ML20010F370

Urges That Requirement to Retest TMI-1 Operators Be Upheld. IE Investigation Into Exam Cheating Was Truncated & Flawed

ML20010F370
Person / Time
Site:	Three Mile Island
Issue date:	09/02/1981
From:	Weiss E HARMON & WEISS, UNION OF CONCERNED SCIENTISTS
To:	Ahearne J, Bradford P, Palladino N NRC COMMISSION (OCM)
References
NUDOCS 8109100101
Download: ML20010F370 (4)

Text

n 5 t^ b o o h v' i-e n --

HARMox & WEISS [-I 9

ins i sraccT, w.w. 9 / p',u suirc see [ig (J ' W 0"

GAIL M. MARMON WASHINGTON,D.C.2ooo6 *,97 y,,, b [3gg ,,g,g cttvu a. weiss wituaM s. .so n o A w. m  % 33 g g l tcc t. eisMoe g Septe mbe r 2, 1981 g

/

Chariman Nunzio J. Palladino 11 /

Commissionar John F. Ahee-ne Commissioner Peter A. Bradfora p DOckETEo i Commissioner Victor Gilinsl.y Usygg Commissioner Thomas M. Rcoerts 2 ggp 7g U.S. Nuclear Regulatory Commission Z-. 3198f -

Washington, D.C. 20555 C8 0f the M S

\ 4 & g##' -

RE: O_per a tor Chea ting , TMI-l 8fgc4 Gentlemen : C On August 20, 1981, Mr. Herman Dieckamp, President of GPU, wrote to Chairman Palladino seeking a reversal of Mr. Denton's decision to require retesting of the TMI-l SRO and RO candi-dates who took exams on April 21-24, 1981. The Union of Con-cerned Scientists urges you to stand behind the requirement to retest the TMI-l operators. In view of the proven cheating during the April 21-24 examination and the f acts which have since come to light casting grave doubt on the integrity of the examination , and on GPU's ability to enforce the strictest standards of integrity, retesting is clearly necessary in order to ensure that all licensed TMI-l operators are in f act competent to perform their duties.

Mr. Dieckamp states that the I&E investigation found no evidence of cheating or irregularities by candidates other than the two caught cheating. In f act, the I&E investigation was both truncated and flawed. Nonetheless, f acts were dis-covered which establish the potential for more widespread cheating and Laint the April 21-24 exams beyond recovery.

With respect to the scope and depth of the I&E investiga-tio n , you should he aware that only 11 of the 3 4 candidates were interviewed by I&E .J/ At that point, the inves tigation was stopped , apparently because the two persons caught cheat-ing had " terminated their employment " with GP U and , in any case, enough inIormation had been gathered to sustain a case against them. J/ Many questions concerning the broader issue of possible widespread cheating were not pursued, including those identified below.

1/ TMI Unit 1/ Investigation of Alleged Cheating on Operator Licensing Examinations, Case #HQS-81-00 3, August 11, 1981 ( hereinaf ter "I&E In-ves tiga tion " ) at 7.

I&E' Investigation at 50. h 'g.,k 2/ g6 0109100101 810902 PDR ADOCK 05000289 G _ _

,._PDR. ,

e HARMON & WEISS

~

Page 2 In addition, Mr. Arnold, Vice President of GPU, informed all interviewees that a representative of GPU management could be present at their interviews. All interviews conducted on-site were done so with a management representative present . Under such circumstances, it is predictable that the interviewees would f eel cons trained to speak freely about " rumors " of cheating or to provide information which might implicate co-workers.

Despite the brevity of the investigation, the following was disclosed:

1. There had been rumors of cheating and the presence of crib sheets during the exam which had gone so f ar as to come to the attention of Mr. Arnold, who informed the I&E investigators on July 29._3/ A crucial question which the I&E report inexpli-cably fails to address is when these rumors first began to circu-late and when GPU management first learned of them. The exam was given 3 months before the investigation began . If management was aware.of the rumors, it had an obligation to inform NRC immediately or at the least to institute its own investigation.

Did GPU take any action upon hearing the rumors of cheating? I&E does not deal with this question.

2. NRC's supervision of the exam was unquestionably insuffi-ci er.t . The exams were unproctored for very long periods of time - up to 3-4 hours. One proctor was observed reading a novel at the front of the room._4/ Several of the interviewees remarked that the prolonged absence of proctors contributed to rumors of cheating._5/ The I&E follow-up of these rumors was superficial at best. Interviews were never even conducted outside the presence of GPU management.

Moreover, the proven cheating took place in sessions which were at least nominally proctored, yet the proctor incredibly failed to notice 'the cheaters passing papers back and forth, leaving answers face up (which continued through-out the whole exam) and talking to each other._6/ tbnsidering the small number of people taking the test, this cheating would have been detected by a reasonably con-scientious proctor. This is strong evidence that the supervision of even the proctored sessiens of the exams was grossly inadequate.

3/ Id . at 8.

4/ Id. at 12.

5/ Id. at 25, 37. Re: rumors of cheating , see also 33, 39.

6/ Id. at 2, 46, 48.

i

a o . t HARMON Oc WEISS Page 3 _.

Nor is it an answer to assert as Mr. Dieckamp does, that the proctoring was " consistent with NRC past practice." Far from providing reassurance about the TMI exams, if tru e , this assertion simply reveals a prevasive negligence on NRC's part.

Surely it provides no exc use for this incident or any justifi-cation for accepting the results of these tests.

3. The f acts suggest either insensitivity or incompetence on the part of GPU management . The same two individuals cheated in the same blatant way during mock exams administel;d by GPU in early April prior to the NRC tests , _]/ yet the word-for-word simu-larities between their answers were not detected by GPU. GPU's supervisor of Licensed Operator Training was aware of "no indi-cations" of cheating during the mock exams._8/ It is apparent that a reasonably diligent reading of the mock exams would have disclosed the cheating.

In addition, the same two persons were involved in a cheati ,g incident in 19 79, in which one of them submitted answers to make-up questions actually written by the other ._9/ While the cheater (but not his cohort) was apparently suspended from licensed acti-vities for some period of time, the GPU investigator remarkably i concluded that he was "a person of unquestioned integrity ," guilty of no " malfeasance." 10/ obviously , the cheater was reinstated and both were allowed to sit for the SRO exam - a position of substan-tial responsibility.

4. A;t least one of the interviewees ref used "for unknown reasons " to answer questions about the possibility of crib sheets being brought into the exam room. While he admitted to hearing rumors of cheating, he refused to reveal specifics or identify individu als . ll/ I&E never pursued this matter , ,

Even though severely limited, the I&E investigation disclosed sufficient f acts to compel retesting of the TMI-l operators. It cannot be disputed that the tests were administered in a manner which permitted and even encouraged cheating - in this case, cheat-ing of a particularly unsubtle sort. The two persons caught were almos t incredibly s tupid. Had they taken minimal care to avoid S/ Id. at 48, 2. .

8/ Id. at 19.

9

/ Id. at 7.

[

10 / Id. at 8.

11/ Id. at 39.

i l

t l

, c- _ . _ _ , _ _ - , , , _ ,-

e s .-

HARMON & WEISS

~

Page 4 word-for-word similarities in their answers, the cheating would never have been detected, despite the passing of papers and verbal consultation between the two persons. There is simply no basis to believe, given the lax and at times nonexistent supervision of the exams, that the results are reliable.

While retesting may be unf air to those who honestly passed, it should pose no great hardship in comparison with the possible i consequences of allowing unqualified people to fill RO and SRO positions. If the candidates were qualified just a f ew months ag o , they should be able to pass the exam again without difficulty.

~

(If not, one must question the correlation between successful performance on the exam and an operator's competence to perform his duties ) .

In conclusion, UCS urges you to require retesting of the TMI-l op erators . The Commission cannot place confidence in the results of the April 21-24 exams.

Very truly yours,

' 1 l

Ellyn . Weiss Counsel for the Union of Concerned Scientists ERW/dmw cc: TMI Service List i

- , . - - - - - - - - ,. . - -