FITNESS REPORTS

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP67-00134R000200050014-8
• Release Decision: RIPPUB
• Original Classification: S
• Document Page Count: 2
• Document Creation Date: December 15, 2016
• Document Release Date: August 6, 2004
• Sequence Number: 14
• Publication Date: March 18, 1964
• Content Type: MEMO

OPTIONAL FORM NO. A10
-1-1- Approved For Rele 2004/08/17 -'('AA=[ t)P,67-00134ROof 050014-8
UNITED STATES GOVERNMENT L ij all t- Fr
Memorandum
DATE: 18 March 1961+
SUBJECT: Fitness Reports
1. I haven't really been delaying answering; I have just been out of
town. This is a formidable problem you have tackled, however, and I wouldn't
be above evading it. I am sure I haven't any solutions, but I do have some
ideas which I am glad to pass along. In answer to your first question, "Is it
likely that the Army Officer Efficiency Report with its 'expected distribution
of 100 officers rated' formula is likely to be any more meaningful than our
fitness reports", I don't know. I know that on its one form the Army tries to
increase variability of ratings between people by use of this hundred officers
formula. On another form they try to increase variability in ratings on one
individual by exhortations to avoid the "halo effect". Similar formulas and
admonitions could be made an integral part of the instructions on any rating
sheet. Whether or not they would be effective could probably be determined best
by looking at the Army results. I don't have the data, but assume we can find
out what sort of distribution the ratings follow on officers' efficiency reports.
But even here I am not sure that the proof of the pudding would be in the eating.
What worked for the Army might not work for us. I suspect that Army officers have
more in common than Agency employees, and that it is easier for an Army rater to
compare any one ratee with a large group of other people. A major problem in
fitness report ratings is defining a reference group. For example, someone rating
an A&E psychologist might do it quite differently depending upon whether he were
rating the individual against an ideal standard, against other psychologists,
against other support people, or against other Agency employees of the same grade.
A forced distribution of ratings among individuals works best when you have one
person rating many others in the same jobs. In that case you can insist on grading
by the curve or on a fixed percentage of people at each rating. If a person is
only rating a few others, it is easy for him to rationalize that they are all
top-notch people; and, in fact, they may be.
2. I do know that the service in the past did not have much better luck
with their fitness reports than we do. About ten years ago I was involved in a
large scale psychological study of Air Force officers. Hundreds of thousands of
dollars and much valuable time of Air Force officers and psychologists was spent
amassing a plethora of data intended to be combined to predict officer effective-
ness. Not one trait was unrated nor one trauma unanalyzed, but the entire study
nearly foundered on the shoals of attenuated criteria; all got the same ratings
on their efficiency reports. The data could not possibly be related to good or
bad performance in such circumstances. Luckily some value was salvaged because
someone had had the foresight to obtain experimental, confidential ratings un-
related to the OER's. I frankly doubt that things have improved much.
Approved For Release 2004/0q 7 IA-RDP67-00134R000200050014-8
 Approved For Rele*96 2004/08/17 CIA-RDP67-00134R00050014-8
3. On your second point, basically I think I am against minor changes in
fitness reports except for more effective instructions. The fitness report
represents a chronological history of the individual's performance, and changes
in the form make for problems of confusion when a reviewer is attempting to
study a personnel file. It may be completely unfeasible, but it seems to me
worthwhile at least to consider the possibility of a fairly radical change. If
one 01. the purposes of the fitness report is to prov3 e a frank opinion of the
individual's performance for the personnel file, it fails. The present system
suspect, will provide mainly superficia- n?ormat on
cu ss o . As
requires that the n vidual be shown his fitness report, a
esu .
o has ever sa on -a review
s ere is more m e a n i n g f u l which is usually presented orally ax?d
informally. It may be that the interest of the Agency could be served by in-
corporating some of this into a portion of the fitness report not shown the
individual. It may also be that this smacks too much of authoritarianism.
Perhaps the fitness report has more value as a motivating device and as a focus
for conversation between supervisor and supervisee than, as a valid rating instru-
ment. Although I am pessimistic about results, I am greatly in favor of a review
of the system. If we can help by reviewing what is being done elsewhere and by
surveying personnel research, please ask.
25X1
Approved For Release 2004/08/11T: 'CIA-RDP67-00134R000200050014-8