analytic suggestions

[Poling, Taylor Leigh]

Hello List,

What are your suggestions in terms of the best technique to use to address the
following question....

I am going to have a large sample of people (hopefully around 300-400 cases)
organized into several categorical groups based on their job category (e.g.,
technical staff, non-technical staff, R&D, administrative support, etc.) as
well as their occupational area (e.g., computer science, physical sciences,
human resources etc). So I will likely have around 10 - 15 group categories.
Then I will have all kinds of variables representing their network use to look
at (some possibilities are: logins, emails sent, amount of files and sizes
downloaded, temporary file storage, etc...basically, anything that might be of
interest that is collected at the network level).

What I want to look at is whether these categories of employees have
distinguishable electronic usage profiles. I want to get at whether there's a
normal baseline profile that is a reliable descriptor of the kind of
electronic behavior we would expect from someone knowing their job category
group.

I was thinking of tackling this with maybe a nearest neighbor discriminant
analysis? Or a decision tree?

Any suggestions, critiques, concerns, etc. would be greatly appreciated!

Thanks!
Taylor

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

[Lombardo, Barbara]

Hello,

Can someone please tell me what character to use in the SPSS .sps SYNTAX
file for COMMENT or documentation lines that will be ignored by the
processor?

Thank you

Barbara

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

[Spousta Jan]

If I recall it right, there are 3 posibilities:

* simple comment.

compute x = 0 /* inline comment */ .

COMMENT yet another comment.

Best regards,

Jan

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Lombardo, Barbara
Sent: Tuesday, October 30, 2007 2:36 PM
To: [hidden email]
Subject: CHARACTER FOR COMMENTS

Hello,

Can someone please tell me what character to use in the SPSS .sps SYNTAX file for COMMENT or documentation lines that will be ignored by the processor?

Thank you

Barbara

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



_____

Tato zpráva a všechny připojené soubory jsou důvěrné a určené výlučně adresátovi(-ům). Jestliže nejste oprávněným adresátem, je zakázáno jakékoliv zveřejňování, zprostředkování nebo jiné použití těchto informací. Jestliže jste tento mail dostali neoprávněně, prosím, uvědomte odesilatele a smažte zprávu i přiložené soubory. Odesilatel nezodpovídá za jakékoliv chyby nebo opomenutí způsobené tímto přenosem.

This message and any attached files are confidential and intended solely for the addressee(s). Any publication, transmission or other use of the information by a person or entity other than the intended addressee is prohibited. If you receive this in error please contact the sender and delete the message as well as all attached documents. The sender does not accept liability for any errors or omissions as a result of the transmission.

-.- --

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

[Shih, Dachuan]

Hi, Taylor,

I'll try tree-based model first. Before doing that, I'll look at the
frequeny table for each potential predictor variable.
It occurs to me the # of categories from your response variable is 10 -
15, with 300-400 cases. In that case, the # of minimum cases for each
terminal node has to be carefully determined.

Hope this helps.
Thomas


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Taylor Poling
Sent: Thursday, October 25, 2007 3:10 PM
To: [hidden email]
Subject: analytic suggestions

Hello List,

What are your suggestions in terms of the best technique to use to
address the following question....

I am going to have a large sample of people (hopefully around 300-400
cases) organized into several categorical groups based on their job
category (e.g., technical staff, non-technical staff, R&D,
administrative support, etc.) as well as their occupational area (e.g.,
computer science, physical sciences, human resources etc). So I will
likely have around 10 - 15 group categories.
Then I will have all kinds of variables representing their network use
to look at (some possibilities are: logins, emails sent, amount of files
and sizes downloaded, temporary file storage, etc...basically, anything
that might be of interest that is collected at the network level).

What I want to look at is whether these categories of employees have
distinguishable electronic usage profiles. I want to get at whether
there's a normal baseline profile that is a reliable descriptor of the
kind of electronic behavior we would expect from someone knowing their
job category group.

I was thinking of tackling this with maybe a nearest neighbor
discriminant analysis? Or a decision tree?

Any suggestions, critiques, concerns, etc. would be greatly appreciated!

Thanks!
Taylor

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