OPINION

COOLAHAN, Senior District Judge.

Rarely does a case involve conflicting interests as important and as difficult to reconcile as those in this litigation. The psychological testing which Jersey City uses to screen applicants for its fire department has been challenged by plaintiffs as an invasion of the applicants' constitutional rights. In plaintiffs' view, conditioning employment on psychological testing of questionable validity puts the applicants to a prohibited choice of sacrificing constitutional freedoms to secure prized employment or of looking for work elsewhere. Defendants contend that the condition is a reasonable and necessary one because the task of fighting fires is no ordinary job in difficulty or importance and because success depends critically on the psychological capabilities of firemen.

There is good reason to scrutinize a government requirement which joins the words psychology and testing. Psychology is not yet the science that medicine is and tests are too frequently used like talismanic formulas. The Court has, therefore, carefully reviewed the extensive evidence generated by a long trial and has not arrived lightly at the conclusion that the defendants' psychological testing is constitutional.

PART I

A. Procedural History and Preliminary Rulings

Plaintiffs' civil rights complaint, brought pursuant to 42 U.S.C. § 1983, seeks relief from actions, by persons acting under color of State law or authority, that allegedly subject them to deprivation of their constitutional rights. Accordingly, plaintiffs complain that the psychological evaluation required by Jersey City as a precondition to employment with the Jersey City Fire Department forces them to forego several constitutionally protected freedoms. Plaintiffs request a declaratory judgment pursuant to 28 U.S.C. § 2201 (Supp. 1977), injunctive relief, and an award of monetary damages. (Amended Complaint.) Jurisdiction is vested in the Court by 28 U.S.C. § 1331, 1343.

This protracted litigation began on April 17, 1974, when plaintiff James McKenna applied to the Court for a temporary restraining order. The Court's dismissal of the complaint for lack of jurisdiction was reversed on appeal. McKenna v. Fargo, 510 F.2d 1179 (3d Cir. 1975). An amended complaint was filed on March 7, 1975, naming five plaintiffs and three defendants. Three of the plaintiffs, James McKenna, Robert McKenna, and Brian Flaherty, have gone through the contested psychological evaluation and hiring procedure. Plaintiffs Joseph Porter and John Cinciarelli applied for employment with the Jersey City Fire Department and were certified as eligible by the Civil Service, but have not taken the physical and psychological examinations. The defendants are Nicholas Fargo, the Jersey City official responsible for appointment and supervision of Jersey City fire fighters, Raymond Gibney, Chief of the Jersey City Fire Department, and Stevens Institute of Technology, a New Jersey corporation under contract to Jersey City since 1966 to conduct psychological evaluations of fire-fighter candidates. (T. 1.09-.11) Defendant Stevens operates the Laboratory of Psychological Studies (LPS), which conducted the testing. Plaintiffs represent themselves only; the Court denied plaintiffs' motion for class action certification for reasons stated in an unpublished opinion filed February 23, 1976. A motion for partial summary judgment in favor of plaintiffs was also denied by letter opinion filed on June 20, 1977.

Trial commenced on June 28, 1977, and ended July 28, 1977, generating 16 volumes of transcript. Proposed findings of fact and conclusions of law were received from all parties by mid-February 1978.

During trial, the Court reserved decision on defendants' motion to dismiss the claims of plaintiffs Porter and Cinciarelli on two grounds. Defendants argued, first, that the case was moot as to these plaintiffs since defendants no longer required psychological testing by defendant Stevens, and, second, that these plaintiffs lacked standing to sue since the Civil Service list pursuant to which they had qualified expired several months before commencement of trial. The Court has concluded that the suit is not moot and that plaintiffs Porter and Cinciarelli have standing to sue even though they have not undergone testing by defendant Stevens.

The evidence indicates that Jersey City intends to continue psychological evaluations and that Stevens withdrew its participation only because of this suit. By letter dated November 10, 1976, Stevens asked Jersey City for a hold-harmless agreement to protect it from any liability that might result from its actions as an agent of Jersey City. (DJ-17) The reply from Jersey City to Stevens stated that its legal department recommended that the agreement not be signed "while a Law Suit is pending" and asked, "Can we proceed as we did in the past without the signed agreement." (DJ-18) Stevens refused to continue doing psychological evaluations without the hold-harmless letter. Since Jersey City still required evaluations as a precondition of employment, defendant Fargo directed that another psychological consultant be located. (DJ-12) By letter of January 31, 1977, Jersey City retained the services of a psychiatric institute for psychological evaluation of fire-fighter candidates. (DJ-13) The Court understands from counsel that this arrangement is ongoing and may involve some of the same procedures used by Stevens. (T. 11.5-.16)

As for the expired Civil Service list, counsel for plaintiffs Porter and Cinciarelli indicated on the first day of trial that they would submit applications to qualify for the next Civil Service examination. (T. 1.5-.6) They appear to have done so. (T. 9.11, 2.113-.114) The Court agrees that they are likely to qualify again since they did pass the test the first time. (P. Post-Trial Br. at 2)

It is apparent that defendants' allegedly unconstitutional conduct has been suspended only in the sense that defendant Stevens has temporarily withdrawn because it cannot now obtain a hold-harmless letter. Clearly Jersey City intends to continue psychological testing. If assured of no liability by a judgment from the Court or a hold-harmless agreement, Stevens would in all probability be willing to resume its role in the testing. Such voluntary cessation of allegedly illegal conduct does not make this case moot. See Defunis v. Odegaard, 416 U.S. 312, 318, 94 S.Ct. 1704, 40 L.Ed.2d 164 (1974). In Defunis the Supreme Court described a line of decisions as standing for the proposition that unilateral termination by the defendant of the challenged conduct cannot be allowed to make a case moot, for to do so would mean that a defendant could evade review by voluntarily ending the conduct whenever sued, thus defeating the public interest in having the legality of the practices determined. For a case to be moot, a court must be able to conclude that there is a reasonable expectation that the wrong will not be repeated. Id. at 318, 94 S.Ct. 1704. In this case, repetition of the challenged governmental action is certain, based as it is on a definite and continuing policy. See Super Tire Engineering Co. v. McCorkle, 416 U.S. 115, 122-125, 94 S.Ct. 1694, 40 L.Ed.2d 1 (1974).

Plaintiffs Porter and Cinciarelli also have standing to sue. Standing to sue exists when the plaintiff himself has suffered some threatened or actual injury resulting from putatively illegal action. See Warth v. Seldin, 422 U.S. 490, 499, 95 S.Ct. 2197, 45 L.Ed.2d 343 (1975). Porter and Cinciarelli were certified as eligible for appointment to the fire-fighter force when suit was brought and are likely to qualify again in the near future. A psychological evaluation was and still is a precondition for employment. The threat of injury to these plaintiffs is not a bare allegation derived from an indirect and improbable chain of events; nor are these plaintiffs' claims for prospective relief based on a threat lacking immediacy or ripeness. Cf. id. at 516, 95 S.Ct. 2197; Roe v. Wade, 410 U.S. 113, 128, 93 S.Ct. 705, 35 L.Ed.2d 147 (1973). Thus these plaintiffs have a personal stake in the outcome of this controversy sufficient to confer standing.

One other preliminary matter requires consideration. Defendant Stevens has argued that plaintiff Flaherty's claim must be dismissed for failure to exhaust State administrative remedies. While, as plaintiffs contend, Flaherty's aborted appeal may well have satisfied any exhaustion requirement, Stevens' argument is, in any event, misconceived. Exhaustion of State administrative remedies is not required prior to commencement of an action under 42 U.S.C. § 1983 in federal court. Hochman v. Board of Ed., 534 F.2d 1094 (3d Cir. 1976). There is, therefore, no bar to consideration of the claims raised by plaintiff Flaherty or any of the other plaintiffs.

B. Elements of the LPS Evaluation

Several personality assessment instruments are involved in this suit. To understand the background and operation of defendants' evaluation and hiring procedure, a brief description of the types of instruments used is necessary. In Part II, how the tests were utilized in the evaluations done by defendant Stevens will be considered.

Self-Report Inventories

1. The Minnesota Multiphasic Personality Inventory (MMPI) consists of approximately 550 numbered statements in booklet form with printed instructions on the cover. (P-3, T. 7.41) An individual taking the inventory reads the instructions and proceeds through the test without being observed or questioned by a psychologist. The MMPI is thus described as a self-administered, self-report inventory. (T. 12.87) The individual is instructed to read each statement and to try to decide whether it is true or mostly true or false, or not usually true, as applied to him; items which the individual finds not applicable or on which he has no judgment are left blank. Answers are recorded on a separate answer sheet by marking in true or false columns. The statements in the inventory range over several areas and refer to opinions, attitudes, observable behavior, and feelings which the subject may find applicable to himself. Answers to certain groups of questions are collected according to ten basic scales, each scale representing a personality characteristic, such as paranoia or depression. Scale numbers are listed on a graph or profile for comparison to other subject groups. (DS-1(a))

2. The Edwards Personal Preference Schedule ("Edwards test" or EPPS) is similar in form to the MMPI. Like the MMPI, it is a self-administered, self-report inventory. The Edwards test consists of about 225 pairs of statements in booklet form. The individual is instructed to indicate which of the two statements is more characteristic of what the individual feels or likes. (T. 12.-99, P-4) The preferences are combined according to 15 scales which reflect "need" categories. The scales are entered on a graph as a profile of personality. (T. 12.-100-.101, P-4(a))

Projective Instruments

Four of the instruments are projective tests. Projective tests use a vague stimulus which the individual is instructed to explain; the subject thus projects on to it fantasies and emotional associations. (T. 7.120, 12.103-.104, 12.115) In contrast, objective instruments, like the self-report inventories, use a single, structured stimulus and limit the subject's response to specific experiences or feelings.

1. The Rorschach test consists of a set series of pictures of inkblots, usually ten in number. The cards range in color from all black and gray to others having several pastel colors. (P-7) The irregular form of the inkblots permits innumerable interpretations, and the vague shapes are roughly suggestive of things ranging from animals to sexual organs. The responses are analyzed in several ways, including, for example, by content, parts of the blot used, or perception of movement. (T. 12.117) Interpretation of the responses provides information about emotional and personality traits.

2. The Thematic Apperception Test (TAT) requires a subject to interpret a picture by telling or writing a dramatic story about what has led up to the event in the picture, what is happening, and what the possible outcome might be. (T. 12.104) The responses are interpreted primarily by analysis of any recurring themes behind the plots, and the way in which the subject uses aspects of the picture to form the story. From a total set of 20 pictures, fewer cards may be used, although different pictures are normally used for men and women.

3. In the Draw-a-Person or Human-Figure Drawing Test, the subject is instructed to draw a person on a blank sheet of paper, and then to draw a figure of the opposite sex on a second sheet. (T. 6.68, 12.128) The drawings can be analyzed according to size, location on page, and quality or amount of detail. (T. 14.74, 6.94, 6.68) The test is usually used for personality assessment, including sexual identity, but also for intelligence measurement. (T. 6.94, 7.25)

4. The Incomplete Sentence Test involved in this litigation was developed by Stevens as a modification of similar instruments. (T. 14.74) The test is a type of projective test, though more structured than the TAT or the Rorschach. A group of 81 sentence stems are presented to the individual, who is asked to finish the sentence with the first idea that comes to mind. (P-6, T. 12.124-.125) Responses are written onto the form by the subject, and it is self-administered, much like the MMPI and the Edwards. (T. 12.125) At LPS the responses were interpreted by the psychologist who administered the Rorschach. (T. 12.125)

Interview and Personal Data Form

1. On the Personal Data Form used by Stevens, the applicant stated his education, employment and military background. The form also contained several questions about the applicant's family history. (P-8)

2. Applicants were given a Patterned Interview by a staff psychologist, who relied in part on the Personal Data Form. (T. 12.130) The purposes of the interview were to inquire as to the applicants' motivations in seeking the position, or whether they or their spouses had reservations about the dangers of the job, and to begin to form a hypothesis of the applicants' personality that could be corroborated by the other testing data. (T. 14.75-.76; 12.129)

PART II

A. Evaluations and Hiring before 1972

The testing or assessment procedure conducted by Stevens for Jersey City began in the spring of 1966. Before then, applicants for both the police and fire departments were given psychiatric interviews by a doctor on the staff of the Department of Police and Fire, who was paid on a fee-for-case basis. (T. 11.50)

More thorough psychological assessment of applicants for the police and fire departments came about because of the civil unrest and riots in Jersey City in 1965 and as a result of the experience Stevens had gained in psychological evaluations done for smaller communities. At that time, and currently, the director of Jersey City's Division of Medical Services was Dr. Patrick McGovern. According to his testimony, Jersey City officials were concerned over the responses of police and firemen to the riots and in particular to attacks directed at them; they "were constantly looking for ways to pick people who would not overreact in situations." (T. 11.50) Dr. McGovern read about a police screening project conducted by Stevens and inquired whether a similar program for firemen could be devised for regular use by a municipality of Jersey City's size. (DJ-2, T. 11.50)

The deputy director of Stevens' Laboratory for Psychological Services (LPS), Dr. Myron Johnson, explained in a letter to Dr. McGovern that Stevens had conducted psychological screening programs for police applicants in three small communities and for the Stevens campus itself. (DJ-3) Stevens had also begun a psychological assessment program for firemen in the town of West New York only several months earlier. The purpose of the West New York project was to detect emotional flaws in applicants that could lead to failures in performance or poor decisions at critical moments and to check for psychosis or imminent emotional breakdown. (T. 9.21-9.-29) Dr. Johnson's review of the available studies and research data in connection with the West New York project indicated there was no relevant, conclusive study on how such testing or assessment should be designed. (T. 9.25-.26)

Dr. McGovern arranged a meeting on March 15, 1966, of Jersey City police, fire, and budget officials to discuss with Dr. Johnson the feasibility of adopting a testing program similar to that used for firemen in West New York and for policemen in several small communities. It was agreed that Stevens would conduct a two-step process of psychological screening for both police and firemen. All applicants would be given the MMPI in conjunction with the physical examination and the other Civil Service tests for intelligence and aptitude. (DJ-1) Only applicants whose MMPI profiles raised doubts as to their suitability would be given a full battery of tests and a psychiatric interview. (T. 12.79) The general qualities or characteristics Stevens would use as a basis for judgment were discussed and agreed upon. (T. 11.54)

In January 1971 Stevens recommended that Jersey City adopt full evaluations for all applicants. (DJ-5) The decision made in March 1966 to give full examinations only to applicants with poor MMPI profiles seems to have been a compromise caused primarily by the limits of Jersey City's budget and the unwillingness of the Civil Service to finance psychological evaluations for all municipalities. (DJ-4(a), DJ-5) Stevens preferred the use of several techniques from the outset, but the municipalities that were interested lacked funding for more than very limited testing. (T. 9.40) With more experience — there having been few authoritative studies to rely upon initially — Stevens prevailed on Jersey City to adopt full assessment for all applicants. (T. 9.40-.41, DJ-5) By 1972, when plaintiffs were evaluated, Jersey City had replaced the two-step MMPI screening with full evaluations for all applicants.

B. Evaluation and Hiring Procedure

The procedure for employment with the Jersey City Fire Department consisted of several stages. The psychological evaluation came at the end of the process and thus only applied to candidates who passed the initial stages. All applicants were required to take the Civil Service examination, which included a physical agility test. If they were certified by the Civil Service Commission, they were given a medical examination at Jersey City Medical Center. After April 1972 the MMPI was given at LPS with all the other tests. (T. 12.89) An investigation of the applicant's background was also made. Applicants who had been certified as eligible and whose medical and background checks were satisfactory were notified by Jersey City to report to Stevens for psychological examination. The eligible applicants were taken from the top of the Civil Service list until the requisite numbers were hired to fill the vacancies.

The testing at Stevens was conducted in two settings. In a group testing room not exclusively used for fire department applicants, candidates were administered the self-report inventories, that is, the MMPI and the EPPS, and semi-structured projective tests, that is, the TAT, Draw-a-Person, and Incomplete Sentences Test. The tests were proctored by a psychiatric technician, although not always continuously. (T. 1.138, 2.49) None of the tests was subject to time restrictions. (T. 1.11)

Applicants were also given an interview with a Stevens psychologist at some point during the day. After the interview, the Rorschach inkblot test was administered by the interviewing psychologist. (T. 1.11) Applicants taking the so-called pencil-and-paper tests in the group testing room would be interrupted, if necessary, for the interview and Rorschach test, when a psychologist became available. (T. 1.11)

From all the data, an evaluation was prepared by the interviewing psychologist. (T. 1.8, 12.141) The following psychological characteristics were used as a basis for evaluating fire-fighter candidates (T. 1.12):

1. Be able to adjust well to close community living;

2. Be able to follow orders explicitly;

3. Be able to withstand substantial stress and tension as generated by life-endangering circumstances encountered in fire-fighting situations, where circumstances beyond the control of the individual are operative;

4. Be able to make decisions under stress;

5. Be able to take calculated but not any unnecessary risks;

6. Be free of abnormal fears related to fire-fighting duties such as fear of heights or enclosed spaces.

A short evaluation report was prepared which summarized Stevens' conclusions and stated a recommendation of accept, reject, watch during probation, or refer for psychiatric help. (T. 1.8, 11.69) After the evaluation was reviewed by the supervising psychologist at Stevens (T. 12.146), the report was forwarded to Jersey City. None of the testing data or interview notes was forwarded to Jersey City. (T. 11.70) Stevens' charge for the evaluation on January 1, 1975, was $105. (T. 11.66)

It was Jersey City's policy to refer all candidates receiving negative evaluations from Stevens to a psychiatrist for a second psychiatric interview. In all cases Jersey City bore the cost of the referral. (T. 11.70) The referral interview was a supplement to the assessment program. Jersey City would forward the Stevens report to the second psychiatrist, who used it for background. (T. 11.79-.80) Jersey City did not, as a matter of policy, follow the Stevens recommendation, but appears to have done so in most cases. (T. 11.70, 11.80)

Candidates who were rejected after the referral interview were notified by the Civil Service Commission of their right to file within ten days an intent to appeal the decision of Jersey City to remove their names from the Civil Service list. (T. 12.12) From 1971 to 1973 the appeals process consisted of a preliminary review and decision by the Local Government Services Division or the Examinations Division of the Department of Civil Service (T. 15.141), followed by an appeal to the Civil Service Commission. The preliminary decision was usually based only on the official documents and any papers that the rejected applicant wished to submit. If further appeal was taken to the Commission, a plenary hearing would be held to resolve serious disputes of fact or complex issues. The hearing officer was a lawyer without expertise in medical matters. (T. 15.141-.142)

Since 1973 a Medical Review Board has been in existence to review medical or psychological issues. The Board consists of a psychiatrist and a clinical psychologist, both members of the faculty of the College of Medicine and Dentistry of Rutgers University, and a Civil Service representative. (T. 12.4) Since 1974 the regulations governing appeals have required submission to the Board of all information pertaining to the applicant, including raw testing data, psychiatric evaluations, and a report of any interviews. (T. 12.7) Refusal or inability to supply the required information on the part of the local authority results in denial of the request to remove the applicant's name from eligibility. (T. 15.144)

When the Medical Review Board receives all the information from the local appointing authority, a copy is provided the applicant and an opportunity is given the applicant to submit a rebuttal report. A hearing before the Board is then held, during which the applicant may appear with counsel and may present witnesses, including doctors. After the hearing, the Board makes a recommendation to the Commission. (T. 12.-8-.9) Exceptions to the report may be made by both parties within 15 days. The Commission thereafter makes the final decision as to whether the candidate's name shall be retained on the hearing list in line for possible appointment. (T. 12.9-.11)

Personal information generated by the evaluation and hiring process was retained by both Stevens and Jersey City. Since the beginning of the LPS evaluations, LPS retained all testing data. (T. 9.43) Only the written summary and recommendation for each applicant, of about one or two pages in length, were forwarded to Jersey City. (T. 12.148, PSJ 1-3) The raw data retained at LPS was stored in a locked file on the eighth floor of the Stevens Center. Only Dr. Springob and Dr. Gottesman, directors of LPS, and research assistants under their supervision had access to the files. (T. 12.-148-.149) The files at Stevens were marked with the applicants' names and an identification number. However, if evaluation data was used for research, it was abstracted from the file, coded, and given an identification number. (P.Ex. 1b, T. 12.149) Stevens retained the raw data primarily because of the applicants' right to appeal (T. 12.149) Defendant Stevens also indicated that there was a possibility that the data might be used in a study to improve the psychological testing norms for firemen, as had been done for policemen, or in a validation project if funding became available. (T. 14.127-.128, 15.5-.6)

The negative evaluations received from LPS by Jersey City were retained by the municipality in a locked cabinet in the office of Dr. McGovern, the director of the Division of Medical Services. Dr. McGovern, his principal assistant, and specific firemen under their supervision had access to the cabinet. (T. 11.64) Although the testimony was not clear, it appears that a copy of the report by the psychiatrist who interviewed applicants after a negative LPS evaluation was sent to the Fire Department Director and possibly to the Director of Personnel for Jersey City. The psychiatrist's report received by the Fire Department Director was kept in a locked file in his office with access limited to him and his secretary. (T. 11.126, 11.78) The Director of Personnel most probably received only the list of names of rejected applicants, but the testimony leaves some doubt on the point. (T. 11.78) There is no evidence that Stevens had written, formal guidelines controlling access to the data, or that Jersey City adopted any such regulations or formal policy. Also, there is no evidence that applicants were notified prior to the testing of possible use of the data in research or validation studies. Initially, rejected applicants were not permitted by Jersey City to see the reports of their psychiatric examinations because the psychiatrists conducting the examinations believed that disclosure would be injurious to the applicants' mental health. The reports were stamped with a warning against disclosure. The Civil Service, however, notified Jersey City that on appeal applicants would be entitled to access and that unless the reports were released, they would not be grounds for removal. Thereafter such reports were stamped "Privileged and Confidential" and were released if requested. (T. 11.64-.65, DJ-7, DJ-8)

C. Plaintiffs' Evaluations

1. Robert McKenna

Plaintiff, pro se, Robert McKenna, was given a psychological evaluation in May and June 1972, and was appointed to the fire department in July 1972. In 1973 he received his under-graduate degree and entered law school in September 1973 while also working as a fireman. Plaintiff was married in 1967 and had two children at the time he was appointed to the fire department.

Plaintiff testified that he came from a "fire department family" and knew that psychological testing was part of the hiring process; from a college psychology course he had also become familiar with certain of the testing instruments. (T. 2.46, 2.100) In April or May 1972, at the Jersey City Medical Center, plaintiff was given the MMPI along with several other applicants. Firemen assigned to duty at the Medical Center distributed the MMPI, read the instructions, and had the group read along. The proctor checked on the group but was not there at all times. (T. 2.49) The remainder of the testing and the interview were conducted at LPS.

According to his testimony, as he went through the MMPI, two questions disturbed him. (T. 2.49, 2.53) The first question, "Do you believe in God?", frightened him because he believed that the fire department would not hire him if he answered "no" and because it was the type of question he could not prepare for like a physical or written examination. (T. 2.57) The second objectionable question was, "Was your father a good man?" This question caused a "fearful reaction" because plaintiff could not honestly give a positive answer, which he believed was the expected one. He also thought that it might lead to official inquiries into his childhood and, consequently, disqualification. (T. 2.59-.60)

During his interview later at LPS, he was asked by the interviewing psychologist to clarify evasive information given about his father's occupation on the Personal Data form. Plaintiff stated that the interviewer's persistent questions forced from him memories about his father that were his "earliest recollections" and which he had not revealed to anyone. (T. 2.67) According to his testimony, after he had recounted an episode involving his father's embarrassing or violent behavior toward him or his mother, the interviewer would follow up each recollection with more questions. (T. 2.67-.72) Several questions were asked about his parents' divorce and his feelings about his mother's remarriage, and about his relationship with his sister. The interview lasted about 30 to 60 minutes. (T. 2.98)

2. James McKenna

Plaintiff James McKenna went through the assessment process in May 1972 and again in the spring of 1974. (T. 1.15, 1.80) For reasons unrelated to this litigation, his name was removed from the Civil Service list in July 1972 and, after appeal, was reinstated in 1974. (T. 1.16) He was given satisfactory evaluations on both the 1972 and 1974 examinations.

At the first examination in 1972, James McKenna was 25 years old and had completed two years of college education. (T. 1.13-.14) He testified that he had an idea of what the testing would be like from other applicants and because he had taken psychology courses in college. Before the testing, plaintiff also read about the MMPI and the Rorschach. (T. 1.24)

He recalled that the testing procedure was essentially the same in 1972 and 1974, except that the MMPI was administered in 1972 at the Jersey City Medical Center and that in 1974 all the tests were given at LPS. (T. 1.116) The tests given in the group setting were administered and supervised by a proctor and were given in an orderly manner. (T. 1.91-.92)

His objections to the assessment procedure were primarily directed at the two interviews and at the administration of the Rorschach test. He was not able to recall each interview separately and therefore gave a general account of both together. (T. 1.46) Discounting for some exaggeration, it appears that during his interviews he was asked a number of questions about his family and personal life. Specifically, he recalled that the interviewer asked about "socializing habits and drinking habits of my family." (T. 1.41) Plaintiff recalled that when he stated that his father drank heavily, the interviewer followed with probing questions about his relationship with his father. He stated that he believed that to have a chance to qualify he had to answer all questions, but admitted that he did not feel "singled out" from other applicants. (T. 1.67) The Stevens evaluation and recommendation were positive. In November 1976 he was appointed to the Jersey City Fire Department for the training and probationary program, and in February 1977 received his full appointment.

3. Brian Flaherty

Of the plaintiffs who were given psychological evaluations, Brian Flaherty is the only one who was not hired. In May 1972, when he was evaluated, he was 27 years old, had completed several semesters of college, and had served in the Army in Vietnam. During 21 months of infantry duty from February 1969 to November 1970, plaintiff saw some combat duty and worked as a public information officer. He was discharged under a 90-day early civilian work release program. (T. 1.126-.130, 2.8)

Flaherty was given the MMPI at the Jersey City Medical Center and the other tests at LPS. In his testimony, Flaherty stated that he was disturbed by the TAT and the interview. He stated that he had difficulty writing the total of 1,500 words for the TAT and had had difficulty with English since high school and also during his Army work as a public relations officer. (T. 2.6-.9)

Plaintiff's more strenuous objections were addressed to the interview. He stated that he was asked general questions about his parents' relationship and any problems they had. (T. 1.143-.144) Several questions were asked concerning any drinking problems in the home or other problems that affected his relationship with his parents. (T. 2.10)

During the interview plaintiff was also asked if he was dating anyone and whether he thought it was a "normal romance." (T. 1.144) Plaintiff testified that he was asked whether the relationship involved premarital sex, whether he planned to marry, and the ethics of such a relationship. Plaintiff objected to the questions and "short-answered the interviewer." (T. 2.11) When asked about childhood bedwetting, plaintiff found the question "comical at first" but was disturbed by it later when he could not understand its relevance. (T. 1.148-.149) The interview lasted approximately 45 minutes. (T. 2.30)

Shortly after the LPS evaluation, plaintiff was notified by Jersey City that the evaluation had been negative and a referral interview had been arranged. (T. 1.150, 2.32) The second interview was conducted by a psychiatrist and lasted about 20 minutes. (T. 1.151, 2.34) Plaintiff was questioned along the same lines as the prior interview at LPS. (T. 2.34)

Subsequently, plaintiff Flaherty received a notification from Civil Service that proceedings would be instituted to remove his name from the eligible list and that he had ten days in which to begin an appeal. (P-38) Plaintiff contacted Civil Service, received general information about his evaluation file, and agreed either to submit a favorable psychological evaluation within two weeks or to have his appeal considered as withdrawn. (P-38) Plaintiff testified that he was financially unable to obtain an independent evaluation. (T. 2.21) After extending the time for submission of the report, Civil Service removed plaintiff's name from the eligible list.

PART III

At trial, testimony and exhibits were received with respect to whether the design of the evaluation process was appropriate, whether the LPS recommendation was properly formulated, and how accurate the ultimate judgment was or could be proven to be. The findings of the Court as to each area will be discussed, but the significance of particular findings will be examined in relation to the legal standard in Part IV.

A. Psychological Factors in Job Performance

Defendants contend that the job of fire fighting has psychological dimensions which are found in few other professions. The testimony of defendant Fargo, now Deputy Chief of the Fire Department, bears out their contention.

Fire fighting is a life-endangering occupation. Chief Fargo stated that in his career he had seen 25 to 30 deaths. (T. 11.-133) Fighting a fire places firemen in situations beyond their control and frequently beyond their expectations. Unlike simulated exercises in the training program, real fires are often neither controlled nor controllable. The unknown quality of the danger and the threat to life occasion emotional demands which can cause errors at crucial moments.

From his experience, Chief Fargo recalled the impact which emotional stress could have on firemen. He stated that he had on several instances seen a fireman freeze halfway up a ladder while going to rescue trapped victims, forcing other firemen to go up the ladder and around him to make the rescue or, failing that, leave other functions aside and raise a second ladder. (T. 11.-120-.122) Similarly, Chief Fargo recalled firemen refusing orders or failing to carry them out until threatened by him. (T. 11.-122)

Overcoming the impulse of panic and being free of abnormal fears of heights and enclosed spaces is essential to survival. Chief Fargo recalled an episode in which a fireman was trapped in a shower stall because of zero visibility and escaped only after three attempts to find the open side. (T. 11.131) Failure in maintaining emotional composure had on occasion caused firemen to lose sufficient mechanical ability even to turn on hydrants.

The nature of firemen's lives in the fire station also makes special emotional demands on them not found in the working day of most people. Chief Fargo described life in a fire station with three or four other firemen as an intimate situation much like a close family but without the privacy of a home. He stated the emotional capacity to adjust to community living with close contact for long hours without becoming antisocial was essential. (T. 11.129)

The criteria used by LPS accurately reflect the emotional capacity and personality traits necessary for successful firemen. Chief Fargo specifically reviewed and approved each of the LPS criteria. (T. 11.-128-.132) Moreover, he was unable to assign to the criteria degrees of importance, finding them "inexplicably interwoven." (T. 11.124) Thus, Chief Fargo's characterization of fire fighting as a life-endangering profession in which psychological elements play a crucial role is entirely credible. (T. 11.119, .132)

B. Appropriate Design

The parties have disputed how the testing process should be labeled and, more important, how its purposes should be categorized. The source of the difficulty apparently is that tests are better suited and more frequently used for assessment of the cognitive and attitudinal qualities involved in job performance than for assessment of job-related psychological or emotional traits. (T. 4.78-.79) In addition, the study of individual personality and psychology not primarily for purposes of job performance is a distinct field of psychology. However, it is this area of psychology, clinical psychology, which is central to the LPS evaluation process.

Clinical psychologists in general are concerned with understanding human behavior and the processes which affect how people behave in a variety of life situations, sometimes, but not primarily, including the work environment. They are trained in the diagnoses of symptoms of poor mental health or emotional disturbances, and for intervention in behavior to provide treatment or therapy in order to aid the individual in being more effective in his activities. (T. 4.145) Industrial psychologists, on the other hand, are largely concerned with the world of work. Their primary concern is to understand what kinds of factors influence the way people behave on the job and to develop ways of increasing an employee's and an organization's effectiveness in terms of work performance. Thus, while industrial psychologists focus on such things as job recruiting, selection, and placement, clinical psychologists are concerned with problems such as the evaluation of psychological states, classifications of mental health, and effective treatment techniques. (T. 3.80-.81, 4.145-.146)

Defendants contend that the LPS testing process, although used ultimately for job selection, was based on valid clinical evaluation procedures. Defendants and defendants' experts have characterized the testing process as a psychological evaluation of emotional or affective traits or characteristics of individual applicants done on a noncompetitive and noncomparative basis. (T. 7.101, 14.42, 14.97-.98) The evidence supports defendants' contentions.

The testing instruments which were used by LPS, and their joint use in a battery, with a patterned interview, are standard techniques and methodology used by clinical psychologists in formulating a diagnosis of an individual's personality. (T. 9.195, 10.-66-.67) Although all of the testing instruments, except for the Rorschach and the patterned interview, were administered in a group setting, the LPS procedure was intended to produce an individualized personality profile for each applicant. The testing instruments which LPS administered in a group setting were, in fact, designed for group administration, and the resulting data, when interpreted by psychologists trained in clinical methodology, could be used in forming an assessment of individual applicants. (T. 9.172-.188, 7.101-.103) The data was analyzed by psychologists experienced in clinical evaluations, who prepared a report and recommendation for each applicant in light of the evaluation and the psychological criteria. (T. 12.76, .139-.148) No attempt was made to weigh the psychological merits of each applicant against those of the other candidates in the group; the LPS evaluation was not a "test" in the sense that most tests, like the Civil Service eligibility test, are designed to produce a ranking of applicants by order of highest score. (T. 14.98) Thus, it is fair to conclude that the LPS evaluation was designed to diagnose and evaluate the suitability of applicants on an individual and noncompetitive basis by adjusting the methodology of clinical psychology to the task.

However, the ultimate use of the evaluation was to screen out applicants whose emotional traits may lead to poor job performance. In that sense, the assessment process resembles industrial screening. Accordingly, plaintiffs contend that an individualized clinical assessment is being used as a basis for job prediction. Plaintiffs urge that the testimony of their experts indicates that such a procedure is unusual and inappropriate because clinical techniques and evaluations were not designed for use in industrial settings. (T. 3.83-.84, 10.132, 10.81)

Defendants concede that the use of clinical instruments in an industrial context is unusual but argue that unique aspects of the job of fire fighting require measurement of potentials for behavior not found in most jobs and that clinical instruments are the only methods available for evoking responses that would be suggestive of what the individual's potential behavior might be. (T. 14.103-.106) In the assessment process, the LPS psychologists sought information on "behavioral potentials," that is, probabilities of how an applicant might react because of his psychological makeup in situations unique to fighting fires and living in the close quarters of a fire station. The behavioral potentials to which LPS referred were stipulated by the parties. (T. 1.12) The behavioral potentials are broadly phrased and relate primarily to the emotional problems of facing life-endangering fires, functioning in dangerous situations beyond the control of an individual, and adjusting to community living.

Defendants distinguish between behavioral potentials and criteria measurements. A criterion is a valued behavior or outcome such as successful job performance or school achievement; generally, criteria are defined so as to be quantifiable and subject to measurement, as, for example, by a grade or rating. (T. 10.5-.6) Behavioral potentials, on the other hand, as defined by defendant Stevens, are an estimate of the kinds and degrees of emotionally trying situations which firemen may face and a judgment as to an applicant's probable reaction. Defendants contend that the emotional dimensions of being a fire fighter, such as the degrees of stress caused by various kinds of fires, cannot be quantified and that the LPS psychologist does not know exactly what situations will be faced by an applicant. (T. 12.83-.84) While defendants were not entirely clear on the point, they seem to argue for a fundamental difference between the two terms. While a criterion is a quantification of a measurable outcome, a behavioral potential is a judgmental estimate of the probability of future behavior. Without deciding whether defendants' assertion is correct that the psychological aspects of the job of fire fighting cannot be reduced to quantifiable terms, the Court believes that the design of the LPS assessment is to arrive at a psychological profile of the applicant and to make a judgment as to whether the applicant's emotional makeup will permit or prevent desirable behavior for a fireman.

There was substantial testimony from defendants' experts that the clinical instruments and tests used by LPS could be helpful in making a judgment as to the behavioral potentials. (T. 7.68, 7.121-.122, 12.-142-.146, 14.83-.86) Dr. Dahlstrom specifically reviewed each of the behavioral potentials and stated how testing for personality characteristics was relevant and which instruments would be useful; he also reviewed prior studies using similar techniques, particularly in military situations. (T. 7.68-.73, 7.116-.122) In his expert report, Dr. Dahlstrom reviewed the possible ways of gathering data for emotional screening and concluded that individual clinical assessments by experienced professional psychologists was the most suitable and appropriate method for the kinds of evaluations and screening LPS was asked to do by Jersey City. (DJ-24A) Plaintiffs' expert, Dr. Lapidus, also testified on cross-examination that in research she had done on predicting mastery of stress in life-endangering situations, she had used some of the self-report inventories and projective instruments used by LPS. (T. 5.102, 9.217-.218) In sum, it appears that the design of the LPS evaluation is based on reputable professional psychology methodology even though its use in this context is disputed within the profession. The use of methods of clinical psychology to determine the psychological makeup of applicants is unusual in a job-screening situation, albeit one that is noncompetitive and noncomparative. However, the large and perhaps crucial part that psychological factors play in the job of fire fighting certainly establishes the relevance of a psychological evaluation. That few other jobs involve the dangers of fire fighting or the necessity for close community living helps explain why psychological evaluations are rarely used or given by industrial psychologists. The use of clinical techniques does not appear to be a fundamentally unsound premise for selection of firemen or conceptually without support in psychological principles. It is, therefore, necessary to consider the evidence as it pertains to the way the design was carried out and to proof of its accuracy.

C. Testing Data

The LPS recommendation was based on the data collected from the battery of clinical instruments and on the evaluation of the data by LPS psychologists. In order to understand how the recommendation was made, it is necessary to consider in detail the kind of psychological information gathered by each instrument.

1. MMPI

The answer to any particular question of the approximately 550 on the MMPI is not significant in itself. Answers become significant only when grouped according to personality scales and after being arranged statistically in a personality profile. The scales and the profile that results from plotting scale scores on a graph to show significant deviation from a norm are the bases for personality analysis provided by the MMPI, not isolated answers. (T. 7.31-.32, 7.119) The LPS psychologist used the profile of scores, as is proper, and did not refer to individual answers. (T. 12.94-.95, 14.-67-.70)

In addition to four control scales which indicate when the MMPI cannot be used because the applicant has answered carelessly or insincerely, as, for example, to cover up even minor average vices ("lie scale"), there are ten interpretative scales. (T. 7.108-.118) Although the MMPI was developed by comparing scores of mental patients with typical Minnesota citizens, the MMPI is now also used for personality assessment of normal individuals. (T. 7.105) The original reference group of Minnesota citizens has been supplemented with specialized norms such as norms for college students, medical and nursing school students, or people over age 60. (T. 7.115) The MMPI has been translated into 50 languages, and is used in many countries. (T. 7.41-.42) The MMPI is administered in the United States several hundred thousand times per year (T. 7.93-.94)

In the LPS evaluation, the ten interpretative scales of the MMPI were analyzed against a special norm developed for policemen in a research study by Dr. Gottesman, after a prior study indicated similar MMPI patterns for policemen and firemen. (T. 14.67-.70, 14.92-.96, 15.17) Plaintiffs' experts testified that research had shown that the MMPI was not consistently accurate in predicting job performance and that the MMPI was usually used to diagnose pathology, not to predict behavior. (T. 3.88, 6.83-.84) However, defendants' expert, Dr. Dahlstrom, testified in detail how each of the MMPI scales could be useful in evaluating applicants in terms of the specified behavioral potentials. (T. 7.116-.119, 8.80-.81) Plaintiffs' expert in clinical psychology also stated that she had used the MMPI in making clinical evaluations. (T. 6.83-.84) Therefore, it can be concluded that the MMPI is a recognized tool in personality assessment for normal individuals, and that its use in evaluating fire-fighter applicants is likely to provide data on the desired emotional characteristics. The question of its accuracy in predicting job performance would be crucial if the MMPI were used as the sole screening device. It is clear, however, that the judgment of the LPS psychologists, based on the whole clinical battery, is the final outcome of the LPS evaluation, and that a certain MMPI profile is not a cut-off, as contrasted, for example, with the use of a minimum score on college aptitude tests to screen out college applications. Since this suit involves the testing procedure after 1972, when the MMPI was not used as the sole screening instrument, the issue of accuracy does not depend on the margin of error of the MMPI or any single test.

2. EPPS

The EPPS is designed to assess "need categories," such as need for aggression or achievement, in accordance with 15 scales representing relatively normal dimensions of personality. (T. 12.100) In the LPS evaluation the EPPS was scored according to standard norms for adult males. An extremely low or high score was used to corroborate similar indications of personality from other tests. (T. 14.71) Dr. Dahlstrom indicated that studies had been done correlating MMPI and EPPS scores. (T. 8.94) Dr. Lapidus stated that the EPPS was developed to be a pencil-paper measure of the needs, wishes, feelings, and preferences as originally obtained in the free-flow stories of the TAT but that use of the EPPS with the TAT would not be a needless duplication. (T. 6.38)

3. TAT

The stories written by the applicants from the TAT cards were reviewed by LPS psychologists for recurrence of themes of disturbing type indicating a potential for abnormal fears or preoccupations. (T. 14.-72) Thus, according to Dr. Gottesman, to a card showing a woman lying in bed, bust exposed, and a man fully clothed standing with his back to the bed, one arm thrown across his face, the most common response is a story of a man rising to go to work. Other responses which might still be acceptable, barring additional evidence, would be stories of a man who has watched his wife die of an illness, or of a rape scene. A deviant response would be a story of a man who plans to kill the woman by setting fire to the bed to watch her die in the flames. (T. 14.86; P-5a)

Cards have different "card pulls," that is, they elicit different lines of fantasy. (T. 8.11) The standard clinical administration of the test uses 20 cards over a two-day session. (T. 12.105) However, in the LPS evaluation five cards were recommended and generally used as the most effective for testing for behavioral potentials. Discretion was left to staff psychologists to use other cards if they felt it necessary. (T. 12.106-.110) The TAT was interpreted by LPS psychologists in relation to other data but was not formally scored. (T. 12.110, 14.72) Defendants' experts disagreed with plaintiffs' contention that objective scoring and use of the same cards for all applicants were required. While such standardization of TAT administrations is possible, and is important in research, interpretation based on the psychologist's trained impressions is an acceptable procedure. (T. 13.109-.110, 8.7, 8.12-.15, 15.72)

4. Rorschach Ink Blot Test

All ten cards of the Rorschach test were administered by the LPS psychologist during interview. The same psychologist later would write the final report. LPS psychologists were not instructed to use any specific means of interpretation; LPS relied instead on their expertise and professional judgment. (T. 12.118) Using common methods of interpretation which analyze the response in terms of content or "determinants," that is, use of color or movement, LPS psychologists would note responses which were not the popular ones described in Rorschach literature. Original responses that were bizarre or aberrant might form the basis for a hypothesis which would be checked against other data. The interpretation discounted for certain reluctance or suspiciousness generated by the testing situation itself in which the applicant was seeking employment and not psychological help. (T. 14.72-.73)

5. IST and DAP

The Incomplete Sentence Test was developed by LPS from existing sentence completion forms specifically for industrial use. (T. 14.74) It was closely modeled on one of the most common and well known incomplete-sentence tests. (T. 6.64, 8.8, 13.74) The standard scoring method for such tests is to total the entire set of responses and formulate an average or frequency of certain responses which is compared to scoring norms. (T. 6.63-.66) There is some evidence that in one instance a single response was not compared to the whole test. (T. 6.63, PSJ-1) However, as with the other test data, the LPS procedure was to cross-check and corroborate any significant responses on the IST with the other data. (T. 14.73-.74) The DAP or Human-Figure Drawing was examined for distorted features or size, or representation of obvious clinical significance. The DAP was used only for indications of phychosis or extreme disturbance. (T. 14.74-.75)

6. Interview

During the interview, according to defendants, applicants were asked about their reasons for seeking entry into the fire force, prior work experiences, general family background, and about significant events in their emotional development. (T. 14.-75-.76, 12.129-.130) Defendants stated that they did not ask questions concerning the applicant's sexual beliefs or practices, and that the purpose of the interview was not to inquire or elicit from candidates any religious beliefs or statements on religious practice. (T. 14.76, 12.131-.133) Dr. Gottesman stated that in his interview of Brian Flaherty he would never have asked any questions concerning premarital sex and that he did not recall or have notes showing that he asked questions concerning the relationship between plaintiff Flaherty and his parents, or concerning the drinking habits of plaintiff's father. (T. 14.79) Defendants did state, however, that if a candidate volunteered specific information about his sexual habits or difficult emotional relationships with family members, the interviewer would follow it up if there was reason to believe it was important. (T. 12.131, 14.77)

There is no evidence that the interviews were "stress interviews," that is, interviews designed to annoy and ridicule subjects to test their capacity for stress, a procedure now in disrepute. Rather, the testimony supports a finding that the purpose of the interview was to provide the LPS psychologist with a preliminary indication of any specific emotional problems or personality traits in an applicant which could be checked against the other test data. (T. 12.129, 14.76)

D. LPS Recommendation

The LPS psychologists who evaluated the testing data for plaintiffs were licensed psychologists with professional training and credentials. (T. 12.139-.141, .148) In the spring and summer of 1972 there were approximately 10 to 15 persons employed by LPS on a full-time or part-time basis for the Jersey City testing program. (T. 14.90) Dr. Gottesman testified that all staff members were screened and supervised by him or Dr. Springob, and that in almost every case the staff member was a licensed practicing psychologist in New Jersey. (T. 14.-90) One such part-time employee was Dr. Johnson, who had until 1968 been Manager of Counseling Services for Stevens. (T. 12.-140) Dr. Springob stated that the credential which LPS looked for was a Ph.D. or, in lieu of that, extensive experience, and that all but one of their current part-time staff had doctorate degrees. (T. 13.62-.64) LPS relied primarily on the training and experience of its staff, although several informal meetings were held to discuss the program. (T. 12.147, 14.91, 14.96) When LPS retained psychologists for the testing, they were instructed to read Dr. Gottesman's monograph and to consider the reasons it lists for rejecting an applicant who would be capable of functioning in most other jobs but whose personality, while not extremely neurotic, still indicated potential for emotional problems as a fire fighter. (T. 14.-93-.95)

The conclusions which were made as to an applicant's psychological fitness were those which were reflected and confirmed by significant measurements on the self-report inventories, in interpretation of the projective instruments, and from analysis of the applicant's interview behavior. (T. 12.141, 14.82) Because of its importance, an example given by Dr. Springob will be quoted in full (T. 12.142-.145):

"Let's assume for the moment that this hypothetical case, I guess is what we are dealing with, the individual is in the interview and the psychologist notes that as they greeted one another, that the individual had a very wet, sweaty palm in the shaking of hands.

That in and of itself may not have been very significant because it is an evaluation situation, and the individual may feel some modest trepidations about that, but if then in the interview there are continuing similar types of signs, such as the individual sitting on the edge of his chair, smoking a lot during the interview, almost chain-smoking, then the psychologist is on reasonably safe grounds to begin to hypothesize that there is anxiety here which is probably more than the situation calls for.

Now, let's assume that the same individual on the Rorschach, in the administration of the Rorschach, begins on the presentation of the first card with considerable delay in responding, and then gives a response to a card that many people very often see as sort of a bat, or flying animal, shape, and gives a response that is reflective of the grayness of the blot, a lot of gloominess, darkness, talks about the threatening aspect of the ink blot itself, and continuously, throughout the administration of the Rorschach there are other types of responses reflective of hesitancy, anxiety, difficulties in dealing with emotional content such as the colored cards, and he sticks very much to the basic form, he is not willing to go beyond that, then you get some added information about the original hypotheses.

Now, in the M.M.P.I., if the same individual then winds up with a pattern of scores that is reflective of very high peaks on Scale 7 and Scale 3, here you get further corroboration of a considerable amount of tension within this individual, anxiety, difficulties in coping with that, dependency qualities of Scale 3, the possibility that there may even be some physiological complaints that the individual comes up with, and you add to this the fact that on his TAT cards, the boy with the violin is experiencing trepidations and fear that his father is going to punish him if he doesn't practice the violin and practice it for three hours while his friends are out playing baseball, you have further evidence of at least anxiety.

You may get some other things coming in, too, for example, the boy's reaction to the authority figure of the father's punishment may be one of feelings of rage, so you may get some evidence about aggressive qualities and resentment of authority and how he would respond to taking orders from other people.

The Edwards Personal Preference Schedule, you may have in this particular type of an instance, somebody who has a very high score on the scale called Succorance, which is reflective of the type of an individual who really is very concerned about getting sympathy and understanding from other people, and at a very high extreme score, then again you are getting reflections of a kind of dependency, and why is the dependency there.

It follows from a psychological theory that there is anxiety involved, and it is an anxiety around the dependency issue.

The fact that this might also be an individual who gets a high score on the Deference Scale where he is almost ultra-conventional, ultra-conservative, conforming, because he is so anxious about doing something on his own that he doesn't dare, he is too threatened by it, so he conforms to the ultimate extreme.

. . . . .

In the figure drawing, you may with highly anxious individuals, find yourself with a presentation of, let's assume it is a male and he is drawing the male figure, he is drawing a fairly diminutive figure in comparison to the size of the female, the footing of the individual is such that he does not appear to be very stabilized in his environment. In fact, he draws no feet on the figure at all, and puts in a lot of embellishments, like a lot of buttons and belts, and there is a lot of midline emphasis in the figure, here you have somebody who is really defending himself and protecting themselves from threat, and where you have that occurring, then you have a lot of anxiety."

Dr. Springob's account of the use of the test data was confirmed by Dr. Gottesman's testimony. (T. 14.80-.85) The Court believes that it conveys accurately how an applicant's emotional characteristics were determined in the LPS testing. From the profile of the applicant's personality, an evaluation was made of the probability that the applicant could act in accordance with stipulated behavioral potentials. A brief summary and recommendation were prepared and delivered to Jersey City. Either Dr. Gottesman or Dr. Springob reviewed the summary recommendations of other LPS psychologists and, in cases of negative or marginal reports, they also reviewed the raw test data. (T. 12.146-.147)

E. Accuracy

While conceding that a formal validation study has not been attempted, defendants contend that such a project is theoretically unsound and therefore not likely to be conclusive, and that the evaluation and hiring procedure, taken as a whole, has a number of safeguards which guarantee its reasonable, if imperfect, accuracy, given the difficulty and need for the screening. Plaintiffs contend that without a formal validation study there is no basis for a judgment as to the relationship between the testing and what the testing is supposed to measure. Plaintiffs argue that a formal validation study is theoretically feasible and presumably, if done, would demonstrate the inaccuracy of the assessment and hiring procedure. Significantly, it is not clear what plaintiffs' position would be if, as seems likely, a validation study was done and the results were inconclusive.

It is, of course, too dangerous to determine the accuracy of the testing process by hiring all the applicants who were screened, including those with negative evaluations, and determining if those who later failed on the job would have been rejected. This variety of validation, called predictive validity, is not advocated by either party. (T. 3.22-.23) The fear of this method of proving accuracy, however, does seem to indicate that the assessment procedure's effectiveness is, at the very least, acknowledged as better than chance.

Plaintiffs' expert, Dr. Stander, testified that a concurrent validity study using the present work force, in particular first- and second-year firemen, is possible. (T. 5.86, 5.139) In general terms, in a concurrent validity study the present work force is evaluated according to job performance and then given the tests which are being validated. Given a work force that had been ranked by peer ratings or by supervisors' grading, those with high rankings would do well on the proposed tests and those with low rankings would do poorly, if the tests are valid and predict accurately. (T. 3.34-.40)

The difficulty with proving a test's validity as a predictor of successful behavior by a concurrent validity study is that in the validity study the population tested, that is, the work force, is a more homogeneous group than the general population of applicants who would take the tests to obtain the job. The current work force consists of individuals who have been through some selection process, and who have remained on the job over a period of time; they have, therefore, similar talents, experiences, and abilities, and are a more homogeneous group than the general population in which the range of differences between individuals would be greater. In general, tests predict individual differences better within a group having a wide range of abilities than in a restricted, pre-selected group where the differences are smaller. Consequently, in a concurrent validity study, when the tests are given to the current work force, they are not as likely to predict which members of the work force are most highly rated because the range of ability is restricted, hypothetically, from satisfactory to excellent. However, if the tests were used as a selection tool, and were given to the group of new applicants in which the range was from unqualified to excellent, the test would in theory actually show a higher degree of success in predicting successful performance — assuming that all of the applicants were later hired and it could be seen which applicants succeeded in fact. More important, tests that could not be shown to differentiate the restricted levels of the work force apparently might still be able to differentiate between broad categories such as above satisfactory and rejection. Thus, because of the restriction of range in the work force, a concurrent validity study is generally believed to underestimate the predictive accuracy of the tests if they were used on the general population of applicants. (T. 3.41-.42) There was also testimony from defendants' experts that the elimination of a portion of the applicants by the Civil Service examination makes a validation study very difficult. (T. 7.127, 8.118) Defendants' experts testified that certain statistical techniques for correcting for restriction of range could not be used with the type of evaluation done by LPS. (T. 8.123, 12.150-.152, 14.109-.114)

A fundamental limitation of a validation study using the current work force should be made clear. Testing present firemen would not directly provide information about firemen who were discharged or about applicants who were rejected. In effect, such a study would only indirectly provide a rejection criterion by gathering data on the characteristics of firemen on the force and by assuming that applicants should be as much like firemen already on the job as possible. (T. 4.43) Defendants point out that the best representatives of ideal firemen are experienced, older fire fighters with years of training, with perhaps different educational and cultural backgrounds from the applicant pool. (T. 14.112) However, the suggestion made by plaintiffs' expert, Dr. Stander, to use probationary firemen or those with two years' training, (T. 5.86), seems reasonable, and it might well be that some form of a study of the work force would be helpful in improving the accuracy of prediction.

Defendant Stevens, when pressed, conceded that a validation study and further research would be advisable, but stated that the costs would be substantial and that it has not made money from the testing program. (T. 14.109) Stevens takes the position that it has a professional obligation to do the assessment because of the life-endangering nature of fire fighting, notwithstanding the undeveloped state of psychological testing for predictive purposes. Defendants conceded the absence of clear proof of accuracy and rely on the general professional design of the clinical evaluation, and the cross-check of the psychiatric interview provided by Jersey City to all applicants receiving a negative LPS recommendation. In addition, applicants are now afforded a hearing before the Medical Review Board and can contest their evaluation with benefit of the right to access to the testing data in their files, as well as the right to present independent psychological evaluations. Questionable evaluations are, therefore, subject to being overridden, which, while putting the burden of going forward on the applicant, improves the fairness of the system.

Considerable testimony was introduced by plaintiffs — not just defendants — establishing the uncertainty and difficulty of proving the validity of personality tests and tests which in general attempt to predict emotional or affective behavior. (T. 3.76-.79, 4.33-.36, 4.58-.59) Plaintiffs contend that absent a conclusive validation study, the testing should not be used in the hiring system. It should be noted that it is hard to see why the same criticism would not apply to municipalities which use only a psychiatric interview without benefit of the data from testing instruments used by LPS.

The testimony presented to the Court does not provide support for a finding of fact as to the degree of accuracy of defendants' hiring procedure expressed in numerical or percentage terms; nor are there grounds for a similar finding as to the margin of error. The Court, however, believes the evidence establishes the following conclusions: because of the unique psychological factors which are crucial to the life-endangering occupation of fire fighting, a psychological and emotional assessment of applicants has an importance that would be found in very few other occupations; that the psychological assessment procedure used by defendants is likely to reduce substantially the risk of hiring firemen who are emotionally unfit for the job; and, finally, that the exact degree of accuracy or the margin of error of the testing cannot be conclusively demonstrated given the current state of validation methodology without serious danger to firemen and society.

PART IV

Plaintiffs' challenge to the constitutionality of defendants' hiring procedure begins on the sound premise that a government benefit, such as public employment, cannot be denied for reasons which amount to abridgment of constitutional rights. Just as the Government cannot by legislation directly bar exercise of constitutional freedoms, so too it is prohibited from commanding sacrifice or abridgment of those rights as the condition of securing public employment. See Elrod v. Burns, 427 U.S. 347, 360-61, 96 S.Ct. 2673, 49 L.Ed.2d 547 (1976); Perry v. Sindermann, 408 U.S. 593, 597, 92 S.Ct. 2694, 33 L.Ed.2d 570 (1972). Cf. Sherbert v. Verner, 374 U.S. 398, 83 S.Ct. 1790, 10 L.Ed.2d 965 (1963). That a citizen could choose to apply for other work is beside the point. The Government cannot put its citizens to the choice of sacrificing constitutional rights and securing governmental employment or refusing to forego constitutional freedom and looking for work elsewhere. The incentive of employment is a powerful one and cannot be used to accomplish indirectly that which could not be done directly. See Keyishian v. Board of Regents, 385 U.S. 589, 605-06, 87 S.Ct. 675, 17 L.Ed.2d 629 (1966). Cf. Elrod v. Burns, supra, 427 U.S. at 359 n. 13, 96 S.Ct. 2673 (accepting public employment not a waiver of First Amendment rights).

The constitutional rights of applicants for employment by Jersey City are infringed, according to plaintiffs, in two ways. First, plaintiffs argue that many questions on the MMPI and EPPS, as well as questions raised during the LPS interview, violate the freedom of belief protected by the First and Fourteenth Amendments. Second, they contend that the psychological testing process as a whole violates a constitutionally based right of privacy. While the issue of privacy is presented by the facts of this case, the problem of protection of freedom of belief is not.

Plaintiffs have picked through the MMPI and EPPS and have compiled a list of some 133 questions which, they contend, inquire into religious beliefs, political opinions, reading habits, sexual preferences, social beliefs, and familial relationships. Such inquiry, plaintiffs allege, results in a chilling effect on First Amendment rights (P. Br. Summary Jdgmt. 23), and might result in a candidate being marked for special scrutiny as a `"socially unacceptable'" individual (P. Supp. Br. Summary Jdgmt. 30). Surely, in this respect plaintiffs have missed the forest for the trees. The facts do not begin to suggest that applicants were tested for social, political, and religious beliefs or sexual attitudes in conformity with an orthodoxy mandated by Jersey City. In fact, Jersey City did not see the raw testing data and relied primarily on the LPS summary. The LPS evaluation was made for the purpose of selecting firemen who had a high probability of withstanding the psychological pressures of fighting fires and living in close quarters. The behavioral potentials which LPS referred to in making a recommendation were obviously relevant psychological and emotional factors, and not an orthodoxy of faith or political belief.

The MMPI and EPPS were designed and are used to make a clinical diagnosis of personality. It is a tenuous inference indeed from the existence of a particular question on the inventories to the conclusion that beliefs are the basis for hiring. As the facts make clear, a question is part of a scale which in turn is interpreted by statistical comparison to a reference group. The significant deviations from the norm for each scale form a personality profile when all the scale scores are entered on a graph. Studies and research on the profiles of various groups have provided ways of interpreting profiles to indicate personality traits and emotional characteristics. In this case, defendants have begun to develop a special norm for policemen and fire fighters; the norm reflects emotional fitness, not beliefs.

The Court does not dismiss lightly the potential for abuse of personality testing; it simply is not found in this case. It might be possible, for example, to develop a special personality norm for individuals who hold radical political ideas or unpopular religious ideas if certain common personality traits could be identified. Under the guise of rejecting the emotionally unfit, a government entity could penalize individuals indirectly because of their beliefs, a result it cannot constitutionally achieve directly. See e. g., McDaniel v. Paty, ___ U.S. ___, 98 S.Ct. 1322, 55 L.Ed.2d 593 (1978); Torcaso v. Watkins, 367 U.S. 488, 81 S.Ct. 1680, 6 L.Ed.2d 982 (1961); West Virginia State Bd. of Ed. v. Barnette, 319 U.S. 624, 63 S.Ct. 1178, 87 L.Ed. 1628 (1943). It might be possible as well to develop neutral questions found to correlate with individuals whose religious or sexual preferences are disfavored by society generally; the facial neutrality of the questions would disguise improper use of the known, underlying correlation. Even more insidious would be a government-sponsored school of psychology that equated dissident political opinions with symptoms of mental illness: only the mentally ill disagree with those in power and only good citizens are normal.

Because these examples do not exhaust the possibilities of abuse, judicial review of psychological testing is essential to prevent intrusions on the freedom of belief behind psychological jargon and methodology. During the trial of this case, the Court allowed great latitude to counsel in introducing evidence. The Court has now reviewed the evidence carefully and in detail, and has concluded that there has been no attempt directly or indirectly to condition employment with Jersey City on any requirement that limits freedom of belief.

Plaintiffs' second argument, based on the constitutional right of privacy, raises important problems of the scope and theory of that constitutional right. There is at the outset a close question whether the constitutional concept of privacy applies in this case. Research indicates that there is no Supreme Court holding or Court of Appeals decision that extends constitutional protection of privacy interests to facts such as those involved here.

The Constitution does not explicitly state that there is a right to privacy. While the source of the constitutional protection afforded privacy interests was only recently substantially agreed upon, a constitutional doctrine has been established variously described as recognizing a right to privacy, see Carey v. Population Serv. Int'l, 431 U.S. 678, 684-86, 97 S.Ct. 2010, 52 L.Ed.2d 675 (1977); Whalen v. Roe, 429 U.S. 589, 606, 97 S.Ct. 869, 51 L.Ed.2d 64 (1977) (Brennan, J.); or protected zones of privacy, see Paul v. Davis, 424 U.S. 693, 712, 96 S.Ct. 1155, 47 L.Ed.2d 405 (1976); Griswald v. Connecticut, 381 U.S. 479, 484, 85 S.Ct. 1678, 14 L.Ed.2d 510 (1965); or fundamental privacy interests, see Whalen v. Roe, supra; Moore v. East Cleveland, 431 U.S. 494, 546, 97 S.Ct. 1932, 52 L.Ed.2d 531 (1977) (White, J., dissenting). Cases on privacy indicate that constitutional protection of privacy extends first to the individual interest in making certain decisions of the most intimate kind, such as those relating to marriage, procreation, contraception, and abortion. See Carey v. Population Serv. Int'l, supra, 431 U.S. at 685,97 S.Ct. 2010; Whalen v. Roe, supra, 429 U.S. at 600 n. 26, 97 S.Ct. 869; Roe v. Wade, 410 U.S. 113, 152, 93 S.Ct. 705, 35 L.Ed.2d 147 (1973). Second, the doctrine has also been interpreted to limit the State's power to intrude on the privacy of the home or to interfere with certain family arrangements, such as the choice of family living arrangements or parents' control over child rearing and education. See Moore v. East Cleveland, supra, 431 U.S. 494, 97 S.Ct. 1932, 52 L.Ed.2d 531; Wisconsin v. Yoder, 406 U.S. 205, 232, 92 S.Ct. 1526, 32 L.Ed.2d 15 (1972); Stanley v. Georgia, 394 U.S. 557, 89 S.Ct. 1243, 22 L.Ed.2d 542 (1969).

A principle that embraces and explains the decisions concerning privacy might be formulated in several ways, but it is not now clear why some personal activities are more private than others. The confused state of the constitutional doctrine of privacy is indicated by the absence of a principle limiting its scope. Virtually every law or regulation intrudes to some degree on the innumerable activities which people conduct, or would like to conduct, in private or free from government regulation. Yet there is no constitutional right to individual anonymity. Whalen v. Roe, supra, 429 U.S. at 604 n. 32, 97 S.Ct. 869. Nor is there "an unlimited right to do with one's body as one pleases," Roe v. Wade, supra, 410 U.S. at 154, 93 S.Ct. at 727, or a right to enjoy all varieties of sexual freedom, see Carey v. Population Serv. Int'l, supra, 431 U.S. at 702, 703, 713, 97 S.Ct. 2010 (Powell, White, Stevens, JJ.) Nor does the constitutional right of privacy prohibit a State from publicizing an official record of an individual's arrest after the charges have been dismissed, even though publication hinders future employment. See Paul v. Davis, supra, 424 U.S. at 712, 96 S.Ct. 1155.

Recent judicial opinions have stated that the source in the Constitution for protection of privacy is the liberty aspect of the due process clause of the Fourteenth Amendment and that decisions which limit the content of laws in order to protect privacy are based on a return to substantive due process. See Moore v. East Cleveland, supra, 431 U.S. 494, 97 S.Ct. 1932, 52 L.Ed.2d 531; Roe v. Wade, supra, 410 U.S. at 167, 93 S.Ct. 705 (Stewart, J., concurring). Cf. "The Supreme Court, 1976 Term," 91 Harv.L.Rev. 72, 128 (1977). In this case, the threshold issue is whether the impact of the psychological evaluations on the privacy interests of the applicants burdens a "liberty" interest under the Fourteenth Amendment and whether the burden is of the magnitude necessary to mandate judicial review of the need for the evaluations. If the impact on the privacy interests is insufficient in kind or degree, the actions by Jersey City are clearly constitutional. Jersey City's hiring and evaluation procedure is a rational attempt to deal with a problem properly subject to the municipality's control. Whether it is unnecessary in whole or part in the Court's opinion would not make it unconstitutional. See Whalen v. Roe, supra, 429 U.S. at 596-98, 97 S.Ct. 869.

But for the reasoning in Whalen, there would not be sufficient support from the Supreme Court decisions on privacy to conclude that the constitutional doctrine of privacy extends beyond familial affairs and independence in intimate personal choices, the two areas described above, to the facts of this case, involving disclosure of personal information to the Government. At issue in Whalen was a New York statutory scheme designed to control the diversion of certain legal but harmful drugs into unlawful channels. A special legislative commission had determined that prescriptions for such drugs were fraudulently issued or refilled by unscrupulous doctors or pharmacists, or were abused by patients. The statute required that prescriptions for the controlled drugs be prepared by the physician on an official form which, when completed, revealed the name of the prescribing physician, the dispensing pharmacy, the drug and dosage, and the name, address, and age of the patient. Data from the forms was processed and stored in a central computer pursuant to regulations which limited access to the data and required that the data be destroyed after five years. A three-judge District Court enjoined enforcement of the reporting provision on the ground that the doctor-patient relationship was a constitutionally protected zone of privacy and that the patient identification provision invaded that zone without sufficient justification. Roe v. Ingraham, 403 F. Supp. 931, 937 (S.D.N.Y. 1975), rev'd sub nom. Whalen v. Roe, supra, 429 U.S. 589, 97 S.Ct. 869, 51 L.Ed.2d 64. The Supreme Court reversed, finding, first, that the statute did not pose a sufficiently grievous threat to cognizable privacy interests to establish a constitutional violation, and, second, that the statute was a reasonable exercise of the State's broad police power. Whalen v. Roe, supra, 429 U.S. at 598, 603-04, 97 S.Ct. 869, 51 L.Ed.2d 64.

From prior cases involving privacy, the Supreme Court concluded that at least two different kinds of privacy interests were within the constitutional doctrine of privacy. One of them, the individual interest in independence in making certain important personal decisions, was derived from the line of cases noted above. Id. at 600 n. 26, 97 S.Ct. 869. The other, the individual interest in avoiding disclosure of personal matters to the public or to the Government, was viewed as being reflected in the language of the opinions in several cases. Id. at 599 n. 25, 97 S.Ct. 869. But cf. id. at 608-09, 97 S.Ct. 869 (Stewart, J., concurring) (no general freedom from disclosure).

Why the Supreme Court concluded that the privacy interest in nondisclosure of private information was not sufficiently threatened in Whalen is crucial to this case. Disclosure of the data contained in the report form to the public was found to be unlikely because the record in Whalen provided no support for an assumption that the security provisions of the statute would fail to prevent State employees from deliberately or negligently revealing the computerized data, and because judicial supervision of any evidentiary use of the data was adequate to prevent unwarranted disclosures. Id. at 601, 97 S.Ct. 869. More important, the Supreme Court recognized that there was a privacy interest in not disclosing personal information to government employees. However, in Whalen, this interest was not unduly burdened. Disclosure to State employees was found not significantly different from the many disclosures of private medical information required by modern medical practice, including, for example, disclosures to insurance companies and hospital personnel. Id. at 602, 97 S.Ct. 869.

The private information disclosed to State employees in Whalen consisted of the patient's name, age, address, and use of a certain drug. In this case, the degree and character of the disclosure is far greater and more intrusive. The evaluation looks deeply into an applicant's personality, much as a clinical psychologist would if requested to do so by an applicant. Fire-fighter candidates are called upon to reveal the essence of their experience of life, the collective stream of thoughts and feelings that arise from the ongoing dialogue which individuals carry on between the world and themselves in the privacy of their being. It involves a loss of the power individuals treasure to reveal or conceal their personality or their emotions as they see fit, from intimacy to solitude.

Involuntary disclosure of such a unique kind to defendant Stevens, acting as the agent of Jersey City, distinguishes this case from Whalen. While, in this case, as in Whalen, the facts show there has not been any public disclosure, the character and amount of information given to the Government alone is itself an intrusion on the privacy interest in nondisclosure of personal information to government employees recognized in Whalen. That privacy interest is sufficiently burdened, the Court believes, to call constitutional protection into play. Though this conclusion is contrary to the holding in Whalen, it is difficult to imagine what more private information could be required by the Government than that involved in this case. The analysis in Whalen, therefore, compels the conclusion that the defendant Jersey City must justify the burden imposed on the constitutional right of privacy by the required psychological evaluations.

The constitutional protection afforded privacy interests is not absolute. State interests may become sufficiently compelling to sustain State regulations or activities which burden the right of privacy. The regulations, however, must be narrowly drawn to express only those compelling State interests. See Carey v. Population Serv. Int'l, supra, 431 U.S. at 685-86, 97 S.Ct. 2010; Roe v. Wade, supra, 410 U.S. at 155-56, 93 S.Ct. 705. Plaintiffs agree that there is a compelling State interest in promoting an efficient fire department, but contend that Jersey City has no compelling interest in the particular psychological evaluation procedure involved here. (Plaintiffs' Proposed Conclusions of Law at 55.) The Court disagrees with this conclusion because it has not accepted plaintiffs' view of the facts, that is, that the evaluation procedure is unreliable and not even rationally related to the goal of improving fire department personnel. (Plaintiffs' Proposed Conclusions of Law at 55.) The Court does not believe that a margin of error is the equivalent of unreliability. There is sufficient support to conclude that the psychological evaluation and hiring procedure taken as a whole is useful and effective in identifying applicants whose emotional make-up makes them high risk candidates for the job of fire fighting. Because fire fighting, like police work, involves life-endangering situations, the State interest is of the highest order. Plaintiffs lose sight of the fact that a fireman who loses emotional control endangers his own life as well as those of other firemen. While a psychological evaluation intrudes on an applicant's privacy, it may save him from the risk of losing his life. The life of a community, as well, depends, at the most basic level, on those whose job it is to protect the community from physical forces, like fire, that have escaped from the control that makes them productive. Property, and the security of the community, as well as lives, are at stake in improving the fire department. The State interest is compelling, indeed, and is served by the challenged evaluation and hiring procedure.

Plaintiffs have suggested three alternatives to the defendants' procedure to support their argument that the procedure is not narrowly drawn, or that less intrusive examinations are available. (Plaintiffs' Proposed Findings at 46-49.) Plaintiffs suggest that evaluations be done only in the cognitive, as opposed to the affective, domain, or that the training program be relied upon. The testimony of Chief Fargo aptly demonstrates the relevance of testing for emotional and psychological factors, and the inadequacy of evaluations based on the limited situations presented in a training program. Plaintiffs' suggestion to place greater reliance on the personal history of the applicant would probably require investigation into an applicant's past, intrusive in itself, and is unlikely to be as useful in discriminating among applicants with average backgrounds.

Defendants' procedure, however, does have the potential for abuse in two respects. First, access to the data was not regulated by any formal Jersey City policy or regulations. There would, of course, be no reason to allow public disclosure or disclosure beyond the Stevens and Jersey City personnel who use the data. Although access to the data was properly limited voluntarily when the program was in operation, rules and sanctions limiting access to defendants' employees on a need-to-know basis are necessary to safeguard the data. See Whalen v. Roe, supra. Retention of the data has been justified by the possibility of obtaining funding for a validation study or for research to improve the evaluative norms and to comply with regulations governing appeals. When securely safeguarded from improper access, the retention of the data for a reasonable time is justified. However, no reason appears from the evidence for retaining the data indefinitely. A limit should be set by defendant Jersey City as to how long the records will be retained and when the existing data will be destroyed. A limit reasonable in light of the legitimate purposes advanced for retaining the data should not prove difficult to establish. It should be noted that the Court has not found that defendants' past conduct with respect to access and retention of the data was improper, but rather that regulations and formal policies must be drawn by Jersey City to insure that when the testing is resumed, it will operate only to serve the compelling interests which justify the burden imposed on the applicants' constitutionally protected privacy interests.

To conclude, the Court has found that defendants' evaluation and hiring procedure did not violate plaintiffs' rights of freedom of belief, and that the intrusion on plaintiffs' constitutionally based privacy interests was justified by the State's need for a procedure which advanced compelling State interests and which was narrowly drawn to further only those interests. Subject to promulgation by defendant Jersey City of rules governing access and retention of the data in accordance with this opinion, the Court finds that resumption of defendants' activities would not be unconstitutional. Accordingly, judgment will be entered in favor of defendants. Counsel are directed to submit an appropriate order.