Eric Grosch, Letter to Dr. Morgenstern on LOR - Semmelweis ...
Eric Grosch, Letter to Dr. Morgenstern on LOR - Semmelweis ...
Eric Grosch, Letter to Dr. Morgenstern on LOR - Semmelweis ...
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
divisi<strong>on</strong>s. (18)<br />
Kingsbury enumerated those allegedly ideal c<strong>on</strong>diti<strong>on</strong>s:<br />
Ratings, <str<strong>on</strong>g>to</str<strong>on</strong>g> be reliable, necessitate (1) averaging three independent ratings, each made <strong>on</strong> an<br />
objective scale; (2) these scales must be comparable and equivalent, made in c<strong>on</strong>ference under<br />
expert supervisi<strong>on</strong>; (3) the three raters must be competent <str<strong>on</strong>g>to</str<strong>on</strong>g> rate.[18]<br />
Paters<strong>on</strong> joined, with slightly different wording, in affirming Kingsbury's ideal c<strong>on</strong>diti<strong>on</strong>s (2) and<br />
(3) and in thus implicitly alluding <str<strong>on</strong>g>to</str<strong>on</strong>g> pitfalls of the method:<br />
. . . Ratings should be accepted and filed for use <strong>on</strong>ly from those who have proved themselves<br />
capable of accurately judging human qualities. . . a rating scheme will not work au<str<strong>on</strong>g>to</str<strong>on</strong>g>matically. It<br />
must be closely supervised preferably by trained pers<strong>on</strong>nel research workers who must c<strong>on</strong>tinually<br />
subject the ratings <str<strong>on</strong>g>to</str<strong>on</strong>g> critical analysis and assist in training executives in proper use of the method.<br />
There is no escape from this requirement.[19]<br />
Paters<strong>on</strong> and Kingsbury omitted menti<strong>on</strong> of what specifics the training they proposed for the<br />
pers<strong>on</strong>nel-research workers should comprise and accomplish but they presumably intended,<br />
am<strong>on</strong>g other things, that the trained supervisors should somehow ensure separate evaluati<strong>on</strong> of<br />
labeled traits <str<strong>on</strong>g>to</str<strong>on</strong>g> exclude Thorndike's halo-effect; then, by averaging, fine-tuning, adjustment and<br />
manipulati<strong>on</strong> of the scores from at least three raters, all of whom knew how <str<strong>on</strong>g>to</str<strong>on</strong>g> provide accurate<br />
ratings (presumably assessed by the raters' mutual agreement <strong>on</strong> each candidate's score <strong>on</strong> each<br />
criteri<strong>on</strong>) obtain a set of ratings c<strong>on</strong>sistent with the aggregate global impressi<strong>on</strong> each candidate<br />
made <strong>on</strong> the raters (the candidate's halo). The circularity of the rati<strong>on</strong>ale seems inescapable.<br />
Prior <str<strong>on</strong>g>to</str<strong>on</strong>g> receiving requests <str<strong>on</strong>g>to</str<strong>on</strong>g> fill out forms c<strong>on</strong>sisting of Likert-scale ratings <strong>on</strong> others'<br />
performance, I have never received any of the extensive training or testing <str<strong>on</strong>g>to</str<strong>on</strong>g> prove myself<br />
“capable of accurately judging human qualities,” nor, I daresay, has any appraiser of my<br />
performance received such training and testing, <str<strong>on</strong>g>to</str<strong>on</strong>g> my knowledge. The origina<str<strong>on</strong>g>to</str<strong>on</strong>g>rs of such forms<br />
seemed <str<strong>on</strong>g>to</str<strong>on</strong>g> assume that the rating scemes would “work au<str<strong>on</strong>g>to</str<strong>on</strong>g>matically,” c<strong>on</strong>trary <str<strong>on</strong>g>to</str<strong>on</strong>g> Kingsbury's<br />
adm<strong>on</strong>iti<strong>on</strong>.<br />
Rugg may have had more insight:<br />
. . . The unordered -- yes, the chaotic -- character of the judgments appears, irrespective of what<br />
traits are c<strong>on</strong>sidered or of what kinds of scales are compared. I now believe that the evidence<br />
establishes the futility of obtaining single “ratings” <strong>on</strong> point scales of such dynamic qualities as<br />
“intelligence,” “pers<strong>on</strong>al qualities,” “general work,” and the like.[20]<br />
Paters<strong>on</strong> cauti<strong>on</strong>ed and predicted:<br />
These rating methods should not be looked up<strong>on</strong> as perfect or final. Further research is necessary,<br />
and industry will profit . . . as progressive, experimentally minded executives realize the scope of<br />
the problem and engage in the necessary research . . . <str<strong>on</strong>g>to</str<strong>on</strong>g> develop newer and more reliable methods