Date: Mon, 7 Apr 1997 09:52:32 -0400
Reply-To: VBMARTIN@AOL.COM
Sender: "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>
From: Vivian Martin <VBMARTIN@AOL.COM>
Subject: best test for nonprobability samples?
Hello,
Well, I plunked down the $185 for the 7.5 Windows grad student pack. I'm in
awe of its capabilities, much as I am in awe of the many power users who
participate in this discussion group. I see that this SPSS pack is a great
bargain, even as I'm beginning to fear it may be more power than I need or
can handle right now.
I've spent the last week or so reading up on nonparametric tests and the 7.0
manuals I purchased when I took a stat course and was introduced to SPSSat
another school. I've put 164 cases , each with 41 variables, in my new
at-home system, but am not a little confused about whether my nonprobability
purposive sample really meets he standards for nonparametric tests as I'd
assumed.
I'd never expected to have to get as deep into stats etc as I have. I'm doing
a grounded study attempting to blend multimethods around the issue of layoffs
and individuals investment in core American values. As a journalist by day (
student by night), my training has been to use numbers and live voices, so I
wanted some of the strengths of quantitative analysis and qualitative
research in my study. I developed a survey with questions from GSS, another
national survey, and a couple of open-ended questions that helped point the
way to subjects for in-depth interviews ( I'm doing 20-25 on top of the
survey data). One segment of my survey group came from job search
centers/workshops in the Greater Hartford area, the other 90+ are mainly mid
managers from the same Hartford insurance company who'd put together a
support group/networking club.( I did a mailing and got more than 60 percent
response.) So I was in deeper than I'd planned, but one of the aspects of the
inquiry-based grad program I'm in ( Vermont College/Norwich University is
that you have to pick up all the interdisciplinary strands, no matter where
they lead.
For me, all roads led to SPSS, but it appears, from manual reading, that
there is an assumption of randomness for all the tests, even ones that some
textbooks would indicate can be used for purposive nonprobability samples
with nominal and rank orders, which is where I fit in. Am I misreading the
manual, or are there manuals/Internet articles that might clarify this for
me? I am definitely not a power user, but would hate to think I need to go
out and buy some sort of Stats for Poets package. Thanks in advance, and
sorry for my dumbness.
Vivian Martin
| 
