ADVANCED

N RESEARCH
RESOURCES

ORGANIZATION

 

 

4330 East-West Highway, Washington, D. C. 20014 = 202 / 986-9000

 

June 27, 1978

Dr. Joshua Lederberg
President

Rockefeller University
New York, N.Y. 10021

Dear Dr. Lederberg:

On the same day that I received your request for a reprint of my
peer assessment article, the Wall Street Journal published an article which
made the reasons for your interest in this topic clear. I am intrigued by
the possibility of applying peer assessment methodology to the peer review
activity. JI hope you won't think it presumptuous of me to take this op-
portunity to convey some of my thoughts on the application of peer assess-
ment methods in the situation you confront.

Appraising the performance of research scientists is of course in-
herently difficult due to the need to define and set standards for crea-
tivity. When these assessments are performed by peers these are added pit-
falls to avoid (e.g., the application of excessively high performance
Standards, the distortion of one's perceptions of another's achievements in
ego-threatening circumstances). When the peers doing the assessments are .
outsiders, the situation becomes complicated further by their need to justify
their existence by finding faults, their access to only limited samples of
performance (which may be selected by the ratee), their probable exposure to
only the outcomes rather than also to the process of the ratee's work, and
their lack of acquaintance with the conditions that may have constrained
performance.

Other questions will arise that are common to all appraisal systems.
Is the appraisal to be formative (meant to eliminate defects) or summative
(pass-fail)? If it is formative, it must furnish objective, informative feed-
back to ratees on how they can do their work better. Are ratees to have the
right to appeal? If so, what adjudicative body will hear such appeals and
what rules and standards for evidence will prevail? Note that the choice of
a peer assessment method sets fairly rigid limits on the types of process and
content that can be employed in an appraisal system.

a division of RESPONSE ANALYSIS CORPORATION, Princeton, New Jersey
This is only a sampling of the issues that arise when the need for
performance appraisal is confronted. I have spent the last five years
intensively studying both the mechanics and dynamics of appraisal. One
outcome of this work has been my proposal of new model for the measurement
of performance which may be very applicable to your needs. The basics of
this model are described in the other piece I have enclosed. Note that
the proposed way of operationalizing the new model permits unlimited ratee
participation in all stages of the rating process, as well as the possibility
of different people being involved in the choice of job aspects on which to
rate performance, the setting of standards for performance, and the actual
rating of performance.

I am basically a researcher but I do make myself available as a
consultant in my specialty areas. Your peer review problem appears to be one
on which I could make useful inputs. If the need for such input arises, please
feel free to call upon me.

Singerely,
i? fifa

ae p ef .

ag ey S//Kane, Ph.D.
Research Scientist

JSK/ygf