Date:         Mon, 13 Nov 2006 10:28:50 -0500
Reply-To:     "Miriam L. Gerver" <mgerver@gmail.com>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         "Miriam L. Gerver" <mgerver@gmail.com>
Subject:      Re: restructuring data to change unique identifier
Comments: To: Richard Ristow <wrristow@mindspring.com>
In-Reply-To:  <7.0.1.0.2.20061109164235.0470cdc8@mindspring.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Thank you, Richard. The syntax you sent worked perfectly! Yes, I
didn't realize that I could leave "department" unselected. And yes, I
selected the indicator variable option (for no rational reason).

Miriam

On 11/9/06, Richard Ristow <wrristow@mindspring.com> wrote:
> Postscript:
>
> At 11:06 AM 11/9/2006, Miriam L. Gerver wrote:
>
> >I ended up with an indicator variable for each department, rather than
> >a variable for the first department associated with a particular email
> >address and a second department associated with that particular email
> >address, etc. So it gave me about 30,000 indicator variables!
>
> I think I see how you did that. In the restructuring wizard, you
> . Panel 1: Selected "Restructure selected cases into variables",
> correctly
> . Panel 2:
> - Selected E-mail address as Identifier variable (correctly)
> + Selected Department as an index variable. THAT'S what's wrong, but
> it's an easy mistake to make: you know you want to do *something* with
> "Department", and those look like the only two choices. (In fact, you
> want to leave "Department" unselected from either category, at this
> point.)
> . Panel 3: (Sort the data or not, is OK)
> . Panel 4: Checked "Create indicator variables", which is the second
> problem.
>
> That generates the following:
>
> SORT CASES BY E_mail Dept .
> CASESTOVARS
>   /ID = E_mail
>   /INDEX = Dept
>   /GROUPBY = VARIABLE
>   /VIND ROOT = ind.
>
> And that gives you what you saw. This is SPSS draft output:
>
> LIST.
> |-----------------------------|---------------------------|
> |Output Created               |09-NOV-2006 16:55:44       |
> |-----------------------------|---------------------------|
> Dept E_mail
>
>    1  abc@abc
>    2  abc@abc
>    3  abc@def
>    4  abc@def
>    5  abc@ghi
>    6  abc@jkl
>
> Number of cases read:  6    Number of cases listed:  6
>
>
> SORT CASES BY E_mail Dept .
> CASESTOVARS
>   /ID = E_mail
>   /INDEX = Dept
>   /GROUPBY = VARIABLE
>   /VIND ROOT = ind.
>
>
> Cases to Variables
> |--------------------------|---------------------------|
> |Output Created            |09-NOV-2006 16:55:44       |
> |--------------------------|---------------------------|
>
> Generated Variables
> |---------|----|------|
> |Original |Dept|Result|
> |Variable |    |------|
> |         |    |Name  |
> |---------|----|------|
> |Indicator|1   |ind1  |
> |         |----|------|
> |         |2   |ind2  |
> |         |----|------|
> |         |3   |ind3  |
> |         |----|------|
> |         |4   |ind4  |
> |         |----|------|
> |         |5   |ind5  |
> |         |----|------|
> |         |6   |ind6  |
> |---------|----|------|
>
> Processing Statistics
> |---------------|---|
> |Cases In       |6  |
> |---------------|---|
> |Cases Out      |4  |
> |---------------|---|
> |Cases In/Cases |1.5|
> |Out            |   |
> |---------------|---|
> |Variables In   |2  |
> |---------------|---|
> |Variables Out  |7  |
> |---------------|---|
> |Index Values   |6  |
> |---------------|---|
>
>
> LIST.
> |-----------------------------|---------------------------|
> |Output Created               |09-NOV-2006 16:55:44       |
> |-----------------------------|---------------------------|
> E_mail  ind1 ind2 ind3 ind4 ind5 ind6
>
> abc@abc   1    1    0    0    0    0
> abc@def   0    0    1    1    0    0
> abc@ghi   0    0    0    0    1    0
> abc@jkl   0    0    0    0    0    1
>
> Number of cases read:  4    Number of cases listed:  4
>
>