|Date: ||Sun, 18 Apr 1999 16:26:07 -0400|
|Reply-To: ||"Daniel B. Swartz" <danswartz@PIEINC.COM>|
|Sender: ||"SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>|
|From: ||"Daniel B. Swartz" <danswartz@PIEINC.COM>|
|Subject: ||New to SPSS - Bear with me|
|Content-Type: ||text/plain; charset="iso-8859-1"|
I'm new to SPSS (using 8.0 for Windows), so bear with me.
I have a question or two here:
I cannot seem to figure out how to tell SPSS to "figuratively" collapse my
data from subscales into one
score (mean) in order to do significance tests for different dependent
variables. I received no book with my software, just a couple pieces of
paper, so I have no documentation from SPSS to help me along the way. I
have read "Guide to Data Analysis" by Norusis - no help with this, or at
least I can't find it in there...
Here is the deal:
I have administered a satisfaction survey with 56 statements/questions, all
with Likert scale responses.
I have 177 cases.
This survey has a total of 12 subscales, with an average of 4 questions in
In order to compare dependent variable differences I need to be able to tell
the program to collapse the means (or give me a mean for the statements just
in that subscale) for each case.
How do I tell SPSS to do this? I have tried doing a bivariate correlation,
Spearman Rho, and can't for the life of me figure out how to tell it examine
the subscales of job satisfaction and compare it with another subscale, say
workload. I am sure this is a simple process, but way beyond me at this
Any help would be appreciated.
Daniel B. Swartz, MA, CSA, CI, CT
Professional Interpreter Exchange, Inc.
Deaf and Interpreter Owned
9101 Cherry Lane, Suites 104-105
Laurel, MD 20708-1133