Date: Wed, 14 Sep 2005 17:30:04 -0400
Reply-To: Richard Ristow <wrristow@mindspring.com>
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: Richard Ristow <wrristow@mindspring.com>
Subject: SPSS FAQs (was, re: cox regression with data weight)
In-Reply-To: <000f01c5b839$cfcb9700$5e64a8c0@russell>
Content-Type: text/plain; charset="us-ascii"; format=flowed
At 04:04 AM 9/13/2005, russell wrote:
>Your (consistent and detailed) responses to questions pertaining to
>weighting in SPSS etc are really useful and illuminating. In addition,
>there are a few threads on the very same issue on this discussion
>list. Will be nice if they can be bundled into one contribution...
Which brings up a thought.
Some questions come up repeatedly on this list, and get answered
repeatedly. I'm sure that some of us are particularly aware, as we
often answer the same or similar questions from time to time ourselves,
perhaps quoting our own old work. Raynald Levesque must have many such
examples, Marta probably a lot; and Hector, Arthur. Probably some of
mine, too.
It would be work, but useful, to prepare overall essays on some of
these topics, and make them available for reference.
Where to make them available? Well, if it's OK with Raynald,
www.spsstools.net is the only reasonable choice. There's already a FAQ
section there, but I don't think the topics duplicate what I'm
suggesting.
I don't generally like the term "FAQ" ("Frequently Asked Questions")
lists, partly because they're often really, "some things I thought of
and decided to write up, but didn't want to organize coherently." But
we can do better if
A. We select topics that do come up repeatedly. We can't be exhaustive,
but to start with, take any that any one of us has answered multiple
times. After that, anybody who notices another repeating topic can post
saying so, and we'll see if somebody's willing to write it up. (By the
way, if you do notice a repeating theme, please give subject line,
date, and poster for instances of it, so an author won't have to repeat
your search.)
B. The author of a topical essay (there should be one author) posts the
draft, or successive drafts, for comment. List members aren't ideal to
comment, being mostly too knowledgeable to catch some missing points,
but are a very useful start.
C. Organizing the essays for reference? Follow the topical outline
that Raynald's used for code examples. It's carefully considered, and
it works, and it spares inquirers from two topical classifications.
Experience will bring out any need for revision.
How to start? I'm not dropping everything, but I'm thinking of doing
some of my own recurring themes. If a few of us do the same, we'll have
a base to work with.
| 
