17
others in the group were compared w ith
these selected individuals and rated ac-
cording to the value assigned the man
they most resembled. A variation of this
method was used in some O.C.S. train
ing in W orld W ar II. Every man rated
every-other man in his class on selected
traits,; lining them up from poorest to
best in a straight array. / > - .
1
The most popular method in use to
day ds thé graphic rating scale. It con
sists of a list of traits or activities to be
evaluated, followed by a line along
which the rater marks. He is guided
by descriptions along this line ranging
from low to the highest degree of that,
tràit. The different types of graphic
scales are numerous, but most often
there is a continum of equally divided
units along the line of qualities and
the evaluation for that factor-is inter
preted numerically.
Another process makes use of a check
list; and has reached its greatestrefine
ment in the "Probst System” developed
by J. B. Probst formerly of the St. Paul
Civil Service Commission. This rating,
consists of one hundred items designat
ing modes of behavior or characteris
tics^ for example: "Indifferent,” "too
much self¿importance,” "always uses
good, jufgfhent,,” "cranky disposition,”
"work alwàÿsf up to d ate/’^etg. The
rater' checks only those o f the hundred
items which he ■ believes apply. Each
factor is assigned a relative value, and
the final scoring is most often done by
the central agency.
The California State Personnel Board
rating carries the Probst Systçm qne step
further. À variety of /mixed items is
also used, but each item is qualified >by
checking,^alw ays,” ''nearly always,”
"generally,” , "seldom,” and "never.’’
Also California has over tw enty/five
différent forms designed fo r various
types of occupations. All ratings are
machine scored hy thç personnel,board.
Some Problems of Development
I Classification' has to I do w ith what
an employee does; service ratingj -ydth
how w e llhe does it. Evaluation of whtft
an employee does lends itself easily to
determination by disinterested, outside
personnel men not connected with the
department; 'But how well he does it
can most often be determined only by
his immediate supervisors. There Is
agreement on the tru th of this conclu-
sion; but as to the best method to fol
low in getting the supervisors’ rating,
there is great disagreement. One, school
of thought believes that the supervisor,
who is closest to the employee’s work,
should do the actual rating. The oppos
ing belief is that the supervisor should
be limited to merely reporting the ob
servable: significant traits of the em
ployee with the central personnel agency
doing the actual scaring either manually
of b y machine.
Neither path is,a-sm ooth one; both
are filled with problems and difficul
ties. Service fating procedures have not
yet reached the super highwayu status.
It is a m atter of deciding which of two
rough roads leading to the same des
tinatio n is easiest.
The choice of the proper'? system
comes down to two factors-—each in
Cpr^iiet w ith the other. On One hand,
the system must be simple,¿flexible,, and
economical; on the other hand it should
be as reliable and valid as possible. A
certain amount of. one factor has to be
sacrificed in order to attain the other.
I The I I' classic aW per sonel people-^eoh-,
sidered a strange and terrible breed in
some quarters—insist that it is impos
sible to make a valid fating of the man
■as a whole, and that specific traits or
items of- /behavior must'» be evaluated
separately. The r e n e g a d ,e personnel
group— also a strange people— m aintain
that regardless of how" much rigmarole
and red, tape-is gone through, it always
'comes back to an overall rating of the
man anyway,’ that it- is a m atter of
judgment, that all such evaluations are
subjective; and- most of all that the
system m ust be simple enough to work.
.These people believe that the whole is.
greater than the sum of the parts; and,
moreover, that the overall personality
and character of the man appear in his
: - (Continued on page