Re: Statistical Analyses of Non-Static Group Question

On Aug 9, 7:41 pm, Doug Morse <mo...@xxxxxxxx> wrote:
Hi,

You've done a great job describing what's distinctive and "unusual" with
your design. To really help, though, I (or anyone really) will also need to
know more about what's "typical" about your design.

So, if you can say bit more about:

(a) who your participants are, how you are selecting them (or how they are
self-selecting), and how they are being assigned to various treatment
conditions (if there are any),
(b) what the participants are doing (i.e.,what's the task or tasks),
(c) what materials or stimuli are being used and, in particular, how the
materials differ due to any treatment conditions or manipulations, and
(d) how you are assessing and testing the participants and how often (i.e.,
what's the measure and its frequency)

that would be great. These four things -- participants, tasks, materials,
and measures -- are essential to understand in order to provide sound
advice.

No need to go overboard, though. I can ask follow-up questions for
clarification or elaboration, if needed. For now, if you can just provide
the basics of what's going on along these four dimensions, that's a great
start.

Cheers,
Doug

On Thu, 09 Aug 2007 15:52:21 -0700, gwcallah...@xxxxxxxxxxxx





<gwcallah...@xxxxxxxxxxxx> wrote:
I don't have a background in statistics so perhaps someone can lend
some guidance with respect to the issue that I present here or tell me
where I can go to find guidance.

The goal is to determine how well a group of participants for a given
program is performing over a given period of time. The number of
participants is not static. The program has a lifecycle where the
number of participants starts out small, builds to a maximum then
declines through attrition until the program is retired. Participants
tend to participate with greater frequency when first added to the
program then taper off with time. A very small percentage of
participants are extremely active from the time they are added until
they leave the program and to a point where they artificially raise
the norm.

The question then is how does one compare group performance from one
sample to the next if the group dynamic is changing?

How does one account for the influx of new participants and the effect
they have on the group?

How does one correct for the overly active participants?

Is it valid to look at only those participants who appear in
subsequent sample periods (i.e. exclude additions and removals)?

Statistics is a broad subject. What form of analysis does this problem
represent? What texts (books) should I study to gain a better
understanding of the problem?

doug morse | psychology vanderbilt | 615.322.5565 |http://edoug.org/dm- Hide quoted text -

- Show quoted text -

The participants of program provide product testing feedback in
various forms such as problem reports, status reports, surveys, and
forum discussions. Each from of feedback is weighted with respect to
its importance to the program. On a periodic basis (monthly) each
participant's feedback contribution is counted and weight adjusted to
produce a score. A mean is calculated based only on those participants
that contributed and a rating is awarded based on the participants
score relative to the mean. Non-contributors automatically receive the
lowest possible rating for the period.

The participants for the program are largely employees of the company
that produces the product; only a few outside participants are allowed
for security reasons. Many of the participants are intimate with
particular aspects of the product, however very few have complete
knowledge. The vast majority are simply users of the product with a
layman's understanding of how the product works.

The purpose for measuring participation is manifold and all closely
interrelated: 1) to ascertain that the product is actually being used
and tested, 2) to weed out those who are simply taking advantage of
the benefits of the program without complying with the requirements,
3) to manage resource allocation (placing the product in the hands of
those who will comply with the program requirements versus those who
won't), and 4) to provide some form of feedback to the participants in
hopes of encouraging those who are seriously under-performing to do a
better job.

While it's fairly simple to understand how well a particular
participant is performing on a periodic basis, it's not so easy to
determine how well the group is performing due to the dynamic makeup
of the group (as described in my previous post). I am hoping there's a
way to make statistical adjustments so that I can say with some amount
of certainty whether the performance of the group is better or worse
and by how much.

I hope provides better background on the problem.

.

Relevant Pages

  • Re: Simple Linear Supply (for teaching my son)
    ... although it could benefit from more modern design. ... zener for voltage stabilization. ... poor load regulation without direct feedback (you have indirect feedback ... voltage sense, and then you have a stable and useful supply. ...
    (sci.electronics.basics)
  • Re: Website for Comments and or Suggestions
    ... How many of your competitor's websites have background music? ... I would again encourage you to spend some time reading and studying some of the material at the links I gave you about web design. ... |> | website and ask for feedback, and I admire and respect that. ... |> | Look, I am not an expert web designer, and my opinions are just that. ...
    (microsoft.public.publisher.webdesign)
  • Re: OOP/OOD Philosophy
    ... Design is about applying recombinant ideas to solving problems. ... And the plans all had feedback. ... my point is that if something goes through the XP methodology ... You still need specific deliverables from the prototyping. ...
    (comp.object)
  • Re: OOP/OOD Philosophy
    ... And the plans all had feedback. ... missing design considerations. ... Once in production software is very hard to ... prototyping ...
    (comp.object)
  • Solving linear equations, the parallel way
    ... of distributed system, or a cluster. ... AX=B can be obtained by performing A^-1*B=X. ... this project is still in design fase and I'm afraid the current setup ... So my question to you is: What is the best (performance & bandwith) ...
    (sci.math.num-analysis)