On Wed, 15 Aug 2001, Melady Preece wrote in part [edited]:
> ... about a project I am about [to] begin. The goal is to use a
> variety of individual predictors (IQ, previous work experience,
> education, personality) to develop a model to predict "success" after
> a vocational rehabilitation program for psychiatric patients.
>
> The problem is how to define success. The current data provides 5
> possible outcomes: Full-time employment, part-time employment,
> ongoing education, volunteer work, or no change.
It is not clear whether "no change" always, or only sometimes, excludes
any of the other categories. It is therefore not clear how "success" may
reasonably be defined (which is in part your question). In fact, you may
wish to define "success" in more than one way, for different analyses.
If "success" be defined solely in terms of outcome _status_, then "no
change" is not a proper value; rather, you should specify the status quo
ante. Perhaps the status quo ante is the same for all your Ss;
nonetheless, to avoid certain kinds of semantic confusion, that status
should be the value reported, for purposes of comparison with the other
outcomes.
If "success" be defined in terms of change of status, then it needs to be
clear that the other 4 categories are in fact changes (and therefore may
usefully be compared with "no change"), and what they are changes from;
and if the status quo ante is in fact not the same for all Ss, it may not
be easy to define "change of status" in terms that are comparable across
Ss.
> Clearly there is no argument to be made that these are linear, but
> even ordinal is questionable.
> I had thought of using a number of logit regression analyses for the
> various outcomes. Or, to use linear regression, rescaling as number of
> hours per week employed, and combining the two employment outcomes;
> scaling education training in terms of program length; combining
> volunteer and no change into number of hours involved in productive
> activity. That would give [me] three outcomes.
All of these are possible. Since you appear to be engaged in essentially
exploratory inquiry, it might be reasonable to pursue all paths, to see
which one(s) seem more productive.
You could rescale all the outcomes into (possibly estimated) numbers of
hours per week of "productive activity". If the status quo ante includes
some productive activity, it might perhaps be appropriate to rescale to
change in the number of hours per week. (I'd always be inclined to
pursue any avenue that occurred to mind, not knowing in advance which
would be most profitable.)
------------------------------------------------------------------------
Donald F. Burrill [EMAIL PROTECTED]
184 Nashua Road, Bedford, NH 03110 603-471-7128
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
