Text Analysis for Surveys 2.0 - Triangulation of three categories

[Ed Sloat]

I am hoping someone on the SPSS thread is using or has used the SPSS
Text Analysis for Surveys software.
 
I am working a project comprised of about 1,000 extended comments. My
question is how I am able to review the intersection of three or more
concept categories I have created.
 
For example, I have a concept category called 'Reduction' (composed of
terms such as cut, reduce, eliminate, less, remove,...), a category
called 'Administration' (reflecting references to parts of the
organizational structure), and a third category called 'retired'
(referencing positions in the organization  that are filled with
once-retired employees who have been re-hired for specific projects and
periods of time).
 
My goal (for this small example) is to investigate the frequency of the
intersection of these three concept categories in the qualitative
information set. I know I can select the three categories, but I am not
sure what the bar graph and associated frequency statistics are telling
me - they don't seem to make sense in isolating these three areas within
the many other concept categories I have defined.
 
Also, anyone know of a good tutorial on the use of SPSS Text Analysis
for Survey's 2.0. The user guide is not very good in this respect.
 
Thanks
 
Ed Sloat
 
 
 
Edward F. Sloat, Director
Research, Planning, and Assessment
Peoria Unified School District
6330 West Thunderbird Road
Glendale, Arizona 85306
623-486-6077
[hidden email] <mailto:[hidden email]>
 

[Paul Thoresen]

Mr. Sloat,

Your best bet might be the web graphs.
The web graphs will show the relationship
between the categories, with more frequency
being denoted by the thickness of the web line.
It is a very handy visual representation of the data.

I think there is another feature that gives the
percentages (but I do not remember the name).
Others may be able to elaborate more.

Good luck.

Paul
Paul Thoresen, M.A.
Surveys & Research Manager
The American Academy of Neurology
1080 Montreal Ave., St. Paul, MN 55116
[hidden email] 651.695.2803



NOTICE:  This message may contain confidential or legally privileged information intended only for the use of the addressee(s) named above. Unauthorized use, disclosure, distribution, or copying is prohibited.  If you have received this message in error, please reply to the sender and delete the original message.  Thank you.
--------------------------------------------------------



From: Ed Sloat [mailto:[hidden email]]
Sent: Monday, April 02, 2007 11:08 AM
To: [hidden email]
Subject: Text Analysis for Surveys 2.0 - Triangulation of three
categories

I am hoping someone on the SPSS thread is using or has used the SPSS
Text Analysis for Surveys software.

I am working a project comprised of about 1,000 extended comments. My
question is how I am able to review the intersection of three or more
concept categories I have created.

For example, I have a concept category called 'Reduction' (composed of
terms such as cut, reduce, eliminate, less, remove,...), a category
called 'Administration' (reflecting references to parts of the
organizational structure), and a third category called 'retired'
(referencing positions in the organization  that are filled with
once-retired employees who have been re-hired for specific projects and
periods of time).

My goal (for this small example) is to investigate the frequency of the
intersection of these three concept categories in the qualitative
information set. I know I can select the three categories, but I am not
sure what the bar graph and associated frequency statistics are telling
me - they don't seem to make sense in isolating these three areas within
the many other concept categories I have defined.

Also, anyone know of a good tutorial on the use of SPSS Text Analysis
for Survey's 2.0. The user guide is not very good in this respect.

Thanks

Ed Sloat



Edward F. Sloat, Director
Research, Planning, and Assessment
Peoria Unified School District
6330 West Thunderbird Road
Glendale, Arizona 85306
623-486-6077
[hidden email] <mailto:[hidden email]>

[Norton, John]

Hi Ed,

I think your best bet here is to export the categories and use SPSS to analyze the intersection.

Essentially, your categories are binary variables; a respondent either did or did not say something in his or her text response which would qualify him or her for inclusion in a given category.  Because, presumably, a respondent can say something to qualify him or her into more than one category, you have the elements for constructing a multiple dichotomy set from these categories.  Using SPSS, you can define a multiple response set using multiple dichotomies (MD).

The trick here is to use SPSS to create 2 MD groups, and then use the crosstab engine in the Multiple Response option (or, for a more presentable table, you can do this within the Tables engine) to cross the MD group with itself.

This way, you can get raw counts as well as proportions for those who said something for one category, as well as something for another.

I hope this helps,

John Norton
SPSS Inc.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Ed Sloat
Sent: Monday, April 02, 2007 11:08 AM
To: [hidden email]
Subject: Text Analysis for Surveys 2.0 - Triangulation of three categories

I am hoping someone on the SPSS thread is using or has used the SPSS
Text Analysis for Surveys software.

I am working a project comprised of about 1,000 extended comments. My
question is how I am able to review the intersection of three or more
concept categories I have created.

For example, I have a concept category called 'Reduction' (composed of
terms such as cut, reduce, eliminate, less, remove,...), a category
called 'Administration' (reflecting references to parts of the
organizational structure), and a third category called 'retired'
(referencing positions in the organization  that are filled with
once-retired employees who have been re-hired for specific projects and
periods of time).

My goal (for this small example) is to investigate the frequency of the
intersection of these three concept categories in the qualitative
information set. I know I can select the three categories, but I am not
sure what the bar graph and associated frequency statistics are telling
me - they don't seem to make sense in isolating these three areas within
the many other concept categories I have defined.

Also, anyone know of a good tutorial on the use of SPSS Text Analysis
for Survey's 2.0. The user guide is not very good in this respect.

Thanks

Ed Sloat



Edward F. Sloat, Director
Research, Planning, and Assessment
Peoria Unified School District
6330 West Thunderbird Road
Glendale, Arizona 85306
623-486-6077
[hidden email] <mailto:[hidden email]>