By george, I think; signs and portents

[From Bill Powers (940724.0845MDT)]

Bill Leach (940723.1712) --

I think you've got it -- that is, as well as I do.

I also am of the opinion that it is only when discussing IT that Martin
seems to maintain this position. In many of the other discussions,
Martin has been as eloquent as You, Mary, Rick, Tom, Dag or whoever in
his expression of the correct view in this matter (again in my
opinion).

I agree completely. Martin is one of our really effective people when it
comes to teaching PCT. We get snarled up only when the discussion starts
bringing in abtract concepts that can't be called "basic." At least not
by me.

···

-----------------------------------
RE: What it means to sit on your legs.

It occurred to me on reading your post that psychology treats behavior
in terms of "signs and portents" rather than understanding organization,
and does so for a good reason.

Suppose you were hiring computer programmers, and wanted to use some
kind of objective test to be fair in your criteria. If you were an
engineer-manager, you would think up some programming problems, like
"Write a program in C to sort this file of numbers by the second digit
in a hexadecimal ASCII representation of the numerical values and output
the result to a line printer." You could then look over the code and see
if the task was really accomplished, and if so, how efficiently.

But what if you were a psychologist being asked to whip up a test to
screen applicants for the same job? Psychologists seldom receive any
kind of training in programming, certainly not enough to devise a
programming problem and evaluate the solution. The psychologist can't
really evaluate how well the applicant will perform as a programmer, so
what is done instead is to look for signs and portents relating to other
factors.

The psychologist will have to look for indirect indications that the
applicant has qualities generally deemed desireable in _any_ employee.
But there is no way to measure these qualities directly. Instead, what
is done is to ask questions that are answered in particular ways by
other people deemed to do well on jobs. Would you rather go to a ball
game or sit by yourself and read a good book? Do you dislike people who
are always telling you what to do? Would you rather have blue or yellow
wallpaper in your bedroom?

Behind such questions there's obviously some rationale, but formally
speaking all that matters is how other people who have done well in jobs
have answered the same questions. There isn't really any technical basis
for the questions themselves. But what else can one do, when one has no
training concerning the job that will actually be filled? If you're
trained as a psychologist, what do you know about mining or marketing or
law or electronics or any of the hundreds of other things that people
learn to do and get jobs doing? And if you have no real theory of
behavioral organization, what else can you ask about but preferences for
wallpaper colors, or whatever else seems to be a sign or a portent of
future behavior?
----------------------------------------------------------------------
Best,

Bill P.

<[Bill Leach 940724.13:59 EST(EDT)]

[ Bill Powers (940724.0845MDT)]

It occurred to me on reading your post that psychology treats behavior
in terms of "signs and portents" rather than understanding organization,
and does so for a good reason.

I am lead to believe that the "MMPI" (Minosota Multiphasic Personality
Inventory) is actually a "decent test" Of course I have that "belief"
without much justification. I know that the test is (was?) used almost
universally in the nuclear power industry but will admit to only casual
evidence that it really is useful (by the way, for what ever it is worth,
in that industry questionable results from the MMPI are the bases for an
interview with a psycharitrist. In addition some significant fraction or
people are interviewed using a "random" selection process).

I do not know how such an exam "stacks up" when taken from the men,
women, blacks, hispanics, asians should all "score" the same, on the
average, view.

Most importantly though, I do not know how well the exam actually
performs its' function.

Suppose you were hiring computer programmers, and wanted to use some
kind of objective test to be fair in your criteria.

Yes, I agree and even then it is still a rather subjective process but at
least if the interviewer is really trying to be "objective" and
recognizes the nature of many of the subjective factors then the process
can work "halfway decently".

It also makes sense that "industry" would want to have a means of
determining in advance if they are potentially hiring a person with a
strong reference for killing others or a reference for something like
"whatever you can get away with is OK".

So what can PCT say to the personel manager? What direct, ready to
implement, practical improvements does PCT offer this person? I am not
being sarcastic here and am quite serious. This manager has "data" that
shows that what he is doing now improves his chances of success in the
process marginally but enough to make a significant difference "to the
bottom line".

It is all well and good to tell him how bad his current system is and he
might even agree with you but unless you can offer him something workable
then all he will do is listen (if that).

-bill

Tom Bourbon [940725.1159]

<[Bill Leach 940724.13:59 EST(EDT)]

[ Bill Powers (940724.0845MDT)]

It occurred to me on reading your post that psychology treats behavior
in terms of "signs and portents" rather than understanding organization,
and does so for a good reason.

I am lead to believe that the "MMPI" (Minosota Multiphasic Personality
Inventory) is actually a "decent test" Of course I have that "belief"
without much justification. I know that the test is (was?) used almost
universally in the nuclear power industry but will admit to only casual
evidence that it really is useful (by the way, for what ever it is worth,
in that industry questionable results from the MMPI are the bases for an
interview with a psycharitrist. In addition some significant fraction or
people are interviewed using a "random" selection process).

Bill L, have you had enough time to think of _one_ good reason why you
believe the MMPI is a decent test? If so, let us know. :wink:

Actually, I believe assessment psychologists prefer to call their tests
"instruments," not tests -- the preferred label must sound more
scientifically legitimate and rigorous.
. . .

Most importantly though, I do not know how well the exam actually
performs its' function.

It depends on whose perspective you adopt when you try to answer the
question. The MMPI is part of a multi-million-dollar-a-year industry and
it is used to justify decisions that have profound effects on people. It
seems to perform its functions _very well_ for those on one side of the
tester-tested divide.

. . .

It also makes sense that "industry" would want to have a means of
determining in advance if they are potentially hiring a person with a
strong reference for killing others or a reference for something like
"whatever you can get away with is OK".

It _does_ make sense. It's too bad behavioral scientists and those who
apply the results of their "knowledge" don't just tell industry that they
(the behavioral experts) don't know how to identify those people with any
level of precision -- sometmes not at all.

So what can PCT say to the personel manager? What direct, ready to
implement, practical improvements does PCT offer this person? I am not
being sarcastic here and am quite serious. This manager has "data" that
shows that what he is doing now improves his chances of success in the
process marginally but enough to make a significant difference "to the
bottom line".

I would tell that person (a) if you are satisfied with the results of
conventional psychological assessments, you are satisfied, and (b) I'm happy
that I don't have to make important decisions about people's lives, using
evidence as bad as that from psychological assessments.

I would also (inelegantly) tell people who care about these issues _and_ who
know about PCT _and_ who want to make some big bucks (that excludes me --
I've taken a PCT vow of poverty), to get to work on developing better
alternatives. I won't do it. My vow precludes it, and I'm too busy getting
in over my head by trying to model much simpler things.

I am a big believer in the idea that, every time we come to a topic about
which we know nothing, or about which our knowledge is _dangerously_
inadequate, we should cry out, "Don't just do something; stand there!"

It is all well and good to tell him how bad his current system is and he
might even agree with you but unless you can offer him something workable
then all he will do is listen (if that).

It is both "all well and good" and the _only_ proper thing (for me) to do.
I cannot be responsible for the folly of what he or she elects to do next.

I'm still out on the limb.
Tom Bourbon
Department of Neurosurgry
University of Texas Medical School-Houston Phone: 713-792-5760
6431 Fannin, Suite 7.138 Fax: 713-794-5084
Houston, TX 77030 USA tbourbon@heart.med.uth.tmc.edu

<[Bill Leach 940725.21:55 EST(EDT)]

Tom Bourbon [940725.1159]

Bill L, have you had enough time to think of _one_ good reason why you
believe the MMPI is a decent test? If so, let us know. :wink:

No, but I might keep trying to think of one... so standby (though you
might not want to hold your breath or delay a meal or anything like that).

Actually, I believe assessment psychologists prefer to call their tests
"instruments," not tests -- the preferred label must sound more
scientifically legitimate and rigorous. . . .

Yes, similar technique used by the "sham" environmentalists. "Air a
great (bipolar of course) controversy" and resoundingly "prove" one view
as aburd and the other as obviously "morally", "ethically" and
"scientifically" superiour while completely ignoring any views that
actually do challenge the validity of your claim.

I detect strong moral and ethical standards in your posting on this
matter (see how "perceptive" I am?). Of course your own moral value
system is no more justifiable using PCT than anyone elses.

I could not help but notice however, that your standard for intellectual
honesty controlled such that you had to maintain some seperation between
what PCT actually tells us "raw" and what it tells us based upon your own
standards (even if the two were almost expressed in the same phrase).

Opinion of course, but that is pretty good for someone that "obviously"
feels as strongly about this subject as you appear to feel.

I think I like your "Don't just do something! Stand there!" :slight_smile:

Of course that is not how a control system error should be handled either
but if conscious effort is involved it might at least be nice to admit
that the whole attempt is based upon SWAG.

OTOH, this "justification" for major, sweeping laws and changes because
some "theorists" MIGHT be right about some physical phenomenon is an
excellent example of when your expression REALLY should be applied.

-bill

From Tom Bourbon [940726.1051]

<[Bill Leach 940725.21:55 EST(EDT)]

Tom Bourbon [940725.1159]

. . .

Concerning my post on uses and abuses of the MMPI.

Bill:

I detect strong moral and ethical standards in your posting on this
matter (see how "perceptive" I am?). Of course your own moral value
system is no more justifiable using PCT than anyone elses.

My own moral system is _every bit_ as justifiable, and unjustifiable, as
that of any other person. My system-level and principle-level references
are as real as anyone else's. You did indeed detect the on-screen results
of my actions to correct my own sense of error when I perceive what I
believe are abuses of large numbers of people in the name of allegedly
precise psychological assessments. Perceptual control in action.

I could not help but notice however, that your standard for intellectual
honesty controlled such that you had to maintain some seperation between
what PCT actually tells us "raw" and what it tells us based upon your own
standards (even if the two were almost expressed in the same phrase).

For someone who sometimes plays the role of "uninformed new-comer," you are
pretty good at picking up on higher level references. :slight_smile:

We cannot use PCT to predict which higher-level references a person will
defend, or how those specific references will come to be, but we can use it
to explain and predict the kinds of things that will happen when a person's
higher-level perceptions are disturbed. Also, as I said in my eaelier posts
on psychological assessments, the statistical adequacy or inadequacy of a
particular assessment "instrument" or the uses to which people put that
instrument are not directly PCT issues; they are matters of statistical fact
-- 80% to 95.4% error is 80% to 95.4% error -- period.

It is of interest to many PCT scientists that so many other behavioral
scientists seem to be satisfied with such huge errors of prediction, and
many of us working with PCT feel outrage that scientists would condone the
use of such inadequate data as a justification for the abuse of people in
large numbers. But we can't make other see the folly of their ways; a few
sincere people within psychology have tried to do that, for decades. They
are tolerated as the court jesters and are pointed to whenever traditional
psychologists want to say they are already aware of the problems with
assessments: "Oh, of course. Old so-and-so has been writing about that very
fact for many years."

I see our best option to be what we are already doing: working to build an
alternative science of behavior, grounded on a working model, not on words
and unworkable lineal models that lead to inadequate predictions about
behavior.

I think I like your "Don't just do something! Stand there!" :slight_smile:

Of course that is not how a control system error should be handled either
but

I'm not sure what you mean here. Why shouldn't error be handled that way?
Seriously.

···

===========

Subject: Re: Defining; testing; defining

<[Bill Leach 940725.22:35 EST(EDT)]

Tom Bourbon [940725.1240]

More on the MMPI.
. . .

Bill L:

Lets see how "clever" you are... Do you have any idea who first began
using this "intrument" for employment screening?

Give up?

The U.S. Government of course. It was used on candidates for people that
would work in close proximity to the President.

I was aware of that. Even when there have been attempts to abolish some of
the abuses of the MMPI as a screen for employment, the Feds have exempted
themselves -- same as they do with the abuse of "lie detector exams" in the
workplace.

Later,

Tom

<[Bill Leach 940727.18:31 EST(EDT)]

Tom Bourbon [940726.1051]

My own moral system is _every bit_ as justifiable, and unjustifiable, as
that of any other person.

I am only responding to this, lest anyone else may think that I thought
otherwise (since I am quite sure that Tom know that I don't challenge his
rights in this regard).

We cannot use PCT to predict which higher-level references a person will
defend, or how those specific references will come to be, but we can use
it to explain and predict the kinds of things that will happen when a
person's higher-level perceptions are disturbed.

This is something that has definately taken me a bit of time to
understand properly. Particularly to understand that the "We cannot use
PCT to..." part is true but that even there PCT will be less in error
than conventional psychology. I am coming around to understand that one
of the "problems" that PCT is facing is directly caused by then thing
that makes PCT so fundamentally different:

PCT is almost "its own worst enemy" in a fashion quite similar to the
nuclear industry (at least the nuclear scientists). You demand
intellectual honesty of yourselves and refuse to assert things that you
can not at least defend through sound logical argument.

Thus you will talk frankly and as honestly as possible about what PCT can
not do, what the existing and (as understood currently) the theoritical
limits are to what PCT will ever be able to accomplish. You try to
remain realistic as to what PCT can do and even what it might do in the
future. You are willing to do a "sales job" but only to the extent that
you can either prove (prefered) or at least logically argue from a very
strong position.

This is both your curse and your virtue!

I think I like your "Don't just do something! Stand there!" :slight_smile: >
Of course that is not how a control system error should be handled either
but

I'm not sure what you mean here. Why shouldn't error be handled that
way? Seriously.

I am serious, but I should not have said "should." A control system that
has an error, will do something. The "something" can include things that
do not involve "external" actions and often probably should.

-bill