Perf Measurement in the News articles LO24641

From: Roger Key (
Date: 05/19/00

Replying to LO24618 --

Hi All

Jack Zigon posted some incentive based articles for comment.

Thanks Jack for the invitation.

Jack and those of you who subscribe to the Deming Electronic Network may
well recognise my name and my interest in Deming. Not surprisingly I
looked at the following article first.

>* Rewards, Interest and Performance: An Evaluation of Experimental

Jack precis is as follows.

>>Rewards, Interest and Performance: An Evaluation of Experimental Findings
>An analysis of more than 25 years of research concerning rewards and
>performance indicates that rewards can be used effectively to enhance
>interest and performance. This article provides a research-based rebuttal
>to Alfie Kohn's opinion-based message that rewards don't work. ACA
>Journal, 12/97.

Now did I look at the article so I could attack it or so I may learn.. I
would hope the latter.

Unfortunately I feel inclined to question the validity of the article. In
the most part it is well written but totally misses the point. Further it
is very journalistic in some of its content and make causative links that
have no substantive basis. If I had received the article as a report from
a post grad student I would have had to spend some time with the student
trying to understand if they were writing a data and scientific method
based report or an article for a general interest magazine.

My biggest grumble with the article is - however - that it has not learnt
anything from two of the people that it seeks to attack / address. The
two people are Alfie Kohn and Dr W. E. Deming. (no surprises here!)
Specifically what I am concerned about is that Deming from the basis of
indepth knowledge of statistics and profound understanding of systems
thinking and Kohn from a more populist but educated basis question the use
of the results from lab studies to real world situations. Deming, Juran
etc note that the individual has impact on less than 20% of the outcome of
the system - this alone is argument against incentives on the simple basis
of why waste time on the 20% when you can have more bangs per buck in the
80+% - the writers seem to either be unaware of this or choose to ignore
it as it will possibly invalidate many of the lab studies they base their
results on. Secondly they seem to be unaware of variation, that
performance may vary but the result is systemically the same. Example
rolling dice. You may get different numbers but the result is the same -
systemicly. There appears to be no addressing of this within the study.
Simulaly many of the assertions and suggestions are simplistic and linear.

It may be that Deming and Kohn are wrong, but I think that this paper
claims to address that but (in Kohn's words about Skinner) They have
studied rats and pigeons and written about human beings. The Drs who have
written this paper have studied heavily simplified 'linear' systems, taken
steps to remove the dynamic complexity, ignored the impact of the
environment, materials / information, methods and equipment and the impact
they may have on the system. Addressed all the (marginal) results to the
people and then written about complex adaptive systems.

They have also identified that verbal incentives are good for everyone.
This flies in the face of the work by Bandler, Grinder, Dilts etc and the
whole field of psycholinguistics and NLP.

The first sentence of the article is also blatantly wrong!

Jack's precis for the article states -

>This article provides a research-based rebuttal
>to Alfie Kohn's opinion-based message that rewards don't work. ACA
>Journal, 12/97.

Kohn bases his assertions on over (to my understanding) 2000 reproducible,
science based published papers. The paper under review is based on 100.
Kohn bases some of his work on W. E. Deming who based his understanding on
a highly detailed understanding of statistics and over 70 years of
personal experience in developing and improving systems, the two writers
in the article do not muster the level of understanding and experience
between them.

The writers of the article, like Kohn, have an axe to grind. They both do
their research in the study on incentive, and probably make their living
from it - as does Kohn. I think Jack's bias shows in his precis - he
makes his living from selling incentives - I show my bias in that I make
my living from trying to help companies get over the damage caused by the

I do not think that this article adds much new to understanding. Their
conclusions - from the stats - is that at best tangible incentives have
minimal impact, so on a cost benefit basis I would ask "Why bother, why
not spend time on making the job more interesting and win in more ways?".
The data also suggests that verbal incentives have more effect. I would
need to see the profile of the people studied as 'internal' sorting needs
verbal incentives like a hole in the head. I am not sure where the peak
is on the 'internal - external' metafilter. But this sounds alarmingly to
me like teamwork and interaction with people allows them to input into the
work they are doing and make it better. This is not incentive it is
called leadership.

However the conclusions drawn by the authors is to support their avowed
position that incentives are a good idea. I am not convinced that the
data, carefully selected, and presented supports this.

There again I am horribly biased!


Roger C. Key
Prescient - The Whole as One
(44) 01639 871062
Learn about Deming and CQI on line at

 - The web site. coming to a PC near you soon! -


Roger Key <>

Learning-org -- Hosted by Rick Karash <> Public Dialog on Learning Organizations -- <>

"Learning-org" and the format of our message identifiers (LO1234, etc.) are trademarks of Richard Karash.