.

Any ideas would be appreciated.

Thanks, Nico

--
Nico Peruzzi, Ph.D.

--
Nico Peruzzi, Ph.D.

Nico Peruzzi

Re: grouping variables

Hmm, tried this. I think it's close - perhaps adding a total column to the
end will do the trick. As is, I don't get a single summary rating for each
vendor. But maybe not - I really need to combine all instances of each
vendor.

Thoughts?

Thanks, Nico

On 8/9/06, Edward Boadi <[hidden email]> wrote:

>
> Hope this helps :
>
>
>
>
>
> DATA LIST FREE /v1 v2 v3 v4 v5 s1 s2 s3 s4
> s5 .
>
> BEGIN DATA
>
> 1 0 0 0 0 3 0 0 0 0
>
> 4 0 0 0 0 2 0 0 0 0
>
> 3 5 0 0 0 1 4 0 0 0
>
> 1 2 3 4 5 4 5 3 4 4
>
> END DATA.
>
>
>
> RECODE v1 to v5 (0=sysmis).
>
> RECODE s1 to s5 (0=sysmis).
>
> exe.
>
> * Custom Tables.
>
> CTABLES
>
> /VLABELS VARIABLES=s1 s2 s3 s4 s5 v1 v2 v3 v4 v5
>
> DISPLAY=DEFAULT
>
> /TABLE v1 [C] + v2 [C] + v3 [C] + v4 [C] + v5 [C] BY s1 [VALIDN,
> COLPCT.VALIDN,MEAN] + s2 [VALIDN,COLPCT.VALIDN,MEAN]
>
> + s3 [VALIDN,COLPCT.VALIDN,MEAN] + s4[VALIDN,COLPCT.VALIDN,MEAN] + s5
> [VALIDN,COLPCT.VALIDN,MEAN]
>
> /CATEGORIES VARIABLES=v1 v2 v3 v4 v5 ORDER=A KEY=VALUE EMPTY=EXCLUDE.
>
>
>
> Regards.
>
> Edward
>
>
> [Edward Boadi]
> -----Original Message-----
> *From:* Nico Peruzzi [mailto:[hidden email]]
> *Sent:* Wednesday, August 09, 2006 3:36 PM
> *To:* Edward Boadi
> *Cc:* [hidden email]
> *Subject:* Re: grouping variables
>
> The s variables are Likert scales.
>
> Thanks, Nico
>
> On 8/9/06, Edward Boadi <[hidden email]> wrote:
> >
> > Nico, I guess the scores are categorical , if the answer is yes , why
> > do you want to compute the mean score for each vendor ?
> >
> > -----Original Message-----
> > From: SPSSX(r) Discussion [mailto:[hidden email]]On Behalf Of
> > Nico Peruzzi
> > Sent: Wednesday, August 09, 2006 3:16 PM
> > To: [hidden email]
> > Subject: grouping variables
> >
> >
> > Hi listers,
> >
> > I have the following data structure from a survey:
> >
> > v1 v2 v3 v4 v5 s1 s2 s3 s4 s5
> > 1 3
> > 4 2
> > 3 5 1 4
> > 1 2 3 4 5 4 5 3 4 4
> >
> > The "v" variables stand for "vendor". The respondent could identify
> > from 0
> > to 5 vendors, which are numbered 1-5.
> >
> > The "s" variables are satisfaction. If the respondent identified a
> > vendor,
> > the rate the vendor. So, in case 1, only vendor #1 was identified and
> > was
> > rated 3. In case 3, the respondent identified vendors 3 and 5, and
> > rated
> > them a 1 and a 4, respectively. Case 4 named all teh vendors and gave
> > them
> > all ratings.
> >
> > I would like create a simple percentage table, and get the mean score
> > for
> > each of vendors 1-5. I'm v13 with Tables module.
> >
> > Any ideas would be appreciated.
> >
> > Thanks, Nico
> >
> > --
> > Nico Peruzzi, Ph.D.
> >
>
>
>
> --
> Nico Peruzzi, Ph.D.
>

--
Nico Peruzzi, Ph.D.

Edward Boadi

Re: grouping variables

In reply to this post by Nico Peruzzi

What about this :

DATA LIST FREE /v1 v2 v3 v4 v5 s1 s2 s3 s4 s5 .<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />

BEGIN DATA

1 0 0 0 0 3 0 0 0 0

4 0 0 0 0 2 0 0 0 0

3 5 0 0 0 1 4 0 0 0

1 2 3 4 5 4 5 3 4 4

END DATA.

recode v1 to v5 (0 = sysmis).

Compute st = (s1+ s2 + s3+ s4 + s5 ).

exe.

MRSETS

/MCGROUP NAME=$Vendors LABEL='Vendors' VARIABLES=v1 v2 v3 v4 v5

/DISPLAY NAME=[$Vendors].

* Custom Tables.

CTABLES

/VLABELS VARIABLES=st $Vendors DISPLAY=DEFAULT

/TABLE $Vendors BY st [VALIDN F40.0, SUM, MEAN]

/CATEGORIES VARIABLES=$Vendors ORDER=A KEY=VALUE EMPTY=EXCLUDE TOTAL=NO POSITION=AFTER.

Regards.

-----Original Message-----
From: Nico Peruzzi [mailto:[hidden email]]
Sent: Wednesday, August 09, 2006 4:24 PM
To: Edward Boadi
Cc: [hidden email]
Subject: Re: grouping variables

Hmm, tried this. I think it's close - perhaps adding a total column to the end will do the trick. As is, I don't get a single summary rating for each vendor. But maybe not - I really need to combine all instances of each vendor.

Thoughts?

Thanks, Nico

On 8/9/06, Edward Boadi < [hidden email]> wrote:

Hope this helps :

DATA LIST FREE /v1 v2 v3 v4 v5 s1 s2 s3 s4 s5 .

BEGIN DATA

1 0 0 0 0 3 0 0 0 0

4 0 0 0 0 2 0 0 0 0

3 5 0 0 0 1 4 0 0 0

1 2 3 4 5 4 5 3 4 4

END DATA.

RECODE v1 to v5 (0=sysmis).

RECODE s1 to s5 (0=sysmis).

exe.

* Custom Tables.

CTABLES

/VLABELS VARIABLES=s1 s2 s3 s4 s5 v1 v2 v3 v4 v5

DISPLAY=DEFAULT

/TABLE v1 [C] + v2 [C] + v3 [C] + v4 [C] + v5 [C] BY s1 [VALIDN,COLPCT.VALIDN,MEAN] + s2 [VALIDN,COLPCT.VALIDN,MEAN]

+ s3 [VALIDN,COLPCT.VALIDN,MEAN] + s4[VALIDN,COLPCT.VALIDN,MEAN] + s5 [VALIDN,COLPCT.VALIDN,MEAN]

/CATEGORIES VARIABLES=v1 v2 v3 v4 v5 ORDER=A KEY=VALUE EMPTY=EXCLUDE.

Regards.

Edward

[Edward Boadi]

-----Original Message-----
From: Nico Peruzzi [mailto: [hidden email]]
Sent: Wednesday, August 09, 2006 3:36 PM
To: Edward Boadi
Cc: [hidden email]
Subject: Re: grouping variables

The s variables are Likert scales.

Thanks, Nico

On 8/9/06, Edward Boadi < [hidden email]> wrote:

Nico, I guess the scores are categorical , if the answer is yes , why
do you want to compute the mean score for each vendor ?

-----Original Message-----
From: SPSSX(r) Discussion [mailto: [hidden email]]On Behalf Of
Nico Peruzzi
Sent: Wednesday, August 09, 2006 3:16 PM
To: [hidden email]
Subject: grouping variables

Hi listers,

I have the following data structure from a survey:

v1 v2 v3 v4 v5 s1 s2 s3 s4 s5
1 3
4 2
3 5 1 4
1 2 3 4 5 4 5 3 4 4

The "v" variables stand for "vendor". The respondent could identify from 0
to 5 vendors, which are numbered 1-5.

The "s" variables are satisfaction. If the respondent identified a vendor,
the rate the vendor. So, in case 1, only vendor #1 was identified and was
rated 3. In case 3, the respondent identified vendors 3 and 5, and rated
them a 1 and a 4, respectively. Case 4 named all teh vendors and gave them
all ratings.

I would like create a simple percentage table, and get the mean score for
each of vendors 1-5. I'm v13 with Tables module.

Any ideas would be appreciated.

Thanks, Nico

--
Nico Peruzzi, Ph.D.

--
Nico Peruzzi, Ph.D.

--
Nico Peruzzi, Ph.D.

Nico Peruzzi

Re: grouping variables

I think this will get me there. Thanks very much.

Nico

On 8/9/06, Edward Boadi <[hidden email]> wrote:

>
> What about this :
>
>
>
> DATA LIST FREE /v1 v2 v3 v4 v5 s1 s2 s3 s4
> s5 .
>
> BEGIN DATA
>
> 1 0 0 0 0 3 0 0 0 0
>
> 4 0 0 0 0 2 0 0 0 0
>
> 3 5 0 0 0 1 4 0 0 0
>
> 1 2 3 4 5 4 5 3 4 4
>
> END DATA.
>
> recode v1 to v5 (0 = sysmis).
>
> Compute st = (s1+ s2 + s3+ s4 + s5 ).
>
> exe.
>
>
>
> MRSETS
>
> /MCGROUP NAME=$Vendors LABEL='Vendors' VARIABLES=v1 v2 v3 v4 v5
>
> /DISPLAY NAME=[$Vendors].
>
> * Custom Tables.
>
> CTABLES
>
> /VLABELS VARIABLES=st $Vendors DISPLAY=DEFAULT
>
> /TABLE $Vendors BY st [VALIDN F40.0, SUM, MEAN]
>
> /CATEGORIES VARIABLES=$Vendors ORDER=A KEY=VALUE EMPTY=EXCLUDE TOTAL=NO
> POSITION=AFTER.
>
>
>
> Regards.
>
> -----Original Message-----
> *From:* Nico Peruzzi [mailto:[hidden email]]
> *Sent:* Wednesday, August 09, 2006 4:24 PM
> *To:* Edward Boadi
> *Cc:* [hidden email]
> *Subject:* Re: grouping variables
>
> Hmm, tried this. I think it's close - perhaps adding a total column to
> the end will do the trick. As is, I don't get a single summary rating for
> each vendor. But maybe not - I really need to combine all instances of each
> vendor.
>
> Thoughts?
>
> Thanks, Nico
>
> On 8/9/06, Edward Boadi <[hidden email]> wrote:
> >
> > Hope this helps :
> >
> >
> >
> >
> >
> > DATA LIST FREE /v1 v2 v3 v4 v5 s1 s2 s3
> > s4 s5 .
> >
> > BEGIN DATA
> >
> > 1 0 0 0 0 3 0 0 0 0
> >
> > 4 0 0 0 0 2 0 0 0 0
> >
> > 3 5 0 0 0 1 4 0 0 0
> >
> > 1 2 3 4 5 4 5 3 4 4
> >
> > END DATA.
> >
> >
> >
> > RECODE v1 to v5 (0=sysmis).
> >
> > RECODE s1 to s5 (0=sysmis).
> >
> > exe.
> >
> > * Custom Tables.
> >
> > CTABLES
> >
> > /VLABELS VARIABLES=s1 s2 s3 s4 s5 v1 v2 v3 v4 v5
> >
> > DISPLAY=DEFAULT
> >
> > /TABLE v1 [C] + v2 [C] + v3 [C] + v4 [C] + v5 [C] BY s1 [VALIDN,
> > COLPCT.VALIDN,MEAN] + s2 [VALIDN,COLPCT.VALIDN,MEAN]
> >
> > + s3 [VALIDN,COLPCT.VALIDN,MEAN] + s4[VALIDN,COLPCT.VALIDN,MEAN] + s5
> > [VALIDN,COLPCT.VALIDN,MEAN]
> >
> > /CATEGORIES VARIABLES=v1 v2 v3 v4 v5 ORDER=A KEY=VALUE EMPTY=EXCLUDE.
> >
> >
> >
> > Regards.
> >
> > Edward
> >
> >
> > [Edward Boadi]
> > -----Original Message-----
> > *From:* Nico Peruzzi [mailto:[hidden email]]
> > *Sent:* Wednesday, August 09, 2006 3:36 PM
> > *To:* Edward Boadi
> > *Cc:* [hidden email]
> > *Subject:* Re: grouping variables
> >
> > The s variables are Likert scales.
> >
> > Thanks, Nico
> >
> > On 8/9/06, Edward Boadi <[hidden email]> wrote:
> > >
> > > Nico, I guess the scores are categorical , if the answer is yes , why
> > > do you want to compute the mean score for each vendor ?
> > >
> > > -----Original Message-----
> > > From: SPSSX(r) Discussion [mailto:[hidden email]]On Behalf
> > > Of
> > > Nico Peruzzi
> > > Sent: Wednesday, August 09, 2006 3:16 PM
> > > To: [hidden email]
> > > Subject: grouping variables
> > >
> > >
> > > Hi listers,
> > >
> > > I have the following data structure from a survey:
> > >
> > > v1 v2 v3 v4 v5 s1 s2 s3 s4 s5
> > > 1 3
> > > 4 2
> > > 3 5 1 4
> > > 1 2 3 4 5 4 5 3 4 4
> > >
> > > The "v" variables stand for "vendor". The respondent could identify
> > > from 0
> > > to 5 vendors, which are numbered 1-5.
> > >
> > > The "s" variables are satisfaction. If the respondent identified a
> > > vendor,
> > > the rate the vendor. So, in case 1, only vendor #1 was identified and
> > > was
> > > rated 3. In case 3, the respondent identified vendors 3 and 5, and
> > > rated
> > > them a 1 and a 4, respectively. Case 4 named all teh vendors and gave
> > > them
> > > all ratings.
> > >
> > > I would like create a simple percentage table, and get the mean score
> > > for
> > > each of vendors 1-5. I'm v13 with Tables module.
> > >
> > > Any ideas would be appreciated.
> > >
> > > Thanks, Nico
> > >
> > > --
> > > Nico Peruzzi, Ph.D.
> > >
> >
> >
> >
> > --
> > Nico Peruzzi, Ph.D.
> >
>
>
>
> --
> Nico Peruzzi, Ph.D.
>

--
Nico Peruzzi, Ph.D.