Incentives

[From Rick Marken (960123.1830)]

Bruce Abbott (960123.1815 EST) --

Or in words you would rather not hear used, by reducing intrinsic error it
[the incentive] leads to the cessation of reorganization, which leaves the
current candidate reference in place or "established." Hmmm, so the
occurrence of the incentive _does_ lead to the establishment of a reference
for striking the key, in your view. For a moment there I thought you were
disagreeing with me. (;->

The incentive does not _lead to_ the "establishment of the reference; the
incentive is a variable in a loop ; as Bill Powers (960123.1345 MST)
pointed out, it is a _dependent variable_. "The incentive by itself can
take no action, nor can it cause ["lead to"] any particular behavior. The
reference specifications for the perception of intrinsic variables are
what drive ("lead to") reorganization or a control hierarchy; not external
variables like food and water.

We've been over this ground [regarding reorganization] before, I know, so
there probably isn't much point in rehashing the issue until we get
honest-to-goodness real data to argue from

Ah. Would that you believed this about reinforcement theory as well!

Best

Rick

[Fred Nickols (971128.1615 ET)]

Rick Marken (971127.1510), in the midst of a long string of messages about
incentives, remarks, as an aside:

NB Fred Nichols: Do you still think that it's control theorists
who use language idiosyncratically? :wink:

I HATE messages like that! The first thing it did was drive me to my
dictionary of foreign terms to confirm what I thought N.B. meant--and I just
today completed the move of my library from one room to another. Drat!

Next, I have the family honor to protect. I am constrained to point out the
correct spelling of my name: it's Nickols not Nichols.

Then, despite my intentions of staying out of the darn discussion about
incentives, I now find myself drawn into it.

For what it's worth, being a dyed-in-the-wool, sometimes radical
behaviorist, I'm of the opinion that incentives work some of the time with
some people for some things and never all the time with everyone for
everything. In my book, they're a hit-and-miss proposition, for all the
reasons cited to date.

I think PCT offers a very good explanation for the spotty effectiveness of
incentives, namely, that if an individual is aware of the "deal," as Bill
Powers put it, and that deal lines up with some goal state not met, most
individuals will do what is necessary to obtain the incentive, provided the
costs don't outweigh the benefits (and y'all can interpret that last
statement just fine without help from me).

From a practical perspective, I don't know that the discussion has shed any

light on the utility or advisability of incentives. Most people I know
realize that incentives aren't worth a damn if no one wants them. They also
know the risks of dangling bunches of really plump, juicy carrots in front
of a group of very hungry mules. Some few know enough to make sure they're
dealing with hungry mules before they try carrots as an incentive. Fewest
in number are those who realize the risks of placing other organisms in
deprived states so as to increase the likely effectiveness of items that
have been labeled "reinforcers."

As a general practice, I nod agreeably when clients talk about the
importance of structuring incentives to support and encourage the behaviors
desired, but I don't "do" incentive systems as part of my consulting
practice. I think they're a waste of time. More important, to impose
contingent relationships between rewards/incentives/carrots and
behavior/actions/results is to try to exercise external control. If I know
anything about external control systems, I know they have to be simple in
nature and few in number (and I owe that insight to Peter Drucker, by the
way). If external control systems are not simple in nature and few in
number, people will manipulate the hell out of them. If anyone has ever
seen an untripped mousetrap with the cheese missing, or tripped with the
cheese missing and no mouse, you know what I mean. It simply is not true,
as so many bone-headed pontificating fools pronounce it so, that "You get
what you measure." That is pure unadulterated bullshit! When you measure
something, and tie others' fortunes to those measurements, you can bet your
bottom dollar that what you get is what they want you to get or what they're
willing to let you have. (I don't think PCTers will have much trouble
articulating the mechanics behind that one.)

What I don't understand is why you got so hot under the collar in connecton
with Bruce Abbott for saying, "I would say that an incentive to a person is
something the person wants." That seems a perfectly reasonable definition
of an effective incentive. It also suggests that a boss might be able to
sit down and work out a program of action on the part of an employee and set
up a systems of incentives that will work in that situation with that
employee. But I'm not a fan of an approach that amounts to springling
incentive dust over everyone and expecting them to concentrate and channel
their energies along the lines I wish. There ain't no such an animal.

If you want to know how I REALLY feel about incentives, ask me again
sometime. :slight_smile:
Regards,

Fred Nickols
Senior Consultant
The Distance Learning Company
http://home.att.net/~nickols/distance.html
nickols@worldnet.att.net

[From Rick Marken (971128.1540)]

Me:

NB Fred Nichols: Do you still think that it's control theorists
who use language idiosyncratically? :wink:

Fred Nickols (971128.1615 ET) --

I HATE messages like that!

And I HATE replies like that!

What I don't understand is why you got so hot under the collar
in connecton with Bruce Abbott for saying, "I would say that
an incentive to a person is something the person wants."

Because (as I said) it was an attempt to make it seem like
psychologists understand purposive behavior when they do not. As
Bill Powers (971127.0941 MST) explained, calling something an
"incentive" implies that that something is an external cause
of behavior. Abbott (and now you, I guess) seem to think that
"incentive" is just an arbitrary sound (like "flamixli") that
happens to be used to describe a wanted perception; what in
PCT is known as the reference state of a controlled variable.

I believe you had argued some time ago that advocates of PCT
couldn't expect to be effective if they used words (like "control",
I believe) idiosyncratically (when, in fact, we use words like "control"
exactly as they are defined in the dictionary). Now
here we have what is clearly an indiosyncratic use of the word
"incentive" by behaviorists. If you look in the dictionary
(if you can find it in your new library;-)) I don't think you will
see "incentive" defined once as "something a person wants" (but
I may be wrong; I'm just guessing because my dictionary is put
away while we paint our library;-)).

If you want to know how I REALLY feel about incentives, ask
me again sometime. :slight_smile:

I was less interested in how you feel about incentives than in
how you feel about the use of language by behaviorists. You didn't
tell me, but I can tell from your post that you cut a lot more
slack for behaviorists (who seemto use language arbitrarily)
than than for control theoriests (who, in my opinion, use it
impeccably).

Best

Rick

···

--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken/

[Ken Kitzke (971129.1100 ET)]

<Fred Nickols (971128.1615 ET) wrote:>
<It simply is not true, as so many bone-headed pontificating fools
pronounce it so, that "You get what you measure." That is pure
unadulterated bullshit!>

I see your BS and raise you two diarrheas!

Measuring is very helpful for knowing how, as an individual or a group, you
are doing. Just seeing the measure getting better or worse or staying the
same can become one's controlled variable that illicits behavior change to
restore the desired perception. It is consistent with PCT and has been
extraordinarily successful at clients in maintaining and improving the
performance of people and companies.

We measure results. Some people will try to fudge results. Some people
lie. It is usually pretty easy to catch such manipulators of the truth. I
encourage my clients to treat that as a big time disturbance and send any
such employees to their competitors.

Measures help people understand what matters most. Today, due to
technology, foreign competition and other factors, product prices may fall
at 10% per year. Workers who think they deserve a raise because they made
5% more products than last year need an education about which measure is
most important to paying their wages: sales dollars or units.

I don't proclaim "You get what you measure." I do proclaim I have seen
performance improve quickly from nothing more than measuring it and
displaying it to those who are responsible. Over and over again. If this
is beyond your realm of experience or understanding, that is unfortunate.
I think it is PCT and explains the behavior flawlessly.

I also claim that if a result is important, you should measure it and
display it for everyone to ponder. I claim that it is difficult to manage
things you can't or won't measure. Does this make me a bone-head?
Actually, without my skull bone, my brains would ooze all over my keyboard.
  8-) I guess I'm guilty.

< When you measure something, and tie others' fortunes to those
measurements, you can bet your bottom dollar that what you get is what they
want you to get or what they're willing to let you have.>

Now, you have illuminated the real culprit behind your emotion and
corresponding unprofessional language. And, here, we might agree. It's
the incentive, the carrot, that makes people look like jackasses. It's the
incentive that makes fools out of people. It's the incentive that doesn't
work and raises conflict. Here we agree, I think, based on PCT? It is not
the measure that is the problem, it's the incentive.

<As a general practice, I nod agreeably when clients talk about the
importance of structuring incentives to support and encourage the behaviors
desired, but I don't "do" incentive systems as part of my consulting
practice. I think they're a waste of time.>

Are you controlling for being paid or for being honest? When my clients
use the word incentive, or the idea of reinforcement, I tell them
emphatically that it won't work reliably and will only cause conflict among
people, including encouraging bold faced lies about results. After
explaining from PCT why they do not work reliably, and thanking Bill Powers
for insight, if they go ahead, I often separate rather than be part of
their predictable mediocre results or total failure.

<More important, to impose contingent relationships between
rewards/incentives/carrots and behavior/actions/results is to try to
exercise external control. If I know anything about external control
systems, I know they have to be simple in nature and few in number (and I
owe that insight to Peter Drucker, by the way).

It sounds like you are bragging about Peter Drucker? I would be
complaining. Is this the same Mr. Drucker, or is it Dr. Drucker the
pontificator extraordinaire, who is the father of external heirarchial
control known as Management By Objectives (MBO)? Objectives are usually
targets for achievement, including often serving as incentives for
executive bonuses.

Westinghouse, my employer for two decades, liked Drucker a lot. Who would
guess that Druker would outlive Westinghouse? The teacher is healthier
than his advice.

Some day, when all the textbooks of the S-R behaviorists are thrown by Rick
Marken on a pile and publically burned, I'll be there chucking my Harper &
Roe *The practice of management* with Peter F. (might be for Fool? 8-))
Drucker's likeness on the cover into the fire with them.

I would spare the man's life, even if he didn't apologize for his
behaviorist management philosophy. But, without any remorse, I would
probably make him sit in cow manure for an unforgettable period until he
would at least give an audience to Bill Powers.

Actually, if Drucker is still alive, he would be about 88 years old. If
alive, I might have mercy on him and just demand he talk to Bill so he can
willingly and purposively burn his own books on management and feel like a
hero.

From one bone-headed pontificating fool to the rest of you. Peace.

Kenny

[From Bill Powers (971129.1009 MST)]

Ken Kitzke (971129.1100 ET)--

A big hurrah for your simple, direct, and (as far as I'm concerned) correct
analysis of incentives and measurements. The man in the trenches is the one
who knows what really works.

Best,

Bill P.

[From Ken Kitzke (971129.2000 EST)]

Bill Powers (971129.1009 MST) said:
<A big hurrah for your simple, direct, and (as far as I'm concerned)
correct analysis of incentives and measurements. The man in the trenches is
the one who knows what really works.>

Bill, I'm so low in the ocean trenches, even the whale dung hits me on the
head. :sunglasses:

But, thanks for the pleasant accolade. It is a reward to me. I hope you
did not say that as an "incentive" so I would feel obligated to thank you
and praise you? :sunglasses:

Rewards are nice, especially coming from those whom you respect. Now, I'm
fired up, feeling invincible and on a role. So, I'm really going out on a
limb.

Rewards can and do encourage behavior. Does that make rewards incentives?
Not for me. Rewards are freely offered and freely accepted without
contingencies set by the provider for the receiver. They are not
incentives because there is no manipulation.

So, here is a big {{{HUG}}} for you. I hope you perceive it as a reward
and not a manipulative incentive. It requires nothing of you in return.
You already did what you needed to do to deserve it.

You have the patience of Job and the kindness of Mother Theresa as I
perceive you. :sunglasses: I don't know what you said or felt so Hans would say
good-bye (anyone who understands PCT would know Hans made this behavioral
choice to control *his perceptions* and you had no control over him, don't
they?)

I bet he is working on a modern mathematical control theory model to show
once and for all that the way you say things will even most likely make
Peter Drucker say "good-bye friends."

Seriously, seldom do you fine people on CSG talk about anything I can
understand or some associated real-life observational experience. Don't
worry. I'll slip back into my cave again and maybe post something just
before Vancouver. Peace.

Kenny

[From Tim Carey (971130.18.30)]

I've been following with interest the discussion on incentives and, while
I'm enjoying it, it seems fairly similar to the discussion on reinforcement
not so long ago. I may be way off the mark but it seems to me that the
discussion is one of deciding whether you agree with an observation or with
an explanation for that observation.

An individual, for example, could observe person A give person B some money
and then notice that person B does more of whatever they were doing before
they got the money. To me, this observation could be interpreted by a
number of different theories. Once you label this particular observation
however as an act of reinforcement or once you label the thing that person
A gave person B as an incentive, you are indicating which particular
theoretical perspective you are explaining the observation from. This
particular theoretical perspective is probably one that attempts to explain
behaviour or actions as the end product in an attempt to predict and
control the behaviour.

An alternative theoretical perspective, however, that viewed behaviour as a
means to an end (an not the end in itself) and was designed to explain the
phenomena of control processes, would also be able to offer an explanation
for the above observation. The explanation from this theoretical
perspective would not use terms such as reinforcement and incentive because
in this perspective they don't exist. Just like controlled variables and
feedback functions don't exist in other theoretical perspectives.

Just a thought .....

Cheers,

Tim

[From Bruce Gregory (971130.0700 EST)]

Tim Carey (971130.18.30)

An individual, for example, could observe person A give person B some money
and then notice that person B does more of whatever they were doing before
they got the money. To me, this observation could be interpreted by a
number of different theories. Once you label this particular observation
however as an act of reinforcement or once you label the thing that person
A gave person B as an incentive, you are indicating which particular
theoretical perspective you are explaining the observation from. This
particular theoretical perspective is probably one that attempts to explain
behaviour or actions as the end product in an attempt to predict and
control the behaviour.

An alternative theoretical perspective, however, that viewed behaviour as a
means to an end (an not the end in itself) and was designed to explain the
phenomena of control processes, would also be able to offer an explanation
for the above observation. The explanation from this theoretical
perspective would not use terms such as reinforcement and incentive because
in this perspective they don't exist. Just like controlled variables and
feedback functions don't exist in other theoretical perspectives.

Yes, I think that's right. I would add that Bruce Abbott seems to be trying to
have it both ways. As a result he is unhappy and so are some of us.

Bruce

[From Bruce Gregory (971130.0730 EST)]

Tim Carey (971130.18.30)]

An alternative theoretical perspective, however, that viewed behaviour as a
means to an end (an not the end in itself) and was designed to explain the
phenomena of control processes, would also be able to offer an explanation
for the above observation. The explanation from this theoretical
perspective would not use terms such as reinforcement and incentive because
in this perspective they don't exist. Just like controlled variables and
feedback functions don't exist in other theoretical perspectives.

One way to look at the situation is that there is a profound difference
between trying to describe the classical world in the vocabulary of
quantum physics and trying to describe the quantum world in the
vocabulary of classical physics. The first approach works. The second
does not. (Who is trying to do what on CSGnet is left as an exercise
for the readert :wink: Hi, Isaac!)

Bruce

[From Bill Powers (971130.0606 MST)]

Tim Carey (971130.18.30) --

An individual, for example, could observe person A give person B some money
and then notice that person B does more of whatever they were doing before
they got the money. To me, this observation could be interpreted by a
number of different theories. Once you label this particular observation
however as an act of reinforcement or once you label the thing that person
A gave person B as an incentive, you are indicating which particular
theoretical perspective you are explaining the observation from. This
particular theoretical perspective is probably one that attempts to explain
behaviour or actions as the end product in an attempt to predict and
control the behaviour.

That is precisely it, Tim. However "atheoretical" and "empirical" a
scientist claims to be, his or her explanations reveal the theoretical
framework.

An alternative theoretical perspective, however, that viewed behaviour as a
means to an end (an not the end in itself) and was designed to explain the
phenomena of control processes, would also be able to offer an explanation
for the above observation. The explanation from this theoretical
perspective would not use terms such as reinforcement and incentive because
in this perspective they don't exist. Just like controlled variables and
feedback functions don't exist in other theoretical perspectives.

I think we should turn this discussion over to you. Clearly and
economically said.

Best,

Bill P.

[Fred Nickols (971130.1220 ET)]

Ken Kitzke (971129.1100 ET)

Fred Nickols (971128.1615 ET) wrote:>
It simply is not true, as so many bone-headed pontificating fools
pronounce it so, that "You get what you measure." That is pure
unadulterated bullshit!>

I see your BS and raise you two diarrheas!

Pass.

Measuring is very helpful for knowing how, as an individual or a group, you
are doing.

Agreed; provided you are measuring something in which you are interested and
want to know about. Measuring the wrong things, measuring to drive others'
behavior, and measuring just for the sake of measuring are among some of the
most destructive activities found in organizations.

Just seeing the measure getting better or worse or staying the
same can become one's controlled variable that illicits behavior change to
restore the desired perception.

Hmm. I assume you meant "elicits" and that makes me ask how you can get
away with using such a bonfide behaviorist term as that and get away with
it? It sounds an awful lot like you're saying that the results of
measurement act as a stimulus to action. Is that what you meant?

It is consistent with PCT and has been
extraordinarily successful at clients in maintaining and improving the
performance of people and companies.

By "it" I assume you meant measuring or using measures in support of
obtaining and maintaining performance levels. If so, I agree that
instituting measures where none before existed frequently generates a lot of
activity. I'm not convinced that the activity is always productive or that
it produces the results sought by those who instituted the measures.

We measure results. Some people will try to fudge results. Some people
lie. It is usually pretty easy to catch such manipulators of the truth. I
encourage my clients to treat that as a big time disturbance and send any
such employees to their competitors.

Ah, I see; play the game by my rules or else. That sounds remarkably like
the game the rest of the world plays. What's different about that?

Measures help people understand what matters most.

Matters most to whom?

Today, due to
technology, foreign competition and other factors, product prices may fall
at 10% per year. Workers who think they deserve a raise because they made
5% more products than last year need an education about which measure is
most important to paying their wages: sales dollars or units.

Actually, I think most workers think they deserve a 5% raise because they
see their bosses walking off with millions and they don't think 5% is too
much for them to ask.

I don't proclaim "You get what you measure." I do proclaim I have seen
performance improve quickly from nothing more than measuring it and
displaying it to those who are responsible. Over and over again. If this
is beyond your realm of experience or understanding, that is unfortunate.

Nope; it's not beyond my experience. Many, MANY years ago, while still in
the Navy, I was moonlighting as the night shift supervisor in a telephone
sales office for Time, Inc. (actually, a subsidiary called Life Circulation
Corporation). I persuaded the office manager to let me post individual
sales results. (He had resisted because he didn't want to stimulate harmful
competition.) Sure enough; sales results more than doubled. A couple of
sales reps left because they couldn't stand the pressure. They had been
marginal performers anyway so no one missed them.

I think it is PCT and explains the behavior flawlessly.

Maybe, maybe not. There are certainly other explanations which, depending
on your theoretical orientation, you would treat as more or less plausible.
That, I believe, was the gist of Tim Carey's very lucid post.

I also claim that if a result is important, you should measure it and
display it for everyone to ponder. I claim that it is difficult to manage
things you can't or won't measure. Does this make me a bone-head?

Nope, provided you're not shooting yourself in the foot. Example: The
president of a large foodstuffs manufacturer once set out to break all sales
records. He set the targets, provided the incentives, and ignored all the
protests and counter-arguments. The sales force went out and broke all
sales records; they, after all, knew what was important. The plants
incurred all kinds of overtime trying to keep up with the orders and
couldn't. Customers defected to other manufacturers who could fill their
orders. The company did break all sales records. It also posted the worst
loss in its history. The president was later fired. Did he measure? You
betcha! Did what he was measuring serve to "communicate what matters most"?
You betcha! Did he measure the wrong thing? I think so; so do lots of others.

When you measure something, and tie others' fortunes to those
measurements, you can bet your bottom dollar that what you get is what they
want you to get or what they're willing to let you have.

I stand by that comment (see the president's story above).

Now, you have illuminated the real culprit behind your emotion and
corresponding unprofessional language.

Emotion? Unprofessional language? Me? But let's digress anyway...

I'm not a professional in the sense that I practice a profession such as law
or medicine or engineering. Nor am I a professional in the sense that I am
a member of some occupation that has a code of ethics to which I adhere. If
I am a professional at all, it is in the sense of being highly skilled and
able to command a premium in pay for applying those skills to the problems
of my clients. All that said, if anyone on this list took offense at my use
of "bullshit," please know that no offense was intended.

And, here, we might agree. It's
the incentive, the carrot, that makes people look like jackasses. It's the
incentive that makes fools out of people. It's the incentive that doesn't
work and raises conflict. Here we agree, I think, based on PCT? It is not
the measure that is the problem, it's the incentive.

Actually, I don't think we do agree. Incentives work all right, it's just
that all too often they don't work as intended. In the story above, it was
the measure that was the problem, not the incentive. The incentive simply
put teeth in a bone-headed measure. The sales reps were laughing all the
way to the bank. They pocketed millions in commissions on confirmed orders.
Failure to deliver wasn't their problem.

<As a general practice, I nod agreeably when clients talk about the
importance of structuring incentives to support and encourage the behaviors
desired, but I don't "do" incentive systems as part of my consulting
practice. I think they're a waste of time.

Are you controlling for being paid or for being honest?

For being paid and for repeat business and for being honest and, most of
all, for choosing my shots. If I confronted every fool notion I heard
uttered, that's all I'd find myself doing and I'd be completely without clients.

When my clients
use the word incentive, or the idea of reinforcement, I tell them
emphatically that it won't work reliably and will only cause conflict among
people, including encouraging bold faced lies about results. After
explaining from PCT why they do not work reliably, and thanking Bill Powers
for insight, if they go ahead, I often separate rather than be part of
their predictable mediocre results or total failure.

Well, that fits with the other things you've said.

More important, to impose contingent relationships between
rewards/incentives/carrots and behavior/actions/results is to try to
exercise external control. If I know anything about external control
systems, I know they have to be simple in nature and few in number (and I
owe that insight to Peter Drucker, by the way).

It sounds like you are bragging about Peter Drucker? I would be
complaining. Is this the same Mr. Drucker, or is it Dr. Drucker the
pontificator extraordinaire, who is the father of external heirarchial
control known as Management By Objectives (MBO)? Objectives are usually
targets for achievement, including often serving as incentives for
executive bonuses.

I admire Drucker but I can't brag about him; he isn't mine. In any event,
please explain for me how you reconcile your admiration for measurement with
your dislike for Drucker and MBO?

Westinghouse, my employer for two decades, liked Drucker a lot. Who would
guess that Druker would outlive Westinghouse? The teacher is healthier
than his advice.

That remark presupposes Westinghouse took his advice, which it didn't and
couldn't because organizations don't do anything, people do. I daresay
Peter Drucker outlived Westinghouse because those who were running
Westinghouse into the ground weren't subject to sufficient checks and
balances on their exercise of authority -- but, what the heck, that's a
whole different subject.

Some day, when all the textbooks of the S-R behaviorists are thrown by Rick
Marken on a pile and publically burned, I'll be there chucking my Harper &
Roe *The practice of management* with Peter F. (might be for Fool? 8-))
Drucker's likeness on the cover into the fire with them.

Book burning? Rick's a man of strong opinions but I doubt he's into book
burning. Actually, it's Harper & Row, and the "F" stands for Ferdinand.

I would spare the man's life, even if he didn't apologize for his
behaviorist management philosophy. But, without any remorse, I would
probably make him sit in cow manure for an unforgettable period until he
would at least give an audience to Bill Powers.

I'm sure Drucker would be heartened by your magnanimity. I don't think Bill
wants an audience with Drucker but I'll let Bill speak for himself.

Actually, if Drucker is still alive, he would be about 88 years old. If
alive, I might have mercy on him and just demand he talk to Bill so he can
willingly and purposively burn his own books on management and feel like a
hero.

Last I heard he is still alive and he just turned 88 early in November.

From one bone-headed pontificating fool to the rest of you. Peace.

Peace to you, too.
Regards,

Fred Nickols
nickols@worldnet.att.net