Xref: utzoo sci.math:4083 comp.sys.ibm.pc:16740
Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!mailrus!ames!pasteur!ucbvax!hplabs!nsc!pyramid!octopus!pete
From: pete@octopus.UUCP (Pete Holzmann)
Newsgroups: sci.math,comp.sys.ibm.pc
Subject: What is a good statistical analysis package for surveys?
Message-ID: <259@octopus.UUCP>
Date: 28 Jun 88 00:30:18 GMT
Reply-To: pete@octopus.UUCP (Pete Holzmann)
Organization: Octopus Enterprises, Cupertino CA
Lines: 29

Hi! Here's hoping the collective wisdom of the Net can help out again.

I'm doing a volunteer project that involves figuring out what the answers
to a survey mean. I'm sitting in front of a stack of 300 survey responses.
There are 40 questions, most with simple pick-one-of-several answers, a
few with pick-any-of-the-following-10-choices answers. A very few have
numeric responses (age, # people in household).

I'd like to find all statistically significant two-way correlations in this
data. I've written a simple dBase program that will make entering the data
easy.

THE QUESTION: Is there a statistical package out there that will make my
life easy in this project? At this point, I really don't care about fancy
graphs and other stuff like that. Correlations are necessary. A dBase
interface would be nice. Handling text (vs. numeric) data is pretty important.

Any opinions would be *much* appreciated, soon. I'll summarize whatever I
learn (from the Net or elsewhere) if there is any interest.

Thanks!

Pete

-- 
  OOO   __| ___      Peter Holzmann, Octopus Enterprises
 OOOOOOO___/ _______ USPS: 19611 La Mar Court, Cupertino, CA 95014
  OOOOO \___/        UUCP: {hpda,pyramid}!octopus!pete
___| \_____          Phone: 408/996-7746