Fowler v. Schwarzwalder, 351 F. Supp. 721 (D. Minn. 1972)
December 6, 1972
Ruth SCHWARZWALDER et al., Defendants,
Daniel R. Sounder et al., Intervenors.
United States District Court, D. Minnesota, Third Division.
*722 *723 Andrew W. Haines, Michael T. McKim, Dolores C. Orey and Roger S. Haydock, St. Paul, Minn., Michael A. Wolff, Denver, Colo., for plaintiffs.
Kenneth J. Fitzpatrick and R. Scott Davies, St. Paul, Minn., for defendants.
James P. Miley, St. Paul, Minn., for intervenors.
MEMORANDUM AND ORDER
DEVITT, Chief Judge.
Principally at issue in this declaratory judgment action is the validity of a written fire fighters test against the claim that it is unfair to minority persons.
Seven Black residents of St. Paul, each of whom took a Civil Service examination on July 8, 1972 to be a St. Paul fire fighter but failed to place among the top 20 applicants bring a class action against the members of the St. Paul Civil Service Commission, its Director and the St. Paul fire chief asking for injunctive relief from the claimed discriminatory practices of defendants in recruiting, examining and hiring St. Paul fire fighters. Intervenors are persons within the top 20 applicants who oppose the requested relief.
Jurisdiction is based on 28 U.S.C. §§ 1343(3) and (4) and the civil rights laws, 42 U.S.C. §§ 1981 and 1983.
The class is properly defined, Rules 23(a) and (b) (2) Fed.R.Civ.P., as:
"All those Black, Indian and other minority persons who have made application to the St. Paul Civil Service Commission for a position as a fire fighter with the St. Paul Fire Department and who have been deemed eligible but who have nevertheless been denied employment."
On October 6, 1972 the Court directed defendants to withhold certification and employment of fire fighters from the eligibility list until further order, 348 F. Supp. 844. Hearings on the merits were held on November 24, 27, 28, 29, 30 and December 1, 1972.
It appears from the files, records and testimony that St. Paul has a 1970 population of 309,980 and a minority population of between 4.6% and 8% depending upon whether Mexican-Americans are included. The exact number of Mexican-Americans was not shown. There are 459 fire fighters on the St. Paul Fire Department. Between six and ten of that number, or between 1.3% and 2.2%, are members of minority groups. The percentage at one time was much higher. Fifteen Black firemen once served on the force. Plaintiffs allege that the present low percentage of minorities is due to discriminatory practices by defendants in recruiting and testing applicants. None of the top 20 applicants in the 1972 test were minorities, and no more than 20 will be certified from the list. The top minority applicant is number 56 on the priority list. Plaintiff's expert, David J. Weiss, said the difference in test scores between minority and non-minority members, when considered in the light of the arbitrary passing score and the limited number of successful applicants to be chosen, compelled the conclusion that the use of the test scores would discriminate against minority group members.
The law is clear in this judicial circuit that statistics reflecting a *724 marked disparity can alone establish a prima facie case of discrimination. Parham v. Southwestern Bell Telephone Co., 433 F.2d 421 (8th Cir. 1970); Carter v. Gallagher, 452 F.2d 315 (8th Cir. 1971). I find that these statistics establish a prima facie case of discrimination. It then becomes the duty of defendants to sustain the burden of presenting evidence to overcome the prima facie case. Griggs v. Duke Power Co., 401 U.S. 424, 91 S. Ct. 849, 28 L. Ed. 2d 158 (1971); Parham, supra, and Carter, supra.
Chief Justice Burger said: "If an employment practice which operates to exclude Negroes cannot be shown to be related to job performance, the practice is prohibited." Griggs v. Duke Power Co., supra, 401 U.S. at 431, 91 S. Ct. at 853.
Defendants recognize their obligation to show job relatedness, or validity, of the challenged test.
The principal issue, therefore, is the validity of the July 8, 1972 test. It is not claimed that the defendants acted with ill-purpose or discriminatory intent in constructing the test. Indeed, purposeful discrimination is not essential to a finding of illegality. Griggs v. Duke Power Co., supra, 401 U.S. at 432, 91 S. Ct. at 854.
Defendants sought to prove job relatedness or validity of the test through testimony of Civil Service personnel who prepared it and of the fire chief and fire department training officer. Their testimony tended to show that the test had content validity in that it sampled the knowledge, abilities and skills demanded of a fire fighter. An expert in the field, Vernon R. Taylor, said that the job analysis was adequate, that the test had a high degree of content validity, and that the questions effectively and fairly measured the abilities needed to succeed as a fire fighter. He said the test was reliable under the recognized Pewter-Richardson calculation of reliability.
Plaintiff's expert, David J. Weiss, characterized the examination as discriminatory against minority group members. Dr. Weiss said it tested specific mechanical and scientific knowledge which assumed an educational and socioeconomic background derived principally from physics and applied mechanics. That background, he said, is most prevalent among non-minority persons and hence the test was potentially discriminatory against minority group applicants. He said the best method of insuring fairness was through use of an examination, the validity of which had been tested by the predictive validity method, that is, by a comparison of test scores obtained prior to appointment with criterion measures of job success after appointment. He also said that a concurrent validity study, conducted on existing firemen, would be appropriate.
Most of the recent cases in the field of employment discrimination have been decided under Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq. Chief among them is Griggs v. Duke Power Co., supra. Although the 1964 Civil Rights Act was not made applicable to public employees (by 1972 amendments they are now covered), the rationale of the decisions, particularly Griggs, affords an appropriately strong analogy to the issues raised under the Constitution and 42 U.S.C. §§ 1981, 1983.
The Equal Employment Opportunity Commission, created by Congress to enforce Title VII of the 1964 Civil Rights Act, has adopted Guidelines on Employee Selection Procedures, commonly abbreviated EEOC Guidelines. 35 Fed.Reg. 12333 (Aug. 1, 1970). These guidelines also have been adopted by the Office of Federal Contract Compliance, the agency charged with the enforcement of Executive Order 11246 against discrimination by government contractors, 35 Fed.Reg. 19307 (Oct. 2, 1971), as well as by state agencies charged with anti-discrimination duties, e. g. California, Colorado and Pennsylvania. The principles embodied in the EEOC Guidelines have received wide judicial recognition as appropriate standards in the field of employment testing. See, e. g., Griggs and *725 Carter. See also Western Addition Community Organization v. Alioto, 340 F. Supp. 1351 (N.D.Cal.1972) and cases cited n. 3, at 1354.
Regardless of the conscientious efforts of St. Paul Civil Service Commission personnel to construct a test which was fair and free from bias, the law requires there also be a substantial showing that the test is job-related, or in the terminology of the psychologists, that it has validity.
The principle is well stated by defendants' expert, Vernon R. Taylor, in his summary of the EEOC Guidelines. The use of a test which adversely affects minorities constitutes discrimination unless the test has been validated and "evidences a high degree of utility," and the user can show that no other suitable selection method is available. Mr. Taylor said further that test validation generally means predictive validity or statistical evidence of validity and that content validity is acceptable only when predictive validity studies are not feasible. These statements summarize, principally, §§ 1607.3 and 1607.5 of the Guidelines. Abstracted from Defts. Ex. 10, "Employment of the Disadvantaged in the Public Service" by Vernon R. Taylor at p. 7.
Section 1607.5, which recognizes the use of content validation, requires that it "should be accompanied by sufficient information from job analyses to demonstrate the relevance of the content." The comparable Office of Federal Contract Compliance (OFCC) Guidelines require a "careful job analyses," § 63-3.5(b) (3).
In this case the defendants have sought to respond to their duty of negating the inference of discrimination by the content validity method. They do not claim they have made any predictive studies to validate the examination.
Thus, the issue of whether defendants have overcome the prima facie showing of discrimination focuses upon (1) whether defendants' efforts to validate the examination by the content validity method were based on a careful job analysis and (2) whether defendants also have shown that the more reliable predictive validity method was not feasible.
It is my conclusion that defendants have not shown the test has content validity, principally because of the absence of a proper job analysis, and have not shown the infeasibility of a concurrent or predictive validity study.
In constructing the test, civil service personnel reviewed the class specifications for the position of fire fighter and selected questions from a card file which appeared to relate to duties and skills of a fire fighter. In view of rumors of a lawsuit challenging the fairness of the test, they removed some questions used in previous tests and changed the wording in others in a good faith effort to eliminate any cultural or educational bias.
The class specifications used consisted of a list of job duties. No effort was made to rate the relative importance of each and to correlate that rating to the number and emphasis of the test questions. One or more of defendants' witnesses testified that the class specifications were the result of a job analysis, but it was not clear when the last formal job analysis was made or where it could be found. The testimony showed that the job specifications were the product of informal exchanges between fire department and civil service personnel. It did not appear that the specifications were based upon a systematic, empirical review of the elements that make up job performance.
It is also worthy of comment that the job specifications do not contain many of the qualities which Chief Conroy and his training officer said were essential to a good fire fighter. Conroy ranked these qualities as desire, mental alertness, good mechanical aptitude, physical strength and ability to get along with people. Few of these qualities are reflected in the job specifications. While Chief Conroy's listing may be empirically *726 suspect, it does indicate that the specifications are incomplete and are not the result of a careful job analysis.
The process by which the test was validated entailed only the judgment of the test constructors. It was a judgmental validation.
While a reading of the test questions prompts the immediate conclusion that they do pertain to the fireman's job and are free of any cultural bias, present standards of psychological science, and the law, require more than this to validate a test.
Three experts in psychological testing gave evidence for the plaintiffs. Their testimony, and the opinions of other psychologists contained in booklets and articles received in evidence, told of the detailed, careful, scientifically conducted procedures required to make a proper job analysis as a necessary part of a "well developed test" essential to establishing content validity under EEOC Guidelines. It would be superrogatory to detail this evidence and the other testimony showing the inadequacy of the job analysis and of the method employed to show content validity. It is enough to say that the evidence was persuasive that a proper job analysis was not done.
Dr. Weiss testified that while the use of content validity methodology is appropriate to verify an achievement test, it is ineffective to validate an ability test intended to predict the suitability of a candidate for a job.
I find that the defendants have not satisfied their burden of showing that the test was properly validated.
Defendants sought to show the infeasibility of the predictive validity method largely through the testimony of their testing expert, Vernon R. Taylor. He said the method was not suitable for testing public employees. Experts on the other side testified to the contrary. Mr. Taylor said the turnover rate was too high to make the test feasible. Evidence showed there was a 5% turnover each year or about 15 men. Plaintiffs' expert thought this rate of turnover would not affect the feasibility of the test. The main issue on this point revolved around the number needed in a group to furnish a broad enough sample to make the test reliable. Mr. Taylor said the number of fire fighters on the St. Paul force was insufficient to make a good test. Plaintiffs' witnesses said any number over 50 would be adequate for this purpose.
It is worthy of note that Thomas J. Bouchard, Jr., Ph.D., author of "A Manual for Validating Selection Tests," introduced in evidence, says in this regard:
"The most acceptable form of validity is criterion-related validity. Content or construct validity is second best and acceptable only when criterion-related validity is not feasible. This means that any job with * * * a large number of people in it (50) must be studied empirically."
The preparation of the Manual was commissioned by the St. Paul Civil Service Commission and other public agencies.
Perhaps the most persuasive evidence of the feasibility of empirical or criterion related validity was furnished by two witnesses who told of the experience of Minneapolis officials following the Carter decision in 1971. They engaged a private firm, Personnel Decisions, Inc., to prepare a model testing program for fire fighters. It conducted a test validation study by the concurrent validity method (recognized by the experts for both sides as appropriate and feasible in tests of this kind) and found test questions which were demonstrably job related. The battery of questions so validated has been used on two occasions in Minneapolis to recruit new fire fighters. In view of the apparent similarity of fire fighter duties in the two cities, it is reasonable to conclude that the same kind of empirical validation would be feasible for St. Paul.
The Court is advised that defendants have already joined with the Civil Service Commission of Minneapolis to provide *727 St. Paul and Minneapolis with fire and police examinations prepared by outside experts in the field which will follow EEOC Guidelines.
The Minneapolis experience in empirical test validation and preparation of job related tests should afford an experience and knowledge readily adaptable with reasonable speed in St. Paul.
It is also claimed that defendants did not affirmatively recruit minority applicants for the fire fighter examination. The record does not support this claim. Civil Service Director Haider testified to the wide public circulation given to all announcements for fire fighter examinations. Chief Conroy told of his efforts, since appointment in 1968, to encourage minority applicants. He instituted a recruitment program under the supervision of Robert N. Harris, a Black fire engineer. Harris called Chief Conroy's program "bold and innovative." Harris, while drawing his city pay, conducted a recruitment and tutoring program for minorities. He spent full time at it for several weeks prior to the 1972 examination and for similar periods prior to the previous two fire fighter examinations. The record shows that Harris had 21 minority members studying for the 1972 test. When Civil Service Commission members, the Mayor and fire chief were contemplating postponing the examination, Harris wrote the Mayor and said, "I implore you to let the test be given as scheduled." Twenty-two of the 344 candidates who took the examination were members of minorities.
Unlike the Minneapolis fire chief in the Carter case, supra, who, "took a strong position against recruitment and employment of Blacks," 452 F.2d at 323, the record here shows that Fire Chief Conroy affirmatively sought minority persons for the department.
The Court concludes that defendants have not rebutted the prima facie case of discrimination in the content of the examination made out by the disparity in statistics, and enjoins defendants from employing persons on the eligibility list resulting from that examination and from use of it in future examinations until it has been validated in accordance with the EEOC Guidelines.
These expressions are intended as findings of fact and conclusions of law in compliance with the requirements of Rule 52 of Federal Rules of Civil Procedure.NOTES
 There is a conflict in the evidence as to the exact number of minorities in the fire fighting force. Fireman Harris, a Black, says there are six. Fire Chief Conroy says there are ten. Apparently there is a dispute as to whether four fire fighters are of American-Indian blood.