Rossini v. Ogilvy & Mather, Inc.

597 F. Supp. 1120, 41 Fair Empl. Prac. Cas. (BNA) 861, 1984 U.S. Dist. LEXIS 22052

• Court: District Court, S.D. New York
• Decided: November 13, 1984
• Docket: 78 Civ. 1713
• Status: Published

Opinion

OPINION

GAGLIARDI, District Judge.

Plaintiffs commenced this action against Ogilvy & Mather, Inc. (“O & M”), an advertising agency, alleging employment discrimination based on sex in violation of Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq. (“Title VII”). 1 The case was tried without a jury from October 3, 1983 through November 15, 1983. The court now makes the following findings of fact and conclusions of law pursuant to Rule 52(a), Fed.R.Civ.P.

Facts

At the end of 1976, O & M employed approximately 800 employee's in New York City. About 200 of those employees were office and clerical workers; the remainder were classified as officials, managers, and professionals. By 1983, the total number of New York employees had grown to 1,045, of whom approximately four hundred were office and clerical workers.

O & M is divided into four major departments and a number of smaller, supporting departments. The largest departments, in terms of numbers of employees, are account management, creative, media, and research. The smaller, supporting departments at O & M include legal, accounting, personnel, and marketing services. Since the inception of this litigation, there have been as many as sixty-six divisions, departments and/or units within O & M at any one time.

The account management department serves client advertisers, and plans and executes advertising programs. The creative department produces the words and images which are heard and seen as advertising. The media department directs the purchase of advertising space in print media or advertising time in broadcast media. The research department obtains information about ad or product performance which can be used by other departments. The staffs of the major departments often are functionally divided into account groups, that is, groups of employees assigned to service a particular client or product.

An individual seeking a professional position at O & M can apply directly to the O & M personnel department, or can apply through an employment agency, a college or graduate school recruitment program, or a letter to or personal contact with a professional staff member in any O & M department. Entry-level professional jobs may be filled by transfer of O & M employees working in clerical or other support positions. Professional positions higher than éntry level frequently are filled by promotion from within the department in which the vacancy occurs.

The personnel department generally recruits candidates and conducts interviews of applicants for professional positions in all departments except the creative department. In addition, interviews usually are conducted by one or more employees in the department or group for which the applicant has applied or is being considered.

Candidates for positions in the creative department generally submit a resume and portfolio as part of their initial application. The creative department has its own personnel manager and deputy or assistant personnel manager, who interview applicants. Candidates for creative department positions also may be interviewed by staff members of that department. On a number of occasions since 1975, O & M has screened creative department applicants by the use of “copy tests,” writing exercises to measure ability to create advertising copy. O & M has ceased using such tests.

Until 1982, O & M had no written job descriptions. O & M has not established any minimal educational or experience requirements for any of its professional positions. O & M does not post job listings or otherwise formally advise its staff of opportunities for promotion or transfer within the New York office. Since March 1976, O & M’s stated policy has been to require department heads and management super: visors (or account group directors)' to advise the personnel office of job openings. 2

O & M’s stated policy has been to evaluate the performance of supervisory employees once a year and that of non-supervisory employees approximately twice a year, although in practice some employees have been evaluated less frequently. An evaluation form is sent to each employee’s department head, or, if the employee works in thé account management department, to the employee’s management supervisor or account group director. The evaluation form is then sent to the personnel department and to the top officials at O & M, and eventually is retained in the employee’s personnel file. A document entitled “category ratings” describes the number ratings, or, during some periods, letter ratings, which may be assigned to an employee as part of the evaluation. The categories in which an employee is rated include “marketing knowledge,” “analytical ability,” “writing ability,” “presentation skill,” “initiative/energy/drive,” “supervisory skill/leadership,” and “ability to take pressure.”

With the exception of a few entry level positions, O & M has no written minimum or maximum salary for any professional job title. Because salaries are kept confidential, employees have access only to salary information regarding their supervisees. Each year, department heads and management supervisors/account group directors are given department or account group salary budgets based on O & M’s financial situation. Those supervisors make recommendations regarding salary increases, but are not given any written guidelines for the amount of increases permissible. Those requests for pay increases are considered by a salary review committee composed of three or four of the highest members of O & M management. In passing upon a proposed raise, the committee considers, inter alia, the employee’s salary history, his or her performance category rating, and the salaries received by those in comparable positions in O & M and in other advertising agencies in New York.

Of the professional and managerial employees at O & M, approximately 150 also have titles as officers. The lowest official title is vice-president. When an employee is elected vice-president, his or her job duties and salary do not change. A new vice-president will receive a small amount of O & M stock and periodically will attend officers’ meetings, but will not, by virtue of the official title, be afforded any additional authority in running the agency.

In May 1970, O & M hired plaintiff Jane Zukofsky as an assistant broadcast operations director in broadcast operations. That unit, which is also known as broadcast forwarding, is part of the account coordination and services department. 3 On November 16, 1970, Zukofsky was promoted to broadcast operations director. From the date of her hire to the present, Zukofsky’s immediate supervisor has been Sally Bieley, manager of the broadcast operations division.

Within a few years of her hire at O & M, Zukofsky began a series of efforts to be transferred from broadcast operations. In or about 1974, Zukofsky met with Jack Silverman, a producer of television commercials in O & M’s creative department, in an effort to obtain a position as an assistant producer. In or about the spring of 1975, Zukofsky submitted a “story board,” a television commercial outline, and other writing samples to John Rand, a high-ranking employee within the creative department. Sometime in 1978, Zukofsky met with Bob White, a senior employee in account management, to discuss the possibility of her transfer to that department. . Zukofsky was not transferred from broadcast operations.

In 1970, O & M hired Carlotta Rossini as an assistant account executive in the account management department. Before working at O & M, Rossini had received a high school diploma and had taken several college courses but never had matriculated for studies leading to a bachelor’s degree. Her prior work experience included a number of -positions in advertising or fields related to advertising. Rossini worked for approximately one year for W.R. Simmons, Inc., supervising individuals doing field studies of magazine readership. Thereafter, she worked for 11 months for Mar-plan, Inc., a research subsidiary of the Interpublic Group, in a position which also required her to supervise the work of employees completing market research questionnaires regarding consumer preferences. In 1967, Rossini worked fob two months as a media buyer at Wunderman, Ricotta & Kline, Inc., an advertising agency specializing in direct response, that is, consumer purchases through coupons in print advertising or a telephone number in broadcast advertising. She then worked as an account executive for Brownstone Associates, a small agency handling advertisements for restaurants and theaters. When that business disbanded due to financial problems, Rossini formed her own two-person agency where she worked from November 1967 through June 1970. At- Rossini/Stevens Associates Inc., Rossini handled accounts including a book club, the Malagasy Republic and Bel Paese cheese. Prompted by difficulties procuring capital and her desire to work for a major ad agency, Rossini closed her firm and came to O & M in 1970.

At O & M, Rossini began work on the junior fashions and home sewing divisions of the Sears account under the supervision of Jim Himonas. In September 1970, she was promoted to account executive and within a few months began work on Sears home fashions. In November' 1971, while continuing her work on the Sears accounts, Rossini also assumed responsibility for Owens-Corning Fiberglass, an account which entailed primarily trade advertising. About a year later, Rossini assumed additional responsibilities on the Cotton, Inc. account.- After some time, she began to work exclusively on that account and in September 1974 was promoted to account supervisor on Cotton, Inc. During the last months of her work on Cotton, Inc., Rossini assumed responsibility for a rug cleaner called “Spray ‘N’ Vac,” which was part of the Glamorene account.

In December 1974, Rossini was moved from Cotton, Inc. to full time work on Glamorene, with responsibility for “Drain Power,” a relatively new drain cleaner being produced by that manufacturer. This was Rossini’s first experience on an account for packaged goods, that is, consumer products purchased “off the shelf” in supermarkets or drug stores. The Drain Power product failed, and in March 1976, Rossini was reassigned to the service establishment division of the American Express (“Amex”) account. That division produced advertising designed to encourage hotels, restaurants, airlines, and theaters to accept the American Express card and to encourage the public to - use that card. Eventually she also assumed responsibility for the Amex service establishment division’s advertising to retail stores.

In 1977 Rossini was elected vice-president of O & M and in April 1981 she was promoted to management supervisor with responsibility for the 1982 World’s Fair- in Knoxville, Tennessee. After that advertising campaign ended in 1982, Rossini was moved from the World’s Fair account back to a management supervisor’s position on the Amex account, where she-has remained to the present.

I. O & M’s Motion for Class Decertification

In a memorandum decision and order dated October 18, 1978, Judge Brieant denied a motion by plaintiffs Zukofsky and Rossini seeking certification of a plaintiff class. 4 Both were held to be inadequate class representatives under Rule 23, Fed.R.Civ.P., because at that time Zukofsky had not complied with the administrative prerequisites to a Title VII action and Rossini, as a vice president of O & M, was found to have a potential conflict of interest with the non-officer members of the putative class.

In a memorandum decision dated April 18,1979, Judge Brieant adhered to his prior determination regarding Rossini’s inadequacy as a class representative, but reversed his earlier decision regarding Zukofsky who, by late 1978, had complied with the Title VII administrative prerequisites. The court thereupon certified Zukofsky as representative of a class comprising “all females (excluding corporate officers) who are, have been, or will be, or were since May 30, 1975 employed by [O & M] as managers or professionals and who have been, are, continue to be or would be affected by the discriminatory practices of defendant.” 5

In the fall of 1982, O & M moved to decertify the class on the basis of the Supreme Court decision in General Telephone Co. of the Southwest v. Falcon, 457 U.S. 147, 102 S.Ct. 2364, 72 L.Ed.2d 740 (1982), issued subsequent to Judge Brieant’s rulings. In General Telephone v. Falcon, the Supreme Court emphasized that a -minority or female employee is not automatically an appropriate representative of an “across-the-board” class of minorities or women challenging all aspects of a Title VII defendant’s employment practices:

We cannot disagree with the proposition underlying the across-the-board rule— that racial discrimination is by definition class discrimination. But the allegation that such discrimination has occurred neither determines whether a class action may be maintained in accordance with Rule 23 nor defines the class that may be certified. Conceptually, there is a wide gap between (a) an individual’s claim that he has been denied a promotion on discriminatory grounds, and his otherwise unsupported allegation that the company has a policy of discrimination, and (b) the existence of a class of persons who have suffered the same injury as that individual, such that the individual’s claim and the class claims will share common questions of law or fact and that the individual’s claim will be typical of the class claims.

Id. at 157, 102 S.Ct. at 2371 (footnotes omitted).

O & M argues that under the Falcon analysis, Zukofsky has failed to demonstrate that her claims are typical of, or raise questions of law or fact common to, the claims asserted on behalf of unnamed class members. 6 That argument is best analyzed by considering separately each type of claim raised by Zukofsky on her own behalf and on behalf of the class.

A. Salary

The court examines first the allegations of salary discrimination. Zukofsky claims that O & M, on the basis of her sex, paid her less money for the same work than it paid to a comparably-qualified man, Bill Williams, the only man in the twelve-person broadcast operations department. 7 She also alleges that she was paid less by virtue of her employment in a largely female department, purportedly deliberately segregated by O & M on the basis of gender. Plaintiffs argue that O & M also has paid less to unnamed class members than to similarly qualified men doing comparable or less responsible work and that O & M discriminatorily has depressed the salary of its female professional employees on a classwide basis.

A number of common factual and legal questions are presented by Zukofsky’s salary claim and those of the class. For example, central to both the individual and the class claim is the question of whether the absence of fixed salary ranges for each job title served to depress women’s wages in a discriminatory fashion. Similarly common issues are raised by the lack of objective salary criteria, i.e., fixed percentage or dollar increases based upon years of experience, level of formal education, account billings, or any factor readily determinable without reference to a supervisor’s subjective evaluation of an employee. 8 A small committee, with relatively constant membership, had ultimate responsibility for the salary increase decisions affecting Zukovsky and the unnamed class members. Cf. General Telephone v. Fal con, supra, 457 U.S. at 162 n. 1, 102 S.Ct. at 2357 n. 1 (Burger, C.J., concurring) (typicality absent where failure to allege that individuals determining plaintiffs promotion also determined whether unnamed class members would be hired); Coser v. Moore, 739 F.2d 746, 750 (2d Cir.1984) (statistical evidence more significant in Title VII case “where a single office makes all the employment decisions”). 9

Significantly, the statistical proof which formed the core of the class salary claim also was relevant to Zukofsky’s individual claim as a possible indicator of O & M’s alleged discriminatory intent. See McDonnell Douglas Corp. v. Green, 411 U.S. 792, 804-805, 93 S.Ct. 1817, 1825, 36 L.Ed.2d 668 (1973). That statistical evidence provides some indication of a pattern of salary disparity between men and women who áre comparable with respect to certain job qualifications. See pp. 1160-1163 infra. 10 The court need not decide at this point whether those statistics constitute a prima facie case of class-wide salary discrimination, but merely holds that they are sufficient to “support an inference that there exists a class in need of protection.” Warren v. I.T.T. World Communications, Inc., 95 F.R.D. 425, 430 (S.D.N.Y.1982); see Paxton v. Union National Bank, 688 F.2d 552, 562 (8th Cir.1982), cert. denied, 460 U.S. 1083, 103 S.Ct. 1772, 76 L.Ed.2d 345 (1983), quoting Donaldson v. Pillsbury Co., 554 F.2d 825, 830 (8th Cir.), cert. denied, 434 U.S. 856, 98 S.Ct. 177, 54 L.Ed.2d 128 (1977) (typicality provision of Rule 23 requires a showing that “ ‘other members of the class have the same or similar grievances as the plaintiff.’ ”). 11

Plainly, given the variety of job titles at O & M, Zukofsky’s salary claim will present some issues which are unique to her. As this court has noted, however,

the fact that the jobs performed by the named plaintiffs are, in some sense, unique, is not a bar to their being class representatives. If it were, no class of professional employees could ever be certified. See, e.g., Lo Re v. Chase Manhattan Corp., 431 F.Supp. 189, 196-97 (S.D.N.Y.1977)____ Moreover, “[i]t is not necessary that each and every issue be raised by each and every member of the class or class representatives.” Vulcan Society v. Fire Dept. of City of White Plains, 82 F.R.D. 379, 401 (S.D.N.Y.1979).

Meyer v. Macmillan Publishing Co., 95 F.R.D. 411, 414 (S.D.N.Y.1982). As the Fifth Circuit stated in analyzing the General Telephone v. Falcon decision and approving the certification of hourly service workers as representatives of a class. including salaried clerical workers, Rule 23 “does not necessarily require a ‘congruence’ that a named plaintiff possess the employment qualifications of all purported class members.” Carpenter v. Stephen F. Austin State University, 706 F.2d 608, 617 (5th Cir.1983); see Paxton v. Union National Bank, supra, 688 F.2d at 562 (“Typicality is not defeated because of the varied promotional opportunities at issue, or the differing qualifications of the plaintiffs and the class members”).

In addition to meeting the four requirements of Rule 23(a), a putative class must also meet one of the three standards set forth in Rule 23(b). 12 That Judge Brieant certified the class pursuant to Rule 23(b)(3) is evident from his order directing notice to class members of their opportunity to “opt out” of the class. See Fed.R.Civ.P. 23(c)(2). 13 The court finds no basis for reversing Judge Brieant’s implicit finding of the predominance of the common questions concerning the class salary claims and of the superiority of class action treatment of those claims. Accordingly, the court concludes that Zukofsky has met the Rule 23 requirements with regard to the class salary claims and appropriately was certified as class representative for those claims.

B. Transfer

Zukofsky also asserts a claim of discriminatory denial of transfer on behalf of herself and unnamed class members. Those transfer claims pertain to the ability of an O & M employee to move from a certain job title in one department to a different job title in another department or sub-department which entails work of a different nature. The transfer claims are to be distinguished from the promotion claims which center on the individual’s ability to move to higher level positions in the same department involving work of generally the same nature.

O & M argues that Zukofsky’s transfer claim is based primarily on a 1975 copy test evaluated by the creative department and therefore is a highly individualized claim not susceptible to class action treatment. Zukofsky contends that it was not simply O & M’s evaluation of her copy test which prevented her from transferring into other, presumably more desirable, departments. Zukofsky points to O & M’s failure to post jobs or to maintain a formal, centralized list of employees requesting transfers, and its failure to make such a list accessible to management considering filling vacancies or new positions. Finally, Zukofsky points to the lack of objective criteria for determining when an employee would receive, or even be eligible for, a transfer. Zukofsky contends that those policies and practices served to deny transfers to her and to unnamed class members in a similarly discriminatory fashion.

Zukofsky thus has raised a number of issues which are common to her transfer claims and to those which might be asserted by other professional women at O & M. That showing of commonality is not necessarily defeated by those factors peculiar to Zukofsky’s own transfer claim. See p. 1131 supra. The question remains whether, under the analysis of General Telephone v. Falcon, supra, and its progeny, Zukofsky has established the typicality of her transfer claim, i.e., that there are unnamed class members “in need of protection” with respect to that issue. See pp. 1131-1132, supra.

Zukofsky introduced no statistical studies directly considering the question of discrimination in transfers at O & M. 14 Instead, as part of their expert’s initial report, plaintiffs proffered a breakdown by year of the numbers of male and female professionals having each job title within certain job families at O & M. 15 Plaintiffs also have indicated by year, for each job family, the average salaries for men and women, respectively, within the encompassed job titles. Plaintiffs’ breakdown of O & M professional jobs indicates that during most of the period from 1975 to 1979, women appeared in the lower ranks of a number of major departments in percentages greater than their overall representation in the O & M professional workforce, which ranged from 46.4% female in 1975 to 55.7% female in 1979. 16 Similarly, in departments such as account management and creative, women occupied a lower percentage of high-level positions than their overall percentage at O & M. 17

In Grant v. Bethlehem Steel Corp., 635 F.2d 1007, 1018 (2d Cir.1980), cert. denied, 452 U.S. 940, 101 S.Ct. 3083, 69 L.Ed.2d 954 (1981), the Second Circuit noted that “a prima facie case [of race discrimination in violation of Title VII] may be made by showing that blacks are concentrated in the ‘lower paying, less desirable jobs ... and were therefore discriminated against with respect to promotions and transfers.’ Teamsters v. United States, supra, 431 U.S. [324] at 329, 337-38 [, 97 S.Ct. 1843 at 1851, 1855, 52 L.Ed.2d 396].” Thus the above-described figures arguably could suggest that unnamed class members in the instant case seek , to raise claims of discriminatory transfers.

Further examination of plaintiffs’ gender breakdown of O & M job families, however, does not support the inference that Zukofsky’s transfer claims are shared by other class members. It has been apparent from the early stages of this action that although it is O & M’s policy initially to look within its own ranks to fill any vacancy, transfers between job titles occur primarily at'the entry level, because the job skills for higher-level positions are learned largely by experience within the same department or job family. Since plaintiffs’ breakdown of O & M’s job classifications shows females to be over-represented in those entry level positions, those statistics do not suggest that there exists a class of women alleging discriminatory denial of transfers into those positions.

Plaintiffs, moreover, have not named any woman other than Zukofsky alleged to have been denied a transfer discriminatorily. No other class member testified that she believed she was denied a transfer on the basis of her sex or that O & M’s transfer procedures had a disproportionately negative impact on her because of her gender. 18

The court concludes that even after trial of the individual and class claims, plaintiffs have failed to produce any evidence that Zukofsky’s transfer claim is typical of claims which would be raised by unnamed class members. See Warren v. I.T.T. World Communications, Inc., supra, 95 F.R.D. at 429. The plaintiff class therefore must be decertified with respect to the claim of discriminatory transfer. 19

C. Promotion

Zukofsky also seeks to represent unnamed class members alleging discriminatory denial of promotion. O & M argues, in essence, that because Zukofsky has failed to establish an individual case of promotion discrimination, she presents no factual or legal issues common to promotion claims which might be asserted by unnamed class members. The court agrees that although Zukofsky’s individual promotion claim is alleged in the complaint, it has been undisputed from the commencement of this action that Zukofsky did not make any effort to obtain, nor did she even want a promotion, that is, a higher position within her own department. Zukofsky thus has never introduced any evidence in support of an individual promotion claim which might be typical of the promotion claims of unnamed plaintiffs.

Zukofsky’s failure to advance an individual promotion claim, however, is not necessarily fatal to her ability to represent unnamed plaintiffs alleging discriminatory denial of promotion. In the instant case, plaintiffs claim that discrimination in both transfers and promotion is manifested by or results from .the absence of formal job descriptions and objective experience and educational requirements for particular job titles. Plaintiffs also argue that 0 & M’s failure to post job openings and its failure to keep a centralized list of those employees seeking other positions within the agency serve to discriminate against females seeking promotions as well as females seeking transfers.

In General Telephone v. Falcon, supra, the Supreme Court remanded the action for a determination whether the named plaintiff, individually claiming promotion discrimination, could represent unnamed job applicants alleging discriminatory failure to hire'. The Supreme Court gave examples of instances in which such representation would be permissible:

If petitioner used a biased testing procedure to evaluate both applicants for employment and incumbent employees, a class action on behalf of every applicant or employee who might have been prejudiced by the test clearly would satisfy the commonality and typicality requirements of Rule 23(a). Significant proof that an employer operated under a general policy of discrimination conceivably could justify a class of both applicants and employees if the discrimination manifested itself in hiring and promotion practices in the same general fashion, such as-through entirely subjective decisionmaking processes. In this regard it is noteworthy that Title VII prohibits discriminatory employment practices, not an abstract policy of discrimination. The mere fact that an aggrieved private plaintiff is a member of an identifiable class of persons of the same race or national origin is insufficient to establish his standing to litigate on their behalf all possible claims of discrimination against a common employer.

Id. at 159 n. 15, 102 S.Ct. at 2371 n. 15 (emphasis in original).

Defendants argue, in effect, that the footnote of General Telephone v. Falcon, supra, quoted above creates two separate tests for determining when an individual challenging one employment practice can represent a class of individuals challenging another practice by the same employer. The determination of which test is appropriate, 0 & M contends, depends upon whether the named plaintiff and putative class members allege claims under the disparate impact theory or the disparate treatment theory. 20 0 & M argues that both Zukofsky’s individual transfer claim and the class promotion claims allege disparate treatment. 0 & M contends that under those circumstances footnote 15 of General Telephone v. Falcon, supra, requires plaintiffs to show “significant proof of a general policy of discrimination” before Zukofsky can serve as representative of a promotion class.

The court finds it unnecessary to determine whether such proof is the sole means by which the commonality and typicality requirements can be established by a named plaintiff complaining of one employment practice who seeks to represent unnamed plaintiffs alleging disparate treatment with respect to another employment practice by the same defendant. Because the class in this case was certified pursuant to Rule 23(b)(3), the class representative must establish not only the commonality and typicality requirements of Rule 23(a), but also must demonstrate that the common factual or legal questions predominate over any questions affecting only individual members of the putative class. See note 12, supra, at 1132; see generally 3B Moore’s Federal Practice P3.45[2]. The court finds that the factual and legal issues common to Zukofsky’s transfer claim and the class promotion claims, see p. 1135, supra, do not predominate over those issues unique to the transfer claim and the promotion claims, respectively. The evidence establishes that in evaluating a candidate for promotion, 0 & M relies primarily on that individual’s performance in his or her present job within the particular department or job family. In contrast, the-performance of an applicant for transfer in his or her present position at 0 & M frequently bears little relation to that individual’s likely success in an entry level position in a completely different department or job family. Accordingly, the most significant question with regard to the promotion claims would be 0 & M’s allegedly discriminatory system for evaluation of a candidate’s performance in his or her present position. Zukofsky’s transfer claim, on the other hand, centers not upon the fairness of the evaluation of her work in broadcast operations, but rather upon the allegedly discriminatory barriers which prevented her entry into areas requiring different job skills. 21 The court concludes that on the basis of her transfer claim, Zukofsky may not serve as the representative of a Rule 23(b)(3) class alleging discriminatory promotions. Cf. Lilly v. Harris-Teeter Supermarket, 720 F.2d 326, 333-34 (4th Cir.1983), cert. denied, — U.S. -, 104 S.Ct. 2154, 80 L.Ed.2d 539 (1984) (named plaintiff alleging discriminatory discharge not appropriate representative of Rule. 23(b)(2) class alleging discriminatory promotions, where promotion claims raised “significant issues of proof separate from those ... implicated by the terminations claim.”) 22

D. Placement

The court considers next plaintiffs’ claim of discriminatory placement. Zukofsky asserts that because of her gender, 0 & M hired her into broadcast operations, an-overwhelmingly female department allegedly characterized by poor pay and low mobility. Zukofsky contends that unnamed class members similarly were placed into largely female and less desirable departments at 0 & M.

0 & M argues initially that Zukofsky cannot represent a class claiming discriminatory placement because she was “placed” in broadcast operations upon her hire in 1970 and the relevant limitations date is May 30, 1975. 23 That argument, however, misconstrues the nature of Zukofsky’s claim. In her placement claim, Zukofsky objects not only to her initial hiring into the broadcast operations department, which was largely at her own behest, but to 0 & M’s alleged maintenance of broadcast operations and numbers of other departments or sub-departments as largely sex-segregated units. Cf. Lewis v. Tobacco Workers’ International Union, 577 F.2d 1135 (4th Cir.1978), cert. denied, 439 U.S. 1089, 99 S.Ct. 871, 59 L.Ed.2d 56 (1979) (allegations that blacks continuously assigned to less desirable department to maintain its racial composition are cognizable under Title VII). 24 The discriminatory segregation allegedly suffered by Zukofsky purportedly results not only from O & M’s decision to hire Zukofsky into broadcast operations, but also from the pattern of hires into or transfers out of that unit.

Zukofsky thus is asserting a continuing violation of Title VII, that is, “persistent discriminatory conduct in the context of ongoing employment relations, e.g., discriminatory layoffs and failure to recall ... or refusals to promote____” Carter v. Delta Air Lines, Inc., 441 F.Supp. 808, 812 (S.D.N.Y.1977) (citations omitted). Such a continuing violation is distinct from the mere continuing impact or effect of a single employment decision prior to the relevant limitations date. See United Air Lines, Inc. v. Evans, 431 U.S. 553, 558, 97 S.Ct. 1885, 1889, 52 L.Ed.2d 571 (1977); Malarkey v. Texaco, Inc., 559 F.Supp. 117, 120-21 (S.D.N.Y.1982), aff'd, 704 F.2d 674 (2d Cir.1983). The limitations issue' therefore does not preclude Zukofsky from asserting her individual placement or sex-segregation claim and from serving as representative of unnamed class members asserting similar claims. See note 16, supra at 1133.

Zukofsky’s individual placement/sex segregation claim, and such claims by unnamed class members necessarily share predominating common legal and factual issues. Central to all such claims is the proof of the existence of sex-segregated departments. Zukofsky, in asserting that the gender composition of her department reflected bias and not just chance, has adduced evidence regarding other departments, and sub-departments allegedly reserved for women. Specifically, plaintiffs also have introduced evidence suggesting that certain job families, or groupings of 0 & M job titles are largely or exclusively female. 25 Those families include accounting (billing)' which, for example, was 87% female in 1977; broadcast production services (business affairs), 100% female in 1977; broadcast production services (talent payments), 86% female in 1977; marketing services (local broadcasting), 86% female in 1977; marketing services (market research), 81% female in 1977;' and personnel, 83% female in 1977.

The court finds such evidence sufficient to permit the inference that Zukofsky’s placement claim is typical of those which might be asserted by unnamed plaintiffs in certain job families. The placement class as presently defined, however, is far too broad since many of its members are employed in job families in which, according to plaintiffs’ own evidence, men are also well-represented. Accordingly, the definition of the class with respect to discriminatory placement/sex segregation must be narrowed to include only those members of the class, as originally defined, employed in job families having five or more employees which, for the years from 1975 to 1979, inclusive, were more than 80% female (“Class II”). 26

E. Training

Finally, Zukofsky claims that 0 & M has discriminated with regard to the training offered her and that offered unnamed class members. Title VII prohibits discrimination on the basis of gender in the administration of training programs. See Batyko v. Pennsylvania Liquor Control Board, 450 F.Supp. 32, 35 (W.D.Pa.1978). Zukofsky has not suggested, through allegation or evidence, how 0 & M discriminated against her with regard to training opportunities. For example, she has not pointed to any 0 & M training program where she or other women employees were rejected or discouraged from participation.

The evidence does indicate that most employees at 0 & M receive on-the-job training because their superiors explain tasks and supervise job performance. Within the account management department, the nature of the on-the-job training appears to vary according to the nature of the account, that is, of the product or service being advertised. Plaintiffs have alleged that 0 & M discriminatorily denied women opportunities to work on certain types of. accounts in which account management staff receive purportedly more valuable on-the-job training. There is no suggestion, however, that training discrimination resulted from differential account assignment within Zukofsky’s department, or in any other manner which affected her individually.

Zukofsky thus has not suggested any issues common to her training claim and those of unnamed class members other than the broad question of whether 0 & M discriminated- against women with regard to training. As indicated in the discussion of General Telephone v. Falcon, see pp. 1129-1130, supra, such a showing is insufficiently specific to meet the requirements of Rule 23. Neither has Zukofsky alleged any factual or legal issues shared by her other individual claims and the training claims of the unnamed plaintiffs. The court concludes that the class alleging discrimination in training opportunities must be decertified. 27

II. Rossini’s Individual Claims

Rossini alleges that throughout her career at 0 & M, the agency has discriminated against her with respect to salary, training, account assignments, and promotion. Rossini further contends that O & M limited her employment opportunities in retaliation for her having filed an EEOC charge and this action. The court will consider first the allegations of discriminatory denial of training opportunities, since O & M alleges Rossini’s lack of on-the-job experience in certain areas as a defense to many of her claims. The court next will consider together the purportedly discriminatory assignments and promotions, since the evidence establishes that promotions in the account management department frequently resulted from lateral assignment to an account likely to require higher level staffing. Finally, the salary and retaliation claims will be discussed.

A. Burdens of Proof and Production for Training and Assignment/Promotion Claims

Rossini’s training, assignment and promotion claims are asserted under the doctrine of disparate treatment, that is, she contends that 0 & M deliberately denied her those employment opportunities because of her gender. See footnote 20, supra. In McDonnell Douglas Corp. v. Green, supra, 411 U.S. at 802, 93 S.Ct. at 1824, the Supreme Court, in the context of an allegedly discriminatory refusal to hire, stated that a plaintiff could establish a prima facie case of disparate treatment in violation of Title VII by demonstrating:

(i) that he belongs to a racial minority; (ii) that he applied and was qualified for a job for which the employer was seeking applicants; (iii) that, despite his qualifications, he was rejected; and (iv) that, after his rejection, the position remained open and the employer continued to seek applicants from persons of complainant’s qualifications.

Those standards have been applied to allegations of discriminatory denial of promotion or of training opportunities. See Betts v. Sperry Division of Sperry Rand Corp., 556 F.Supp. 562 (E.D.N.Y.1983); Adams v. Gaudet, 515 F.Supp. 1086, 1095 (W.D.La.1981); Brazer v. St. Regis Paper Co., 498 F.Supp. 1092, 1097-98 (M.D.Fla.1980).

Once the plaintiff has established a prima facie case, the defendant employer then assumes the burden of production, that is, of articulating, through admissible evidence, a “clear and reasonably specific” legitimate, non-discriminatory reason for disfavoring plaintiff. See Texas Department of Community Affairs v. Burdine, supra, 450 U.S. 248, 255 & n. 9, 101 S.Ct. 1089, 1094 & n. 9, 67 L.Ed.2d 207 (1981). However, “[w]here the defendant has done everything that would be required of him if the plaintiff had properly made out a prima facie case, whether the plaintiff really did so is no longer relevant.” United States Postal Service v. Aikens, 460 U.S. 711, 715, 103 S.Ct. 1478, 1482, 75 L.Ed.2d 403 (1983). Thus, if the defendant has met its burden of production, the court need not determine whether plaintiff has established a prima facie case, but rather must proceed to determine the ultimate question of whether “ ‘the defendant intentionally discriminated against the plaintiff.’ ” Id. (quoting Texas Department v. Burdine, supra, 450 U.S. at 253, 101 S.Ct. at 1093).

With regard to the Title VII defendant’s burden of production, the Second Circuit recently stated:

Moreover, the evidence produced by the employer [to rebut a prima facie case] should be objective and competent. Subjective evaluations are not adequate by themselves because they may mask prohibited prejudice. Knight v. Nassau County Civil Service Commission, 649 F.2d 157, 161 (2d Cir.), cert. denied, 454 U.S. 818 [102 S.Ct. 97, 70 L.Ed.2d 87] (1981); see Crawford v. Western Electric. Co., 614 F.2d 1300, 1313-17 (5th Cir.1980); United States v. N.L. Industries, Inc., 479 F.2d 354, 372 (8th Cir.1973). Objective competent evidence may consist of the required qualifications for the subject position measured against plaintiff’s qualifications including background in the relevánt field. Knight, 649 F.2d at 1611. Such criteria do not become subjective simply because they are evaluated by plaintiff’s' superiors. Id.

Sweeney v. Research Foundation of the State University of New York, 711 F.2d 1179, 1185 (2d Cir.1983). Once a Title VII defendant has made such a showing, the plaintiff, who retains the burden of persuasion throughout the proceeding, then must “demonstrate that the proffered reason was not the true reason for- the employment decision ... either directly by persuading the court that the discriminatory reason more likely motivated the employer or indirectly by showing that the employer’s proffered reason is unworthy of credence.” Texas Department v. Burdine, supra, 450 U.S. at 256, 101 S.Ct. at 1095.

This court has described as follows some of the factors relevant to the issue of pretext:

The reasonableness or lack thereof of the employer’s explanation is probative in this regard. See Loeb v. Textron, Inc., 600 F.2d [1003,] 1012 n. 6 [(1st Cir.1979)]. Thus, “[t]he more idiosyncratic or questionable the employer’s reason, the easier it will be to expose it as pretext.” Id. For example, evidence that the claimant was considerably more qualified for the contested position than the person selected would be probative of pretext because it is unlikely that an employer would exercise its good faith business judgment in such a manner. See Aikens v. United States Postal Service Board of Governors, supra, 665 F.2d at 1959-60; Lieberman v. Gant, 630 F.2d 60, 67-68 (2d Cir.1980). Conversely, the lack of a significant discrepancy between the qualifications of the claimant and the successful applicant would not be probative of pretext because an employer may choose from among several qualified candidates without violating [the equal employment laws]. See Burdine, supra, 450 U.S. at 259 [101 S.Ct. at 1096], Murphy v. Middletown Enlarged City School District, 525 F.Supp. 678, 693, 707 (S.D.N.Y.1981). Another example of pretext would be if the employer claimed it did not select an employee for a particular position because its records indicated he had not applied, and plaintiff adduced credible evidence that the employee did apply.

EEOC v. Trans World Airlines, Inc., 544 F.Supp. 1187, 1220 (S.D.N.Y.1982). 28 Evidence of pretext also may include statistical evidence regarding the defendant’s workforce. See p. 1131 supra; Rodriguez v. Board of Education of Eastchester Union Free School District, 620 F.2d 362, 367 (2d Cir.1980).

B. Training

Before the court reaches the consideration of specific assignments and promotions allegedly denied Rossini because of her gender, it must consider Rossini’s claim of discriminatory denial of the on-the-job training that an account management employee receives by virtue of his or her assignment to a particular account. 0 & M, in defending its assignment and promotion decisions, frequently cites the experience of the chosen employee on product accounts similar to the account at issue. In particular, O & M has asserted that prior successful packaged goods experience is relevant to an account management employee’s success on a new packaged goods assignment. Rossini does not dispute that experience on similar accounts is a criterion relevant to account management staffing decisions. She alleges that because the best account assignments, including packaged goods, have been denied to women, consideration of such experience perpetuates past discrimination.

Initially, 0 & M argues that the statute of limitations precludes consideration of any account assignment prior to May 30, 1975. As discussed note 16, supra at 109, however, liability in this action may be premised upon discriminatory employment practices occurring before that date, if such practices form part of a continuing pattern of discriminatory conduct. Moreover, even if account assignments prior to that date are not part of such a pattern, they still may be considered as “relevant background evidence” to the allegedly discriminatory practices appropriately before the court. United Air Lines, Inc. v. Evans, supra, 431 U.S. at 558, 97 S.Ct. at 1889. The court concludes that, at a minimum, it may consider information regarding earlier assignments as background evidence relevant to Rossini’s post-May 1975 assignment/promotion claims.

Claims of disparate treatment in training opportunities, such as that raised by Rossini, have been analyzed under the McDonnell Douglas model. See Brazer v. St. Regis Paper Co., supra, 498 F.Supp. at 1097-98 (M.D.Fla.1980). In the instant case, 0 & M claims that prior to 1975, it assigned Rossini to non-packaged goods accounts largely because those accounts became available at times when Rossini needed additional responsibilities. As the Second Circuit emphasized in Sweeney v. Research Foundation, supra, 711 F.2d at 1187 n. 11, in the absence of evidence of discrimination, “an employer must have freedom to run its organization efficiently. While the employee’s wishes should be taken into account in the employer’s decision-making, they do not govern.”

Rossini has cited Connecticut v. Teal, 457 U.S. 440, 102 S.Ct. 2525, 73 L.Ed.2d 130 (1982) and argued that even if 0 & M favored some women in, for example, the assignment of accounts with packaged' goods training potential, the agency nevertheless is not relieved from liability for its alleged discrimination against her in that area. In Teal, the Supreme Court held that a racially-balanced work force did not relieve an employer from liability to a minority employee adversely affected by a test shown to have disparate impact on minorities. The Supreme Court noted that Teal involved a disparate impact claim which did not require a showing of intent. In a disparate treatment case, such as Rossini’s training claim, in contrast, “an employer’s good faith efforts to achieve a nondiscriminatory work force, might ... assist an employer in rebutting the inference [necessary to liability] that particular action had been intentionally discriminatory....” Id. at 454, 102 S.Ct. at 2535. The court, therefore, finds that evidence of gender-neutral assignments of accounts to other O & M staff is relevant to the issue of the allegedly discriminatory denial of packaged goods experience to Rossini.

Testimony by a number of female account management employees therefore lent support to O & M’s argument that it did not discriminatorily deny Rossini packaged goods experience prior to 1975. Ann Iverson, Rochelle Lazarus, and Sharon McGavin all testified to their work on O & M packaged goods accounts in or before 1976. Although Rossini submitted figures indicating other women working in account management from 1970 to 1976, she adduced no other examples of women discriminatorily denied packaged goods experience. In addition, several men who began work at O & M around the time Rossini did and with whom she compares herself spent a significant portion of their time on non-packaged goods accounts. Bob Haller, for example, worked on the Sears home appliances and home entertainment accounts. Henry Ferris, after an initial assignment to Maxwell House and Maxim coffees, took on responsibility for the Rex Restaurant account. In April 1971, Ferris began work on the Cunard Lines account where he continued for the next several years. Bob White spent most of his first ten years at O & M on accounts such as Sears, Shell Oil, and Cotton, Inc.

Moreover, Rossini herself noted in a December 1975 memorandum to Francis Houghton, then the personnel director at O & M, that the agency’s “non-package[d] goods clients outnumber those in package[d] goods by almost 3 to 1.” Those figures suggest a non-discriminatory basis for Rossini’s assignments to non-packaged goods accounts, that is, that O & M simply had more available positions on non-packaged goods accounts. She also stated (emphasis added) “[w]e all agree that my best prospects are in the non-package[d] goods accounts...,” thereby indicating, at least to some degree, her accession in O & M’s assignment decisions.

O & M also alleges Rossini’s poor performance in 1975 on the Drain Power account as the agency’s reason for denying her thereafter packaged goods work and the concomitant on-the-job training. Allen Clark, a management supervisor on Drain Power’s parent account, Glamorene, rated Rossini’s performance on Drain Power from October 1974 through October 1975 as category 4, “marginal performance in present assignment,” although she had received higher evaluations in all her previous work at 0 & M. Clark wrote that Rossini lacked “numerical ability. In turn, her knowledge of package goods marketing is weak. Further, she had great difficulty retaining the key facts of our client’s market and brand (e.g. Nielsen, [a survey of consumer preference] research). This in turn affected the quality of her judgment in related areas.”

Rossini attacks as suspect the subjectivity of such an evaluation. “It is true that an employer may not use wholly subjective and unarticulated standards to judge employee performance for purposes of [job advancement].” Knight v. Nassau County, supra, 649 F.2d at 161. The standards applied by Clark, however, while necessarily evaluated in a subjective manner, are not so idiosyncratic or unclear as to deprive Rossini of a fair opportunity to rebut the charges of her inadequate performance.

Rossini has introduced evidence which she claims indicates her skill in the numerical aspects of account management, thereby purportedly demonstrating the pretextual nature of Clark’s criticism. She also argues that much of Clark’s criticism simply masked his disagreement with her strategy to emphasize the safety features of Drain Power, a strategy which if implemented, Rossini contends, might have saved the failing product.

Specifically, Rossini cites a November 1975 letter she wrote to Glamorene reporting 0 & M’s revised Nielsen projections for Drain Power and other drain cleaners. Rossini argues that a high level account management employee noted on that letter that Clark had rejected any changes in its format. Rossini contends that the indication shows that Clark had approved the letter and that his subsequent criticism of her ability to analyze such data was a pretext for sex discrimination. It appears from the notation on the November 1975 letter, however, that Clark simply had agreed with Rossini’s lengthy recitation of the figures, but not that he thereby had approved whatever analysis of data the letter contained. Moreover, William Weed, Clark’s supervisor on the Glamorene account, testified that the November 1975 letter did not reflect Rossini’s analysis or “viewpoint” with regard to the Nielsen study or a safety strategy for the product. Weed cited the last paragraph of the letter: “Hopefully, our startup of advertising ... with the guarantee strategy, combined with your promotional activities, will deter further losses____” He concluded that an appropriate letter recommending a change in strategy “would take quite a different form____ Changing the strategy on a brand is a major issue, and [the letter] would start out by saying that that’s what it was undertaking to do, and it would make a specific recommendation to that effect and present the supporting evidence.” The court concludes that Rossini has not established the November 1975 letter to be an example of the numerical/analytical skills which Clark contended she lacked.

Rossini also submitted her April 1976 letter to Glamorene discussing “DART tests,” a statistical report of consumer response to drain cleaner commercials. That letter was graded “A” in a notation by Charles Fredericks, Clark’s supervisor at that time. Rossini again contends that such a notation by a high level account management official proves that she actually did possess the skills in question. However, there was no testimony by Fredericks indicating exactly what he was approving with the “A” grade. Weed testified that the April 1976 letter touched on strategy only in . the summary and conclusion. He added:

A. And the last sentence says, “We hope,” — again, it’s hope, not fact or conviction, “We hope the upcoming split run copy test will give us a definitive answer.”

So it really relies on the upcoming copy test to give an answer on safety. That’s the only time we see “safety” mentioned.

Q. Both letters [the November 1975 and the April 1976] concluded on the “hope” note?

A. Yes.

Q. Is that the custom? Is that the form that these letters usually take?

A. No.

Q. Would you explain that?

A. Well, I guess we’re always hoping for things to get better, but we usually try to anticipate the problem so that we can make certain that they improve and recommend action for improvement or correction.

The record, , therefore, provides the court with no basis for evaluating the April 1976 letter as a good example of numerical/analytical account management skills. Weed’s testimony again suggests the weakness of Rossini’s proposals for a safety strategy, thereby undermining her contention that those recommendations were the actual basis of Clark’s criticism.

Rossini cites two other examples of her numerical/strategic skills, which she contends reflect the pretextual nature • of Clark’s criticism. In 1975, she wrote two memoranda to Clark citing certain consumer research statistics and urging ads for Drain Power with an emphasis on safety. However, Rossini has not provided the court with any basis for judging whether those memoranda are good examples of strategic use of advertising statistics, or for determining Clark’s reaction to her proposed focus on safety. Rossini herself did not explain the purported merits of those memoranda. Neither did she offer testimony from any other individual experienced in account management who would explicate her bald assertion of the significance of those documents. The court finds that the evidence does not establish that Clark’s evaluation of Rossini’s numerical analytical skills was erroneous and therefore merely a pretext for discriminatory denial of packaged goods work to Rossini.

Rossini also introduced evidence of a 1981 “Effy,” an award conferred by the advertising industry in recognition of effective marketing. Rossini thus suggests that regardless of the earlier criticism of her skills, she later showed herself to be accomplished in all aspects of account management. The evidence, however, did not conclusively establish whether that Effy, for the Amex “assured reservations campaign,” was given to Rossini for her individual accomplishment or to 0 & M in recognition of contributions by all of the Amex staff. Neither does the record establish the nature of Rossini’s role in the commended advertising campaign.

Finally, Rossini cites a number of documents prepared in 1982 for the marketing of a new financial services division of Amex. Prudence Piggott and Bruce Car-lisle, two other 0 & M employees on that account testified that the documents, introduced to show Rossini’s strategic marketing abilities, actually were drafted primarily by Piggott.

The court concludes that Rossini has not established by a preponderance of the evidence the pretextual nature of 0 & M’s criticism of her numerical/analytical skills, abilities undisputedly relevant to packaged goods advertising’. 29 Neither has she carried her burden generally with respect to establishing the discriminatory denial of packaged goods training experience.

C. Assignment and Promotion Claims

(1) General Issues

In the context of the McDonnell Douglas framework, the parties have raised a number of legal issues relevant to the court’s analysis of the promotion/assignment claims. The court will discuss those issues before addressing the particular positions in dispute.

(a) Title VII Coverage of Lateral Assignments

First, 0 & M asserts that lateral assignments to specific accounts represent simply the employer’s decisions regarding staffing patterns and argues that “the judiciary certainly ought not to have to determine why each particular account assignment was made.” That argument ignores the broad scope of Title VII which extends beyond promotions, salary, hiring and other practices with an immediate financial impact on the alleged discriminatee. Title VII prohibits employment practices which discriminatorily “limit, segregate, or classify ... employees ... in any way which would deprive or tend to deprive any individual of employment opportunities.” 42 U.S.C. § 2000e-2(a)(2); see footnote 22, supra. Since account assignments in the account management department unquestionably were related to employment opportunity, the court must look at .the challenged assignments not to determine why they were made, but rather to determine whether they were made on the basis of gender. See Texas Department v. Burdine, supra, 450 U.S. at 259, 101 S.Ct. at 1096 (under Title VII, “the employer has discretion to choose among equally qualified candidates, provided the decision is- not based on unlawful criteria”).

(b) Need for Proof of Application

0 & M also argues that the court may consider the denial of assignment or promotion only with respect to those positions for which Rossini actually applied in a timely fashion. Rossini counters that because 0 & M does not post job openings, she could not have been aware of all the positions for which she might have been qualified but which she did not receive as a result of 0 & M’s alleged discrimination.

In Teamsters v. United States, supra, 431 U.S. at 365, 97 S.Ct. at 1869, the Supreme Court held that given a prima facie showing of classwide discrimination, an individual plaintiff could not be denied relief simply because he had failed to apply for positions routinely denied to other minority group members. In Grant v. Bethlehem Steel Corp., supra, 635 F.2d at 1017, the Second Circuit found that black construction workers had made a prima facie case of discriminatory promotion denial, despite their failure formally to apply for every opening át issue.

Each of the three named appellants clearly and repeatedly made his interest in a job as foreman known to at least one of the superintendents. This was sufficient to put the superintendents on notice that these men wanted a foreman’s job. Under Title VII “a nonapplicant can be a victim of unlawful discrimination ... when an application would have been a useless act serving only to confirm a discriminatee’s knowledge that the job he wanted was unavailable to him.” Int’l Brotherhood of Teamsters v. United States, supra, 431 U.S. at 367 [97 S.Ct. at 1870], Dothard v. Rawlinson, 433 U.S. 321 [97 S.Ct. 2720, 53 L.Ed.2d 786] (1977). Faced with an admittedly entrenched discriminatory system that had historically shown no inclination to make blacks foremen, appellants were not required to keep beating their heads against the wall by reapplying.

The court finds that given the failure of O & M to post job openings and the possibility of discrimination resulting therefrom, it is appropriate to consider certain positions for which Rossini did not make a timely formal application. See Rich v. Martin Marietta Corp., 522 F.2d 333, 347-48 (10th Cir.1975). If Rossini alerted the relevant O & M account management personnel to her interest in a particular account a reasonable period of time before an appropriate position on that account became available, that position can be included as part of Rossini’s claim.

(c) Adequacy of Notice at Trial of Disputed Assignments/Promotions

O & M next argues that the court should consider as part of Rossini’s assign ment/promotion case only those opportunities raised by her at trial. The court agrees and indicated during Rossini’s testimony that such a limitation was necessary to assure that 0 & M had the opportunity to articulate by testimony or documentary evidence a legitimate, non-discriminatory reason why Rossini was denied a particular position. Moreover, that notice requirement imposed no special burden on Rossini; whatever denied positions were cited in her post-trial briefs plainly were known to her and her counsel at the time of trial. The court therefore adheres to its prior ruling that specific consideration of promotions or assignments -denied Rossini will be limited to those opportunities which she raised at trial.

0 & M contends that Rossini’s post-trial brief alleges discriminatory denial of the following opportunities not raised by Rossini at trial:

1. 10/75 Avon — Henry Ferris [received assignment]

2. 11/75 Pepperidge Farm — Charles Guariglia

3. 1/76 Amex — Personal Cards — John Jones

4. 1/76 Amex Travelers Cheques — Ed Vick

5. 2/76 Pepperidge Farm — Earl Bahler

6. 7/78 Mercedes & Uniroyal — Fred Lamparter

7. 5/79 Lever & Shaeffer — John Reed

8. 5/79 Kimberly Clark — Fred Lamparter

9. 1/80 Menley-James — Peter Schaeffer

10. 5/81 Block Drug — Fred Lamparter

Rossini did not testify directly to any interest in the two aforementioned positions on Amex. Those openings in the personal card and travelers checks divisions had become available two months before Rossini was assigned to the service establishment division of the Amex account. At trial, Rossini stated that it was only after she had worked in the service. establishment division of Amex for some period of time that she indicated to her superiors an interest in working on the other portions of that account. The court finds that such a general allusion did not give O & M adequate notice of her intent to challenge the January 1976 Amex assignments.

At trial, Rossini alluded to a Pepperidge Farm management supervisor job available after 1978, and cited her 1972 memo to management expressing a desire to work on Pepperidge Farm. Neither that 1972 memo nor Rossini’s testimony adequately alerted 0 & M to her intention to contest the aforementioned 1975 and 1976 assignments to Pepperidge Farm account supervisor positions.

With regard to Avon, Rossini mentioned at trial two positions. She testified, however, that both of them were filled by women, not by Henry Ferris or any other male employee. In view of the evidence indicating that any account, over a period of time, can have numbers of openings, the court finds that Rossini’s testimony regarding Avon was not sufficient notice that the Avon assignments to Hank Ferris and Fred Lamparter were disputed.

Rossini has not cited and the court has not found any indication at trial thát positions on the Mercedes, Menley-James, or Block Drug accounts were at issue. Accordingly, the court concludes that Rossini is precluded from asserting that any of the ten assignments discussed above reflected discrimination. 30

(d) Statistics as Evidence of Pretext

Finally, the court considers plaintiffs’ breakdown of the number of men and women occupying various job titles in account management. Such statistical evidence regarding the larger' workforce is relevant with regard to motive or pretext in an individual disparate treatment claim. See p. 1131, supra. Moreover, plaintiffs’ figures indicate that females, in proportion to their total numbers at 0 & M, generally are overrepresented in the lower ranks of account management and underrepresented in the upper ranks. See pp. 1133-1134, supra. 31

The Second Circuit in Lieberman v. Gant, 630 F.2d 60, 69 (2d Cir.1980), however, has cautioned the courts regarding the probative nature of such overall workforce statistics in an individual case.

The Court noted [in McDonnell Douglas v. Green, supra, 405 U.S. at 805 n. 19, 93 S.Ct. at 1826 n. 19] that such evidence [of discriminatory intent] might include the ... composition of the employer’s workforce, though adding a caveat, especially pertinent here [in a case alleging discriminatory denial of tenure], about the weight this sort of evidence deserves:

We caution that such general determinations, while helpful, may not be in and of themselves controlling as to an individualized hiring decision, particularly in the presence of an otherwise justifiable reason for refusing to rehire.

The support for Rossini’s case to be gathered from the overall staffing patterns is limited somewhat by her failure to provide some measure of the statistical significance of the apparent underrepresentation of women in upper account management positions, particularly in the management supervisor position which Rossini sought throughout the late 1970’s. Such a measure compares the actual number of women in a particular position with the “expected value” or likely number of women in that position if such assignments were randomly made within the employee population, i.e., without regard to gender. See Hazelwood School District v. United States, 433 U.S. 299, 308 n. 14, 97 S.Ct. 2736, 2742 n. 14, 53. L.Ed.2d 768 (1977); Castaneda v. Partida, 430 U.S. 482, 496 n. 17, 97 S.Ct. 1272, n. 17, 51 L.Ed.2d 498 (1977); Boykin v. Georgia-Pacific Corp., 706 F.2d 1384, 1391 (5th Cir.1983), cert. denied, — U.S. -, 104 S.Ct. 999, 79 L.Ed.2d 231 (1984). 32 Although such an analysis is not always required, see Hazelwood School District v. United States, supra, 433 U.S. at 311 n. 17, 97 S.Ct. at 2743 n. 17, in this instance the raw data concerning the gender composition of the account management hierarchy lend only limited support to her argument that she was denied promotions and assignments as a result of intentional sex discrimination by 0 & M.

Rossini also introduced statistics indicating that on the average men have fewer years of 0 & M experience when they are promoted to account supervisor than do women. Those figures indicate an average of 4.82 years of such experience for women and 2.48 for men. The average O & M tenure of women appointed as vice-president/deputy or management supervisor was 9.6 years for females and 4.91 for males. Again, Rossini has submitted data of limited usefulness because those figures do not reflect the account management employee’s advertising experience prior to 0 & M. Neither do those figures distinguish between types of experience at 0 & M, and accordingly include years of secretarial work or work outside account management at 0 & M as part of the employee’s 0 & M experience prior to appointment as a supervisor in the account management department. The court therefore concludes that those tenure statistics do not contribute significantly to establishing an inference of discriminatory motive by 0 & M which would be relevant to Rossini’s individual case.

(2) Specific positions

Against that background, the court will examine the assignments or promotions Rossini alleges were discriminatorily denied her on or after May 30, 1975. The court must determine whether any of those employment decisions constituted a violation of Title VII and whether they indicate a pattern' of continuing discrimination which could subject 0 & M to liability for actions before May 30, 1975.

The first assignment Rossini contests was the denial of lateral move to an account supervisor position on Sears, vacated by the promotion of Richard Hosp in early 1976 and filled by Arthur Schulbaum in July 1976. 0 & M argues that Rossini had not requested that assignment since it was in December 1975 that she made her last inquiry to upper management regarding an account supervisor position on Sears.

Under the standard described at pp. 1145-1146, supra, the court finds that the December 1975 inquiry was sufficient to alert 0 & M to Rossini’s interest in the position at issue. The court finds, however, that Rossini has not established as part of her prima facie case that she was qualified for the job. See pp. 1139-1140, supra. Specifically, the Sears account supervisor position was at the Chicago office of Ogilvy & Mather International and Rossini did not establish her willingness to go to that office. On the contrary, throughout the trial, Rossini indicated her disinclination to leave New York and her feeling that such a move, even if only temporary, would be a considerable detriment to her professional progress. Even when Rossini was specifically cross-examined about her knowledge of the location of the disputed Sears job, she at no time stated that she maintained any interest after she learned that the account had been moved to Chicago. The court concludes that Rossini has not shown that ‘she was qualified for the July 1976 Sears account supervisor position and therefore that she has not established discriminatory treatment with respect to that assignment.

Within a year of her 1976 assignment as an account supervisor on the service establishment division of the Amex account, Rossini began to request transfer to other divisions of that account in order to broaden her experience. She alleges that in 1977 and 1978, 0 & M assigned two less-qualified men, Russ Rhodes and Peter Schaeffer, to account supervisor positions in the Amex card division, the group responsible for advertising to persuade consumers to obtain and use the American Express card.

0 & M introduced significant evidence of non-discriminatory reasons for the selection of those two men. Russ Rhodes had substantially more education and experience than Rossini. He has a B.A. from Yale and an M.B.A. in marketing from the Harvard Business School. From 1955 to 1960, he worked in advertising at Proctor & Gamble; in 1960 he went to Norman Craig and Kummel and later directed that agency’s European operations with responsibility for its cosmetic, detergent and drug advertising throughout the United States and Europe; from 1969 to 1976, he was self employed in Europe, primarily as a marketing consultant and writer.

While formal degree or experience requirements which are not job-related may mask discrimination, see James v. Stockham Valves & Fittings Co., 559 F.2d 310, 355 (5th Cir.1977), cert. denied, 434 U.S. 1034, 98 S.Ct. 767, 54 L.Ed.2d 781 (1978) (invalidating high school diploma requirements as not job-related), Rossini does not dispute the relevance of an 0 & M employee’s educational background to his or her account management assignments. On the contrary, Rossini has challenged as part of the subjectivity of 0 & M’s promotion procedures the agency’s failure to set fixed education and/or experience requirements for various job titles. Educational, attainment, therefore, is not intrinsically suspect when proffered by 0 & M as one of its reasons for not selecting Rossini for a given assignment or promotion.

Rossini notes that in 1976, a high-level account management supervisor wrote upon the Rhodes’ appointment to an account supervisor position on Avon that Rhodes “would be hard to move to many other accounts.” Rossini has not pointed to any evidence suggesting to be pretextual 0 & M’s conclusion that the Amex card division was one of the few other accounts appropriate for Rhodes.

Rossini also cites the areas of weakness that were noted in a generally favorable 1980 evaluation of Rhodes. Because that evaluation post-dates Rhodes’ 1977 appointment to the contested Amex card division position, it has little probative value with regard 0 & M’s motives in making that appointment. Even if Rhodes’ later evaluations suggest that, in retrospect, his appointment to the Amex card division job may have been a mistaken business decision, such a conclusion, without more, does not indicate 0 & M’s discriminatory motive at the time the appointment was made. Such a situation plainly is distinguishable from one in which an employer chooses a candidate who at the time of the employment decision appears markedly less qualified than the alleged discriminatee. It is in the latter instance only that discriminatory motive at the time of appointment is suggested. See Lieberman v. Gant, supra, 630 F.2d at 67-68; EEOC v. TWA, supra, 544 F.Supp. at 1220.

Neither has Rossini shown to be pretextual O & M’s asserted reasons for the selection of Paul Schaeffer to succeed Rhodes in 1978. Bob White, Rossini’s supervisor on Amex testified that Paul Schaeffer, a Dartmouth graduate, was given the account supervisor position in the Amex card division because Rossini at that time was more useful to O & M by continuing her in the service establishment division. White also cited Schaeffer’s extensive packaged goods experience. Apparently those factors outweighed White’s recommendation on Rossini’s 1977 evaluation that she be given “serious consideration” should Russ Rhodes vacate the top position in the Amex card division. O & M’s professed desire to maintain Rossini in the service establishment division has not been shown to be “unworthy of credence.” Texas Department v. Burdine, supra, 450 U.S. at 256, 101 S.Ct. at 1095.

As discussed at p. 1142, supra, in the absence of evidence of discrimination, the employer retains broad authority with regard to the assignment of personnel. Rossini does not dispute the relevance of Schaeffer’s seven years of packaged goods experience at another major ad agency to the position he filled in the Amex card division. Rossini has not established that O & M discriminated by denying her packaged goods training or by giving her, a poor evaluation of her sole packaged goods assignment. Accordingly, Rossini has not shown to be pretextual O & M’s asserted reasons for Schaeffer’s assignment to the Amex card division.

Rossini next asserts a series of discriminatory denials of assignments to the travelers’ check division of Amex. In February 1978, Earl Bahler was rehired as an account supervisor in that division. O & M argues that Bahler was more qúalified by virtue of, inter alia, his B.A. in economics and his Dartmouth M.B.A. Rossini does not dispute the relevance of those educational credentials to the positions at issue. See p. 1149, supra. The critical evaluation of Bahler’s performance, cited by Rossini, post-dates his appointment to the Amex travelers’ check division and in any event, constituted only a few negative comments in a highly positive overall evaluation. The court concludes that Rossini has not established by a preponderance of the evidence that O & .M acted from discriminatory motives in appointing Bahler and not Rossini to the Amex travelers’ check division.

Rossini also cites as discriminatory the June 1980 appointment of Paul Kurnit as management supervisor on Amex travelers’ checks. Brendan Ryan, a top level supervisor who in 1980 was responsible for coordinating all branches of the Amex account, testified that Kurnit was chosen as “a packaged goods expert” able to handle the demands of a complex financial account. Again, Rossini does not contest the relevance to the position at issue of Kurnit’s experience in packaged goods, including four years at another major advertising agency. She notes only that Brendan Ryan subsequently removed Kurnit from Amex at the client’s request some time after he was appointed to that account. Ryan testified that although there was “bad chemistry between one client [at Amex] and Paul Kurnit, ... he did an excellent job” on the account. The court cannot conclude that those events subsequent to Kurnit’s appointment to management supervisor on Amex travelers’ checks suggest that 0 & M’s initial reliance on his superior packaged goods experience was merely a pretext for discrimination against Rossini. See pp. 1149-1150, supra.

Finally Rossini contends that in November 1980, O & M discriminated against her by naming Mike Néwbrand to a management supervisor’s position in the Amex travelers’ check division. Again Rossini does not dispute Newbrand’s superior packaged goods experience as a management supervisor on the General Foods account nor does she contest the relevance of that experience to the Amex position at issue. 33 Accordingly, the court finds that Rossini has failed to establish by a preponderance of the evidence the pretextual nature of 0 & M’s professed reasons for choosing Newbrand.

Rossini also alleges discrimination with regard to two positions on the TWA account. Ira Berkowitz became an account supervisor on that account in July 1979. O & M has cited Berkowitz’s experience on the airline part of the Amex account. Rossini does not dispute the relevance of that experience to the TWA account, but notes instead that Berkowitz was rated “unsatisfactory” in client relationships and had previously been replaced on the Union Carbide account. The evaluation to which Rossini refers was completed in 1975, and the evaluator added, “We agreed that client’s dissatisfaction was largely unfair in light of the effort Ira has put forth, but that a change in account assignments was necessary under the circumstances.” In view of Berkowitz’s positive ratings in client relations subsequent to 1975 and prior to his appointment to TWA, the court cannot conclude that 0 & M’s articulated reasons for that appointment were a pretext for discrimination. 34

The other challenged TWA appointment occurred in September 1980 when Kelly O’Dea became an account supervisor on that account and soon thereafter became a management supervisor. O & M attributes that choice to O’Dea’s packaged goods experience, training expressly sought by the client as necessary to understanding its business in the wake of airline deregulation. Again, Rossini does not dispute the relevance to the TWA account of that qualification. Instead, Rossini quotes a 1977 evaluation of O’Dea which states: (emphasis in original): “He must now concentrate on improving his skills to lead a brand versus just crossing projects off a' project list.” The quoted remark, however, was preceded by several sentences which indicate that the comment in question was not intended as criticism of O’Dea’s past performance but rather to set a goal for work in his new, more responsible position. The preceding sentences read (emphasis in original): “Kelly has done an excellent job motivating and training his assistants. Kelly is qualified to be a supervisor as evidenced by his recent promotion.” O’Dea’s subsequent evaluations rated him as outstanding or very good in the area of leadership. The court therefore concludes that Rossini has not demonstrated to be pretext 0 & M’s articulated reasons for the contested TWA assignments.

Rossini contends that she was discriminatorily denied a promotion to the management supervisor position on Clairol in October 1980. 0 & M asserts that it chose Earl Bahler for that position largely because he already was serving at the management supervisor level, in contrast to Rossini, who was an account supervisor at that time and would have required a promotion to fill the job. As discussed at p. 1142, supra, in the absence of some evidence of discriminatory intent, an employer can deploy its workforce as it sees fit. Rossini has not demonstrated any indicia of discrimination with regard to 0 & M’s choice for the Clairol account, particularly since Bahler’s predecessor as management supervisor was a woman, Rochelle Lazarus. See p. 1142, supra.

Rossini alleges that 0 & M discriminated against her in January 1981 when it appointed Phillippe Defechereux account supervisor on the Peugeot account and then promoted him to management supervisor on that account in February 1982. 0 & M claims that those employment decisions were based on Defechereux’s marketing abilities, developed on the General Foods packaged goods account; his international background; his fluent French; and his hobby of racing cars. Rossini does not dispute the relevance of those qualifications to the account at issue.

In response, Rossini cites sharp criticism of Defechereux’s relationships with his subordinates, as reflected in a 1977 evaluation from Ken Roman and a 1978 evaluation by Steve Humphrey, both high level account management personnel. In his 1980 evaluation of Defechereux, however, Humphrey wrote:

[Defechereux’s] relationship with his subordinates has improved significantly. His people feel they’re learning a lot and getting good guidance and direction. He lets Account Executives manage projects when they can, but he’s there for help when needed. This achievement is particularly impressive because Phillippe has often been put into the position of training newly promoted Account Executives who are very bright but need a lot of training in basic marketing/advertising skills, writing, and presenting.

Phillippe’s relationship with Creative partners is excellent. People in the Media and Research departments value his support.

The court concludes that Rossini failed to establish by a preponderance of the evidence that 0 & M on the basis of gender picked a less-qualified individual for the Peugeot account.

Rossini testified that sometime after she was assigned to Amex, 0 & M discriminatorily denied her management supervisor positions on Cotton, Inc. and the Calandre women’s fragrance division of the Paco Rabanne account. 0 & M cites efficient use of agency resources as the reason for the appointments to both of those positions. Doug Bruce, who was under-utilized as a management supervisor on other accounts assumed the part time management supervisor duties on Cotton, Inc. without extra cost to the agency. Calandre, as á relatively small account, did not require a separate management supervisor. In the interest of efficiency, 0 & M assigned Calandre to the management supervisor on the rest of the Paco Rabanne account. Rossini has not established by a preponderance of the evidence that those articulated administrative reasons were merely a pretext for discrimination.

Finally, Rossini alleges discrimination with regard to two management supervisor positions on Warner Amex, a cable television company formed by Warner Communications and Amex. Earl Bahler filled one of those positions in March 1980; Ira Berkowitz filled a Warner Amex position as an account supervisor in September 1980 and was promoted to management supervisor on that account in June 1981. Neither in its trial memoranda nor through testimony of its witnesses does 0 & M specifically address the Warner Amex positions. 0 & M therefore has not met its burden of production with regard to Rossini’s claim of discriminatory assignment; the agency has failed to articulate a legitimate reason for the selection of two male employees for those positions. See p. 1140, supra.

Accordingly, the court need consider only whether Rossini has established a prima facie case of discriminatory denial of the Warner-Amex assignments. With regard to the first element of the prima facie case, an application for the position at issue, Rossini testified that she “let it be known” to 0 & M management that she was interested in a number of management supervisor positions, including Warner Amex. For the reasons discussed at pp. 1145-1146, supra, Rossini has established a sufficient application for the Warner Amex management supervisor position which was filled in March 1980. With regard to the Warner Amex position filled in September 1980, Rossini erroneously asserted that it was a management supervisor position. In fact, Ira Berkowitz, who was chosen for that position, began work as an account supervisor. By June 1981 when Berkowitz was promoted to management supervisor, Rossini already was serving as management supervisor on the Knoxville World’s Fair Account. The court concludes that absent any evidence that O & M intended immediately to promote the September 1980 Warner Amex appointee to management supervisor, Rossini’s inquiry regarding management supervisor positions on Warner Amex may not appropriately be regarded as an application for the September 1980 account supervisor slot on that account.

With regard to the March 1980 Warner Amex management supervisor position, Rossini offered no testimony specifically directed to showing as part of her prima facie case that she was qualified for that position. Rather, Rossini concluded generally that within a few years of her appointment to account supervisor, she was capable of filling any management supervisor position with the possible exception of General Foods, which required extensive packaged goods experience. Rossini provided no description of the Warner Amex account or of the nature of the skills required for a management supervisor on that account.

The evidence in this case indicates that supervisory account management positions on different accounts are not fungible, but rather require different skills and experience. See p. 1155, infra. Accordingly, in the absence of more specific evidence regarding the nature of the Warner Amex management supervisory job, and the suitability of Rossini’s particular background and abilities for that position, the court concludes that Rossini has not established that she was qualified for that job. Although the burden of “establishing a prima facie case of disparate treatment is not onerous,” Rossini’s lack of specificity concerning the March 1980 Warner Amex position precludes a finding that she “was rejected under circumstances which give rise to an inference of unlawful discrimination.” Texas Department v. Burdine, supra, 450 U.S. at 253, 101 S.Ct. at 1093. 35

(3) “In Place”Deputy Management Supervisor Promotions

In her trial memoranda, Rossini argues that 0 & M discriminated against her by not including her among sixteen male and female employees given “in place” promotions to deputy management supervisor, with no change of account assignment or, apparently, of salary. The court does not reach the merits of that claim for, like the challenges to a number of account assignments, the issue of the May 1978 in place promotions was not appropriately raised at trial. See pp. 1145-1146, supra. Rather at trial, Rossini testified that in or about December 1979, her supervisor Bob White stated that he wanted- to promote her to deputy management supervisor or full management supervisor. Neither Rossini’s testimony nor anything in her pre-trial submissions suggested that the May 1978 in place promotions were at issue. In the absence of such specification, the court concludes that it would be unfair to require 0 & M to defend those employment decisions.

With regard to the deputy or full management supervisor position for which White purportedly recommended her in or about December 1979, Rossini has not established a prima facie case. She has not introduced evidence suggesting the availability at that time of such openings, other than those previously held to have been accorded to other 0 & M employees on a non-discriminatory basis. Accordingly, the court concludes that Rossini has failed to establish by the preponderance of the evidence that 0 & M discriminated against her in account assignments or promotions. 36

D. Salary

Rossini has charged 0 & M with salary discrimination and alleges generally that since she began work there in 1970, the agency has paid her less than similarly qualified men performing similar work. Her claim must be evaluated in the first instance pursuant to standards which arise from the relationship between Title VII and the Equal Pay Act, 29 U.S.C. § 206(d) (“EPA”). 37 That relationship is discussed at length in County of Washington v. Gunther, 452 U.S. 161, 101 S.Ct. 2242, 68 L.Ed.2d 751 (1981). Gunther interpreted the Bennett Amendment, section 703(h) of Title VII, 42 U.S.C. § 2000e-2(h), which incorporates as a defense to Title VII claims of sex-based salary discrimination, the defenses to an EPA claim. In that case, the Supreme Court ruled that although EPA claims are limited to allegations that a member of one gender is being paid more than a member of the. other gender for performing the same work the salary discrimination prohibited by Title VII is not so circumscribed.

Title VII claims of gender-based salary discrimination are not limited to unequal pay-equal work claims cognizable under the EPA. Courts also have found ac tionable under Title VII claims that an employer deliberately depresses women’s wages even without allegations that those women performed work substantially equal within the meaning of the EPA, to that performed by men. See Gunther v. State of Washington, supra. Such Title VII claims are premised on evidence, often statistical, indicating intentional discrimination. See Wilkins v. University of Houston, 654 F.2d 388 (5th Cir.1981), vacated and remanded on other grounds, 459 U.S. 809, 103 S.Ct. 34, 74 L.Ed.2d 47 (1982) (violation of Title VII suggested by university’s pay structure in which females comprised 18 of the 21 individuals being paid less than lowest salary level for a particular job title); Melani v. Board of Higher Education of the City of New York, 561 F.Supp. 769, 776 (S.D.N.Y.1983) (gross statistical disparities in the treatment of similarly qualified men and women may constitute prima facie proof of intentional sex discrimination).

In instances where a Title VII plaintiff brings a sex-based salary discrimination claim which would be cognizable under the EPA, the prima facie Title VII case has the same elements as the prima facie EPA claim. See Odomes v. Nucare, Inc., 653 F.2d 246, 250 (6th Cir.1981); Melanson v. Rantoul, 536 F.Supp. 271 (D.R.I.1982); see also Francoeur v. Corroon & Black Co., 552 F.Supp. 403, 408 (S.D.N.Y.1982). To establish a prima facie case under the EPA, a plaintiff must show that the defendant paid higher wages to members of the favored gender for equal work on jobs requiring equal skill, effort, and responsibility under similar working conditions. See Corning Glass Works v. Brennan, 417 U.S. 188, 94 S.Ct. 2223, 41 L.Ed.2d 1 (1974). The EPA plaintiff does not have to prove that the job “duties performed are identical. A violation of the [EPA] may be proven if the ‘skill, effort and responsibility’ required in the performance of the jobs is ‘substantially equal.’ ” Usery v. Columbia University, 568 F.2d 953, 958 (2d Cir.1977) (quoting Hodgson v. Corning Glass Works, 474 F.2d 226, 234 (2d Cir.1973), aff'd sub nom. Corning Glass v. Brennan, supra). The determination of substantial equality must be based on the work activities actually performed, and not on job title or classification. See Thompson v. Sawyer, 678 F.2d 257 (D.C.Cir.1982).

In Usery v. Columbia, supra, the Second Circuit interpreted the “substantially equal” requirement in view of regulations promulgated by the Secretary of Labor at 29 C.F.R. §§ 800.114 to 800.163 (1976). Specifically, that court discussed in detail the equal effort element of an EPA prima facie case:

Under the [EPA,] “effort” is the physical or mental exertion required in performing a job. So long as the ultimate degree of exertion remains comparable, the mere fact that two jobs call for effort different in kind will not render them unequal.... 29 C.F.R. § 800.127 (1976). Nor will effort expended on additional tasks assigned to male employees necessarily suffice to justify a pay differential. If the additional tasks do not consume a significant total amount of all of the employees’ time, ... or if female employees also perform duties which require additional effort, ... or if third persons who perform the additional tasks as their primary job are paid less than the male employees in question, ... in these situations the additional effort is insufficient to differentiate the male positions under the [EPA],

568 F.2d at 959 (citations omitted). Regulations promulgated by the Secretary of Labor describe in similar detail the equal skill and equal responsibility elements of an EPA prima facie case:

Skill includes consideration of such factors as experience, training, education, and ability. It must be measured in terms of the performance requirements of the job. If an employee must have essentially the same skill in order to perform either of two jobs, the jobs will qualify under the Act as jobs the performance of which requires equal skill, even though the employee in one of the jobs may not exercise the required skill as frequently or during as much of his working time as the employee in the other job. Possession of a skill not needed to meet requirements of the job cannot be considered in making a determination regarding equality of skill.

29 C.F.R. § 800.125 (1980). The Secretary’s regulations with regard to responsibility provide in relevant part: “Responsibility is concerned with the degree of accountability required in the performance of the job, with emphasis on the importance of the job obligation.” 29 C.F.R. § 800.129 (1983). Supervisory duties are an important element of a job’s “responsibility” requirements. See 29 C.F.R. § 800.130; Hill v. J.C. Penney Co., 688 F.2d 370, 373 (5th Cir.1982).

In the instant case, Rossini compares her salary with those paid seven men who, between 1969 and 1971, were hired for or promoted into the same job titles she occupied at O & M during that period. She alleges that -on the basis of her gender 0 & M paid her less throughout her career than those allegedly comparable men.

Again 0 & M argues that the court is precluded by the statute of limitations from considering the salary paid Rossini prior to May 30, 1975. As discussed supra, the court must determine whether any of 0 & M’s acts after that date constituted part of a pattern of discriminatory employment decisions, which arguably could subject 0 & M to liability for an entire course of conduct. The court finds, however, that Rossini has failed to establish, for the post-May 29, 1975 period, a prima facie case of salary discrimination under the Title VII/EPA standard described above. The court bases this conclusion largely on the evidence that individuals in supervisory account management positions occupied by Rossini since 1975 did not necessarily perform work of equal skill, effort, and responsibility.

As discussed above, the type of product advertised substantially affected the nature of the skills demanded of individuals working on that account. The most notable example of such differing requirements is the numerical and analytic skills undisputedly necessary for supervisors of packaged goods accounts. It is apparent that certain accounts, such as those with larger billings or more-deadlines, may entail more effort by the responsible supervisory account management personnel. Rossini has provided no evidence with regard to the nature of the accounts supervised by the allegedly comparable men, nor indeed, has she gone into significant detail with regard to the nature of her own work on most of her post-May 1975 assignments. 38

Neither has Rossini established the equal responsibility component of the EPA/Title VII standard by evidence of the number of individuals supervised by each of the allegedly comparable males. Cf. Orahood v. Board of Trustees of the University of Arkansas, 645 F.2d 651, 655 (8th Cir.1981) (in Title VII case incorporating EPA standard, no substantial equality of work where male, inter alia, supervised more employees). In addition, the evidence here suggests that supervisory employees of the same job titles may bear varying degrees of responsibility for designing and implementing an advertising program, according to the size and nature of the account. Cf. Horner v. Mary Institute, 613 F.2d 706 (8th Cir.1980) (jobs not substantially equal where male developed and implemented courses and female only taught classes selected by someone else). Rossini has failed to introduce evidence indicating the nature of the account responsibilities of the allegedly similar males. Accordingly, the court concludes that Rossini has failed to establish a claim of unequal pay for equal work in violation of the EPA standard incorporated by Title VII.

Rossini also has failed to present evidence of such intentional sex discrimination of a nature which, while not actionable under the EPA, would violate the broader protections of Title VII. Although the number of males arguably comparable to Rossini is too small to permit a sophisticated statistical analysis, the more simple averages submitted by Rossini do not demonstrate pay disparities of a magnitude which suggests intentional discrimination. 39 For the period since May 1975, the disparity between Rossini and the lowest paid of those males has been no greater than $3500 annually. All of the seven men with whom Rossini compares herself have college degrees, and three have advanced business degrees. As discussed previously, Rossini does not contest the relevance of such academic training to account management work, nor does she dispute the propriety of increased financial awards to individuals having such training. In addition, several of those men had as many as seven years of prior experience at major advertising agencies or in the advertising departments of large corporations before they came to 0 & M. Rossini’s prior experience, in contrast, was limited to advertising-related fields and smaller agencies, primarily her own. Finally, by 1977, two of those men had been promoted to management supervisor and therefore had greater job responsibilities than Rossini, who was still an account supervisor at that time. Since Rossini failed to establish her claim of promotion/assignment discrimination, the court cannot conclude that higher salaries to males occupying higher account management positions than Rossini reflected intentional gender discrimination by 0 & M. Cf. Melani v. Board, supra, 561 F.Supp. at 783 (inappropriate to consider employee rank in Title VII salary analysis, given evidence of discriminatory assignment to rank). The court concludes, that the gender disparities in the salary figures adduced by Rossini are insufficient to suggest that O & M in violation of Title VII intentionally depressed Rossini’s salary because of her gender.

E. Retaliation

Rossini alleges O & M retaliated against her after she filed the EEOC charge in 1976 and this action in 1978. 40 “To prove a claim of retaliation under 42 U.S.C. § 2000e-3(a), plaintiff must show that the defendant took some adverse employment action because of plaintiff’s action taken against perceived employment discrimination.” Ferrara v. Ciba-Geigy Corp., 548 F.Supp. 703, 706 (S.D.N.Y.1981), aff'd mem., 697 F.2d 290 (2d Cir.1982); see Lieberman v. Gant, supra, 630 F.2d at 69-70. Rossini claims that O & M punished her for her Title VII challenge by attempting to remove her from the New York office and by assigning her as management supervisor to the 1982 World’s Fair, a purportedly undesirable account.

The evidence establishes that subsequent to 1976, Rossini was offered opportunities at O & M International offices in London and Puerto Rico. In addition, she was recommended for pro bono work to be performed at the office of a non-profit institution. The utility of such assignments in furthering Rossini’s advancement at O & M was hotly contested, with 0 & M introducing evidence of high level account management personnel whose careers had advanced after pro bono work or after assignments to other offices of 0 & M International. However, even accepting Rossini’s contention that such positions were undesirable, the evidence establishes that she declined to accept them and was not punished therefor.

Rossini next argues that her assignment to the World’s Fair account constituted retaliation for her claims of discrimination. It is undisputed, however, that her position on that account was that of management supervisor, a promotion from her Amex position of account supervisor. Rossini also claims the World Fair account was undesirable because of its inevitable termination at the close of the fair. However, when her work on the World’s Fair ended, she was reassigned to Amex as a management supervisor and thus had achieved a promotion on Amex. Rossini argues that while on the World’s Fair account she supervised only two or three staff members, in contrast to the dozen for whom she had been responsible as an account supervisor on Amex. She also points to the fact that while on the World’s Fair account, she reported to at least one other management supervisor. The court has noted previously, however, that staffing patterns at 0 & M vary considerably from account to account, depending upon the nature of the product being advertised. The court cannot conclude, solely on the basis of the number of staff assigned to the World’s Fair account, that Rossini’s promotion to management supervisor on that account actually was a retaliatory demotion.

Finally, Rossini contends that the financial stability of the World’s Fair was questionable, and that its economic problems both indicated the undesirable nature of the account and were responsible for her disagreements with that client. In fact, the record establishes that the World’s Fair met most of its obligations, including its bill to 0 & M, which was paid pursuant to a contract negotiated not by Rossini but by her superior on the account, Steve Humphrey. Bill Fransisco, the World’s Fair official responsible for dealing with 0 & M, testified about his serious dissatisfaction with Rossini’s performance, which culminated in his request for her removal from the account. He stated that his unhappiness stemmed not from Rossini’s actions to protect the financial and legal interests of 0 & M and the World’s Fair, but rather from her delay in providing a clear explanation for taking particular actions. Fransiseo stated because Rossini was unable to coordinate 0 & M staff working on the World’s Fair, he received frequent unnecessary phone calls’ from 0 & M employees. He testified further that Rossini dealt with him and other World’s Fair personnel in a condescending and arrogant manner.

The court finds the testimony of Fransisco, who has no further business dealings with 0 & M, to be convincing evidence that the poor evaluation Rossini received for her performance on the World’s Fair was neither discriminatory nor retaliatory. Neither Fransisco’s testimony, nor the rest of the evidence, suggests that the World’s Fair was such an undesirable account that Rossini’s assignment thereto itself constituted retaliation.

III. Class Issues

A. Alleged Class-Wide Salary Discrimination

(1) Burdens of proof and Production

Plaintiffs have alleged that 0 & M regularly pays less to women performing the same work as comparably qualified men.

The plaintiffs in a Title YII action may establish a prima facie case of class-based discrimination by “demonstrating the existence of a discriminatory pattern and practice.” International Brotherhood of Teamsters v. United States, 431 U.S. 324, 359 [97 S.Ct. 1843, 1866, 52 L.Ed.2d 396] (1977) (quoting Franks v. Bowman Transportation Co., 424 U.S. 747, 772 [96 S.Ct. 1251, 1267, 47 L.Ed.2d 444] (1976). In order to prove such a pattern or practice, plaintiffs must demonstrate more than “isolated or ‘accidental’ or sporadic discriminatory acts.” They must show that discrimination was “the regular rather than the unusual practice” of the defendant. Teamsters v. United States, supra, 431 U.S. at 336 [97 S.Ct. at 1854].

Melani v. Board of Higher Education of the City of New York, 561 F.Supp. 769, 772 (S.D.N.Y.1983) (footnote omitted); see Eastland v. TVA, supra, 704 F.2d at 618. Plaintiffs’ initial post-trial memorandum suggests indirectly that the court may use a disparate impact analysis of the effect of subjective criteria used by 0 & M in setting salary. The thrust of plaintiffs’ memoranda and their trial presentation, however, was that 0 & M intentionally depressed class members’ salaries on the basis of their gender. Plaintiffs, for example, rely primarily upon cases such as Melani v. Board, supra, and Trout v. Lehman, 702 F.2d 1094 (D.C.Cir.1983), vacated on other grounds, — U.S.-, 104 S.Ct. 1404, 79 L.Ed.2d 732 (1984), which employ a disparate treatment analysis of allegations of gender-based salary differences. Accordingly, in the interest of adequate notice to O & M of plaintiffs’ theory, the court will consider the class-wide salary discrimination allegations as a claim of disparate treatment. Cf. Sainte Marie v. Eastern Railroad Association, 650 F.2d 395, 399 n. 2 (2d Cir.1981) (where case tried on the basis of disparate treatment theory, appellate court would not consider claim- of disparate impact). 41

Evidence of discriminatory intent necessary to proof of such a claim may be established by statistics indicating gross disparities in the treatment of the plaintiff group. See Hazelwood School District v. United States, 433 U.S. 299, 307-308, 97 S.Ct. 2736, 2741, 53 L.Ed.2d 768 (1977); Eastland v. TVA, supra, 704 F.2d at 618. But see Segar v. Smith, 738 F.2d 1249, 1278 (D.C.Cir.1984) (where statistics are finely tuned ... “gross disparities need not be shown to permit an inference of discrimination”). In particular, courts have looked for statistical evidence of significant salary differences between similarly qualified men and women in jobs which are not of a “widely disparate nature.” Melani v. Board, supra, 561 F.Supp. at 777; Sobel v. Yeshiva University, 566 F.Supp. 1166 (S.D.N.Y.1983). Such disparities can be an indication that women’s wages are “depressed because of intentional sex discrimination in violation of Title VII.” Washington v. Gunther, supra, 452 U.S. at 166, 101 S.Ct. at 2246: Melani v. Board, supra. 42

[W]hile plaintiffs must demonstrate to the court’s satisfaction that their statistical comparisons are meaningful, they need not present a perfect statistical analysis at the prima facie case stage____ The appropriate degree of refinement óf the plaintiff’s statistical analysis, moreover, may depend on the quality and control of the available data____ As the District Court observed, “plaintiffs cannot legitimately be faulted for gaps in their statistical analysis when the information necessary to close those gaps was possessed only by defendants and was not furnished either to plaintiffs or to the Court.” Trout [v. Hidalgo], 517 F.Supp. [873], 883 [D.D.C.1981) ] (footnote omitted).

Trout v. Lehman, supra, 702 F.2d at 1101-1102 (citations omitted).

Although a plaintiff can establish a prima facie case of disparate treatment solely on the basis of statistics, see Segar v. Smith, supra; Hudson v. International Business Machines Corp., 620 F.2d 351, 355 (2d Cir.), cert. denied, 449 U.S. 1066, 101 S.Ct. 794, 66 L.Ed.2d 611 (1980); Melani v. Board, supra, 561 F.Supp. at 777, the sufficiency of statistical evidence will be evaluated in light of any testimony in the record regarding plaintiffs’ individual experience of discriminatory treatment. See United States v. Teamsters, supra, 431 U.S. at 338-39, 97 S.Ct. at 1855-56; EEOC v. Federal Reserve Bank of Richmond, 698 F.2d 633, 643 (4th Cir.1983) reversed on other grounds, sub. nom. Cooper v. Federal Reserve Bank of Richmond, — U.S. -, 104 S.Ct. 2794, 81 L.Ed.2d 718 (1984); Ste. Marie v. Eastern Railroad Association, supra, 650 F.2d at 405-06.

If plaintiffs’ statistics establish a prima facie case of class-wide intentional discrimination

the burden shifts to defendant to rebut the inference of discrimination by showing that plaintiff’s statistics are misleading or by presenting legitimate non-discriminatory reasons for the disparity. Texas Department of Community Affairs v. Burdine, 450 U.S. 248 [101 S.Ct. 1089, 67 L.Ed.2d 207] (1981). Defendant’s burden is production not persuasion. If defendant meets its burden, plaintiff may show that the asserted explanations are inaccurate or otherwise unworthy of credence. The ultimate burden of persuasion remains at all times with plaintiff.

Eastland v. TVA, supra, 704 F.2d at 618-19; see Trout v. Lehman, supra, 702 F.2d at 1113 n. 10, Melani v. Board, supra, 561 F.Supp. at 781. The Court of Appeals for the District of Columbia recently discussed the special nature of a defendant’s burden when plaintiffs have established by means of statistics a prima facie case of class wide discrimination.

[T]o make an initial showing of disparate treatment in such cases the plaintiff class will typically have presented statistical evidence showing pervasive disparir ties and eliminating most, if not all, potential nondiscriminatory explanations for the observed disparities. See Vuyanich [v. Republic National Bank of Dallas] 521 F.Supp. [656,] 663 [(N.D.Tex.1980), vacated and remanded on other grounds, 723 F.2d 1195 (5th Cir.1984)]. Though the employer is not required to meet a burden of persuasion of rebutting the disparate treatment claim, the nondiscriminatory explanation must cast sufficient doubt on the plaintiffs’ proof to permit the trier of fact legitimately to decline to draw an inference of discrimination from that proof. The bare articulation of a nondiscriminatory explanation, while sufficient to rebut an individual plaintiff’s low-threshold McDonnell-Douglas showing, generally' will not suffice as a rebuttal to a typical class-wide showing of pervasive discrimination. Burdine made this much clear____

Indeed, there may be some cases where the plaintiffs initial evidence, combined with effective cross-examination of the defendant, will suffice to discredit the defendant’s explanation.

450 U.S. at 255 n. 10 [101 S.Ct. at 1095 n. 10] (emphasis added). The typical pattern or practice case is closely analogous to the situation the Court envisions in the last sentence of the quoted passage____ Thus in both individual and class action contexts the defendant faces the same rebuttal burden; it must present sufficient evidence to permit the trier of fact to decline to draw the inference of dis crimination from the plaintiffs’ proof. But in the class action pattern or practice case the strength of the evidence sufficient to meet this rebuttal burden will typically need to be much higher than the strength of the evidence sufficient to rebut an individual plaintiff's low-threshold McDonnell Douglas showing.

Segar v. Smith, supra, 738 F.2d at 1269 (footnotes omitted).

(2) Statistical Studies

(a) Plaintiffs’ Studies

At trial, the court reserved decision as to the admissibility of certain expert reports. For purposes of clarity, the court describes those as well as the statistical reports already received in evidence.

Plaintiffs’ reports were prepared by Dr. Donald Wise, an economist and statistician associated with the New Jersey consulting firm of Mathtech, Inc. Dr. Wise’s first report (“Pl.Ex. 229”) contained a number of studies. Pl.Ex. 229 includes a number of descriptive statistical compilations, by year between 1970 and 1980, of average salaries of professionals at 0 & M by gender; of average salaries of newly hired professionals at 0 & M by gender; and of average annual salary increases for professionals at 0 & M by gender. The descriptive studies also include listings of the average salaries earned by men and women in each job title, by year, during that period.

Pl.Ex. 229 also contains what Dr. Wise describes as “the most important” of his studies, the salary regression analyses.

Multiple regression analysis permits the calculation of the effect of a variety of factors called independent variables on a particular result called the dependent variable. The use of the computer both reduces the time necessary to perform this analysis and makes it possible to include a large number of independent variables whose effect on the dependent variable may be thus determined. This technique has been recognized as useful for measuring the average salary differential, if any, between men and women or blacks and whites while controlling for or factoring out the effects of other productivity-related characteristics which may lead to differences in compensation. See Trout v. Hidalgo, 517 F.Supp. 873, 877-78 (D.D.C.1981) [affirmed in part and reversed in part on other grounds sub. nom. Trout v. Lehman, supra ]; Greenspan v. Automobile Club of Michigan, 495 F.Supp. 1021, 1061 (E.D.Mich.1980); Fisher, Multiple Regression in Legal Proceedings, 80 Colum.L.Rev. 702, 721-25 (1980) [(“Fisher”)].

Melani v. Board, supra, 561 F.Supp. at 774. The “sex” or “gender” coefficient in such studies, that is, the coefficient of the independent variable representing gender, is a purported measure of the difference in salary correlated with gender when independent variables measuring certain productivity characteristics are held constant. See Sobel v. Yeshiva, supra, 566 F.Supp. at 1174; Vuyanich v. Republic National Bank of Dallas, 505 F.Supp. 224, 267-79 (N.D.Tex.1980), vacated and remanded on other grounds, 723 F.2d 1195 (5th Cir.1984) (“Vuyanich I”).

The regression studies in Pl.Ex. 229 analyzed the salary differences between non-officer male and female employees at O & M while controlling for certain differences in education, experience and tenure at O & M. Dr. Wise also performed regressions on the salaries of newly hired employees, for each year between 1970 and 1980, and on annual salary increases for professional men and women at 0 & M during that period.

In addition to producing the coefficient of each independent variable as an estimate of the impact of that variable on salary, plaintiffs’ regression analyses also provide a measure of the statistical significance of the gender coefficient. Statistical significance refers to the likelihood

that the true coefficient is actually zero; that is, that the independent variable [in the instant case, gender] to which it corresponds has no effect on the dependent variable [in the instant case, salary or salary increase]. Fisher, supra, [80 Colum.L.Rev.] at 717. If what is known as the “t-statistic” for the particular coefficient is large enough, then we can reject the hypothesis that the true coefficient of the variable in question is equal to zero____ For example, if a 5% level of significance is used, a sufficiently large t-statistic for the coefficient indicates that the chances are less than one in 20 that the true coefficient is actually zero. Fisher, [80 Colum.L.Rev.] at 717. The magnitude of the t-statistic necessary to reject such hypothesis varies with the desired level of significance. Thus the t-statistic would need to be larger with a 5% level of significance than with a 10% level of significance.

The magnitude of the t-statistic is also dependent on the particular type of hypothesis to be tested. When an analyst uses a “two-tailed hypothesis” — that is, where he wishes to determine whether there is any relationship, positive or negative, between the independent and the dependent variable — a t-statistic of approximately two means (in the ease of large samples) that the chances are less than one in 20 that the true coefficient is actually zero. Cf. id.

Vuyanich I, supra, 505 F.Supp. at 271-72. Thus, in the instant case the t-statistic represents a measure of the probability that the observed salary differentials between men and women (the gender coefficient in each analysis) actually resulted from random factors. Because the regression analyses seek to measure positive or negative relationships between gender and salary or salary increases, a t-statistic of approximately two indicates that the chances are less than one in 20 that the gender coefficient is the product of chance factors and the true coefficient of gender is zero. Plaintiffs’ expert testified that for the regressions of the type he performed, the t statistic was roughly equivalent to the number of standard deviations. See Segar v. Smith, supra, 738 F.2d at 1261 n. 7.

Plaintiffs’ reply report (“Pl.Ex. 230”) was proffered primarily to respond to the contention by O & M’s expert that the regressions in Pl.Ex. 229 reflected the problem of “serial correlation.” That problem is explained most simply by a report prepared for O & M by Dr. Herbert Robbins of Columbia University:

The method Dr. Wise uses for his regression analysis systematically overestimates the. statistical significance of the regression coefficients and makes the statistical significance of the differences between the salaries of males and females appear larger than should be the case ---- This error is caused by his apparent neglect of what is known as “serial correlation.”'

Dr. Wise’s regression analyses seem to have been performed by putting into a single equation the salaries extending over ten different years and into the spring of 1980. Thus, if a person were at O & M for all eleven years, his or her salary and data appear and are counted eleven times____

Suppose for simplicity that the individual salaries of females and males did not change over the period 1970-1980 and that their background descriptive variables also did not change. In this case, the inclusion of all of the yearly data for each of the eleven years is equivalent to using the same data eleven times over. An estimated regression sex coefficient that is not significantly different from zero for any one year may easily become so when compounded over eleven years. Boosting the statistical significance of results by spuriously multiplying the number of observations is not a legitimate statistical technique.

Accordingly, in Pl.Ex. 230, Dr. Wise stated that he had revised Pl.Ex. 229, by “estimat[ing] separate regression equations for each year from 1970 to 1980. This approach eliminates any possibility of serial correlation in the analysis.” It is undisputed that Pl.Ex. 230 corrected the serial correlation problem in Pl.Ex. 229. 43 In addition to addressing the serial correlation problem, Pl.Ex. 230 contains a number of studies Dr. Wise performed on 0 & M’s data base, while excluding the independent variables to which Dr. Wise had “strong objection.”

0 & M at trial objected to the admissibility of Pl.Ex. 229 and Pl.Ex. 230 on a number of grounds. 0 & M flagged two programming errors in those reports. First, each individual was given two years seniority for each year he or she actually was employed at 0 & M. Second, many of the regression studies included a variable indicating as a measure of experience years between the date of an individual’s last degree and his or her date of hire at 0 & M. In the initial version of plaintiffs’ reports, Dr. Wise had programmed the computer to indicate as zero the time between date of degree and date of hire at 0 & M for those individuals for whom 0 & M’s computerized records (the “PERS 3600 tapes”) did not indicate date of degree. 0 & M asserted that plaintiffs’ initial report thus incorrectly assumed that approximately 250 individuals began work at O & M as of the date of their last degree.

Early in the trial, plaintiffs produced revised studies (“yellow report”) which corrected the double seniority and the date of degree-date of hire programming errors. Although the yellow report was never marked for identification or formally introduced into evidence, O & M thoroughly cross-examined Dr. Wise with regard to those studies. Moreover, the yellow report was based on information contained in the computer tape of plaintiffs’ data base, given to O & M months before trial. O & M thus could have, and apparently did, rerun studies on plaintiffs’ data to check the regressions in the yellow report. The court therefore finds that O & M will not be unfairly prejudiced by consideration of the yellow report, which corrects the two aforementioned programming errors.

O & M raised a number of additional objections to Pls.Exs. 229 and 230, as amended by the yellow report. First, as O & M’s computer programmer testified and Dr. Wise virtually conceded, those studies included only the 449 O & M professional employees included on the PERS 3600 tape, thereby omitting all such employees who left O & M in 1975, 1976, or 1977. Secondly, O & M objected to Dr. Wise’s technique of “constructing” a date of graduation in order to obtain a measure of experience for certain O & M employees. Dr. Wise recorded such individuals as graduating from high school at age 18, from junior college or technical school at age 20, from college at age 22, and from post-graduate training at 24. Finally, O & M contends that although it alerted plaintiffs to allegedly serious accuracy problems in the educational data in the PERS 3600 tapes, Dr. Wise did not correct those data by reference to individual personnel files O & M had given to plaintiffs. The court finds that such objections are relevant to the weight rather than the admissibility of those studies by plaintiffs’ expert. See Wilder Enterprises v. Allied Artists Pictures Corp., 632 F.2d 1135, 1144 (4th Cir.1980); Phelps Dodge Corp. v. Atchison, Topeka & Santa Fe Railway Co., 400 F.2d 20, 24 (10th Cir.1968). Accordingly, Pl.Exs. 229 and 230, as amended by the yellow report, must be considered.

Plaintiffs’ rebuttal report, Volumes one and two, was admitted without objection. Volume one (“Pl.Ex. 283”) repeated certain regression analyses performed by O & M’s expert, but included in those analyses approximately 40 O & M employees omitted from O & M’s initial studies. The added individuals were classified as “managers” in work force records O & M kept pursuant to Title VII. See 42 U.S.C. § 2000e-8(c); 29 C.F.R. §§ 1602.7 et seq. 44 Volume two of plaintiffs’ rebuttal report (“Pl.Ex. 301”) contains Dr. Wise’s study of whether there are statistically significant differences in the structure of the regression models for “salaries of those 0 & M employees hired before May 1975 and of those hired thereafter.” It also contains “Table I,” containing other versions of the regression analyses comparing salaries received by men and women employees at 0 & M between 1975 and 1979.

Finally, plaintiffs offered a supplemental report (“green report”) amending the regression models in Pl.Exs. 229 and 230, as corrected by the yellow report. In preparing the green report, plaintiffs’ expert sought to address the problem of the constructed date of graduation previously raised by 0 & M. Plaintiffs returned to the personnel files and obtained the actual date of graduation for 144 0 & M employees for whom that information was constructed in plaintiffs’ earlier studies.

0 & M objected to the introduction of the green report since it was based on information plaintiffs culled from personnel files after the trial began, even though plaintiffs had been given copies of those files many months prior to the commencement of trial. 0 & M argues that the added dates of graduation in the green report therefore were not in the computer tapes which reflected the plaintiffs’ data base and which were given to 0 & M pursuant to court order many months before trial. 0 & M also notes that when, in mid-trial, plaintiffs gave it the computer tapes on which the green report was based, plaintiffs did not indicate that those tapes, unlike any others exchanged during and before trial, contained information not included in plaintiffs’ original data base. The green report thus differs from plaintiffs’ other corrected reports which were simply new computer runs on the data base already provided to 0 & M. 0 & M correctly asserts that in order to evaluate the studies in the green report, its expert would have to return to the personnel files to check the accuracy of the 144 newly obtained dates of graduation.

The court finds that admission of the green report would have imposed an unduly burdensome task upon 0 & M. Such an imposition would have been particularly unfair since 0 & M, months before trial, gave plaintiffs its expert’s report criticizing Dr. Wise for his decision to “make up” or construct the value of certain missing variables when data pertinent to those variables were not available. The court concludes that plaintiffs cannot submit as part of their rebuttal a study founded on material outside the original data base. See generally Fed.R.Civ.P. 16(e); 3 Moore’s Federal Practice ¶ 16.18; Recommendations on Pretrial Proceedings in Cases with Voluminous Data, 39 New York City Bar Association Record 49 (1984). 45 0 & M’s objection to the admission of the green report therefore is sustained. 46

(b) 0 & M’s Studies

0 & M introduced statistical studies prepared by its expert, Dr. Orley Ashenfelter, a professor of economics at Princeton University. 0 & M’s initial report (“D.Ex. 126”) included studies of gender in hiring and in accessions into job title by transfer or promotion. It also contained regression analyses of gender differences in salary for each year from 1976 to 1979 for 0 & M employees hired since May 1975, holding constant certain indicators of education, prior experience, tenure at 0 & M and former salary. Other regression analyses in D.Ex. 126 measured salary differences for male and female professionals at 0 & M while controlling for salary as of May 1975 as well as the above-mentioned productivity factors.

0 & M also submitted a supplementary report (“D.Ex. 127”), which contained regression analyses of salaries similar to those in D.Ex. 126. Studies in the supplementary-report, however, deleted the controversial independent variables of salary prior to 0 & M and job title at 0 & M, while adding variables to hold constant college major and college standing, time since degree, other training, 0 & M department, and age. Certain studies in D.Ex. 127 also added a variable to reflect the 0 & M employee’s military service experience.

D.Ex. 127 excluded non-college graduates from the 0 & M employees studied in D.Ex. 126. 0 & M therefore submitted a reply report (“D.Ex. 129”), repeating its prior studies, but including in all but one of them, 0 & M professional employees who had not completed college. In addition, 0 & M in D.Ex. 129 amended its previous studies in a number of other ways in response to criticisms by plaintiffs’ expert. Plaintiffs had objected to 0 & M’s use of the Gourman Report, a ranking of a large number of American colleges, as a measure of standing of the college attended by 0 & M employees. Plaintiffs argued that that report failed to include among its high-ranking schools a number of nationally recognized women’s colleges, i.e., Wellesley, Mount Holyoke, Smith, Yassar and Barnard. In D.Ex. 129, 0 & M included those women’s colleges among the institutions of higher standing reflected by the studies’ variable. In addition, 0 & M’s expert responded to plaintiffs’ other criticisms by reducing to one variable the measure of training outside college and by reducing the number of variables used to measure job tenure.

On rebuttal, 0 & M sought to introduce regression studies, D.Ex. 698a-698n, similar to those in its earlier expert reports, except that most of the rebuttal studies included the previously omitted managers and, in certain instances, added a variable controlling for military experience. Plaintiffs objected to parts b, d, e, f, g and h of D.Ex. 698, purportedly on the same basis that 0 & M had objected to plaintiffs’ green report. Plaintiffs contend that by including the managers in their rebuttal studies, 0 & M was offering a report premised on a new data base.

Plaintiffs’ data base argument with regard to D.Ex.' 698 is without merit. Plaintiffs do not and cannot argue that the 0 & M employees classified as managers were excluded from the 0 & M data base given to plaintiffs prior to trial. There is no indication that 0 & M returned to the personnel files to obtain data on those managers. Consequently, plaintiffs could have checked D.Ex. 698 by running studies based on computer tapes already in their possession and would not have had to check the accuracy of 0 & M’s manager data against the personnel files. Finally, the court notes that plaintiffs’ expert reports, given to 0 & M before trial, which specifically criticized the omission of non-college graduates in certain 0 & M studies, made no mention of the omitted managers.

Accordingly, the court concludes that 0 & M’s expert inadvertently omitted managers from his studies and that the omission, was first drawn to his attention during the course of trial. Because D.Ex. 698, which includes the managers,' did not incorporate any evidence newly gathered from the personnel files, they are admissible as appropriate rebuttal evidence.

(3) Discovery of Officer Personnel Files

Plaintiffs contend that the probative value of all the statistical studies in evidence has been severely limited by an order denying them discovery of the personnel files of approximately sixty men and thirty-five women who were elected vice-president at 0 & M between May 30, 1975 and 1980. The 0 & M employees included in the data bases of the regression studies accordingly excluded such individuals. Plaintiffs claim that this limitation on the universe studied is an artificial one, since the evidence clearly established that election to vice president did not change the nature of the job duties performed by the employee.

The history of the dispute over the officer data is a long and tortured one. On August 4, 1980, Magistrate Washington ruled that 0 & M was not required to produce personnel files or other records of officers. By order dated December 10, 1980, Judge Brieant affirmed the magistrate’s order but granted plaintiffs leave to seek further relief from the Magistrate after they had obtained “a representative sample of the paperwork by which an officer is made” and had deposed a high-ranking 0 & M employee about the procedure for electing officers. 0 & M thereafter produced one to two page biographical sketches of approximately 30 officers. Plaintiffs correctly contend that those biographies were insufficient to permit detailed statistical analysis since they frequently lacked information regarding the officer’s prior work history, educational background, or date of hire. The magistrate denied plaintiffs’ request for further officer discovery.

Plaintiffs again appealed the magistrate’s ruling to Judge Brieant, who on April 13, 1981 affirmed the magistrate’s order as not clearly erroneous under 28 U.S.C. § 636. Nevertheless, Judge Brieant during oral argument prior to his decision, expressed serious reservations about the discovery limitations.

Why don’t you give [plaintiffs’ counsel] the files and tell her that it is not relevant, but in order that I don’t have to defend later on I am getting rid of this problem____

But you are allowing the Magistrate to set a trap for you because if you win this case on the merits you are going to be confronted on appeal with the contention that there is a simple group of files which should have been given to me and I didn’t get them because the Magistrate refused to give them to me and — assume for the purposes of our discussion, and I have not decided this motion, but assume that I decide to stick with the Magistrate____

A trap has been built for you and your client. Do you understand that?

Without opinion, in 1981 I denied plaintiffs’ motion to reconsider Judge Brieant’s discovery order. Plaintiffs renewed their motion for reconsideration at trial, and decision was reserved. During the course of trial, plaintiffs limited the requested discovery to the personnel files of the ninety-five individuals who had been elected officers between the May 30, 1975 limitations date and the effective end of class-wide discovery in 1980.

O & M argues initially that review of Judge Brieant’s discovery ruling is precluded since it is the law of the case. The Second Circuit recently emphasized the discretionary nature of the law of the case doctrine:

In Slotkin v. Citizens Casualty Co. of New York, 614 F.2d 301, 312 (2d Cir.1979), ce rt. denied, 449 U.S. 981, [101 S.Ct. 395, 66 L.Ed.2d 243] (1980), we reiterated our long-established view that the law of the case is, at best, a discretionary doctrine which “does not constitute a limitation on the court’s power but merely expresses the general practice of refusing to reopen what has been decided. Dictograph Products Co. v. Sonotone Corp., 230 F.2d 131, 134-36 (2d Cir.), petition for cert. dismissed per stipulation, 352 U.S. 883 [77 S.Ct. 104, 1 L.Ed.2d 82] (1956).” Thus, judges of coordinate jurisdiction are not bound by each other’s rulings, but are free to disregard them if they so choose. The only limitation placed upon a trial judge’s decision to disregard a previous ruling by a judge of coordinate jurisdiction is that prejudice not ensue to the party seeking the benefit of the doctrine. See First National Bank of Hollywood v. American Foam Rubber Corp., 530 F.2d 450, 453 n. 3 (2d Cir.), cert. denied, 429 U.S. 858 [97 L.Ed.2d 157, 50 L.Ed.2d 135] (1976). In this context prejudice does not mean the harm which results from a failure to apply the doctrine; rather, it refers to a lack of sufficiency of notice and an opportunity to prepare armed with the knowledge that one judge is disregarding the ruling of another. Id.

United States v. Birney, 686 F.2d 102, 107 (2d Cir.1982). The court emphasized a trial judge’s discretion to revise prior evidentiary rulings.

The judge who actually presides at trial must be accorded wide discretion in deciding either to admit or exclude evidence, since it is he who determines the course which the trial is to take. See Cruz v. United States Lines Company, 386 F.2d 803, 804 (2d Cir.1967). When a trial judge believes evidence to be relevant “it could be reversible error to blindly adhere to [a] prior order.” Control Data Corp. v. International Business Machine Corp., 421 F.2d 323, 327 (8th Cir.1970). Thus in deciding whether to admit the proffered evidence the trial record to that point should be the decisive factor, not a pre-trial order of another judge.

Id. If the evidentiary ruling at issue here is reversed, 0 & M will not be prejudiced in the manner described in United States v. Birney, since both sides would be granted equal opportunity to use and respond to newly admissible evidence.

Accordingly, the court now reconsiders the correctness of the discovery ruling regarding officer data in light of the evidence already admitted at trial. The court notes first the broad scope of discovery provided by the Federal Rules of Civil Procedure and the similarly broad definition of admissibility provided in the Federal Rules of Evidence. Rule 26(b)(1), Fed.R.Civ.P., states that “[p]arties may obtain discovery regarding any matter, not privileged, which is relevant to the subject matter involved in the pending action____” Rule 401, Fed.R.Evid., defines relevant evidence as “evidence having any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence”, and Rule 402, Fed.R.Evid. states that “[a]ll relevant evidence is admissible,” except as otherwise provided.

More specifically, as the Supreme Court stated in McDonnell Douglas v. Green, supra, 411 U.S. at 804-805, 93 S.Ct. at 1825, information regarding a defendant employer’s treatment of minorities throughout its workforce is relevant to the showing of intent required for a disparate treatment claim. In Kohn v. Royall, Koegel and Wells, 496 F.2d 1094, 1100-1101 (2d Cir.1974), the Second Circuit noted that an individual claim of discriminatory failure to hire by a law firm would entail proof of that defendant’s employment practices far beyond the limited decision not to hire plaintiff. In Blank v. Sullivan & Cromwell, 418 F.Supp. 1 (S.D.N.Y.1976), this court relied on Kohn v. Royall, supra, in holding that even if a Title YII plaintiff alleging sex discrimination by a law firm did not claim that she was illegally denied partnership, evidence with regard to partnership decisions was relevant to the issue of the defendant firm’s intent in refusing to hire the plaintiff. Thus, proof of an employer’s treatment of women in other positions is relevant to a Title VII plaintiff’s allegation that the defendant employer deliberately discriminated against her on the basis of her gender. See Webb v. Westinghouse Elec. Corp., 81 F.R.D. 431 (E.D.Pa.1978); Hubbard v. Rubbermaid, Inc., 78 F.R.D. 631 (D.Md.1978).

More significantly, plaintiffs here contend that the officer data serve not only as proof of O & M’s treatment of women generally, but may be reflective of discrimination specifically affecting the plaintiff class of non-officer females. O & M’s expert offered an initially persuasive argument why the officer information does not bear directly on the question of discrimination against non-officer females at O & M.

The reason that people get into confusion about this, I think is because there is a confusion about the differences between whether or not evidence of discrimination outside the class is available for making a decision about the class itself. That’s really where the issue I think arises____

Well, look, if somebody asked you the following question____what’s the difference between the heights of men and women in this room? You could ask that question.

That’s a definition of the group.

Then I would say you do not need to go out and sample the people out here on the street to find out what their heights are____

Q That would add data that would skew the result because it wouldn’t have anything to do with the people in the room? A That’s right....

If you just want to know what the answer is to the question of heights between males and females in this room, you don’t think you need [any] data [other] than that____

So if you define the question to be, I want the differences between salaries of males and females, equal qualifications, and in the class [of non-officers], then I say that you don’t need any other information to get that figure.

Dr. Ashenfelter thus testified that if the issue is a comparison of the salaries of women in the class, that is, non-officers, with those of similarly qualified non-officer males, there is no need to examine officer data. That testimony, unquestionably correct as a matter of statistics, nevertheless is based upon an incorrect characterization of the legal issue before the court.

Dr. Ashenfelter presumes that the only males at 0 & M comparable to the female non-officer class members are male non-officers. The evidence establishes, however, that at least at the level of vice president, an officer title makes no difference in an 0 & M employee’s job functions. Thus, for example, a male and a female non-officer account supervisor performing the same functions generally would continue to perform the same functions if the male was elected to officer status and the female was not. It is 0 & M’s stated policy that officers are given no salary increase as a result of their election, but are only accorded a number of shares of 0 & M stock, not relevant to the salary issue under consideration here. Accordingly, there is no basis in the record or in Title VII law for limiting to non-officers the group of males compared to plaintiff non-officer females.

The court is well aware of the significant burden upon the parties of reviewing the officer files and including those employees and their data in new versions of the regression studies previously submitted. Nevertheless, the omitted officers represent a relatively large number of employees in proportion to the number studied in regression analyses already submitted by both sides, some of which include as few as 250 employees. It is possible, therefore, that the inclusion of data regarding the omitted officers could change the salary disparities indicated as the coefficients of gender from statistically insignificant to significant, or vice versa. For example, if 0 & M was paying numbers of non-officer males more than comparably qualified non-officer females with similar functions, evidence of that discrimination would have been removed from court scrutiny if during the period from 1975 to 1980 the males were elected vice-presidents.

The court must conclude that the data on the ninety-five individuals elected officer between May 30, 1975 and 1980 are relevant to the question of whether O & M intentionally discriminated against female non-officer class members. Accordingly, within thirty days of the date of this decision, O & M will turn over to plaintiffs’ counsel the officer files at issue, which will be subject to the protective orders already in place. The data from those files are to be entered on computer tapes and then exchanged by the parties under the supervision of the magistrate, as was done with the non-officer data. The parties will then analyze the new information. In order that neither side can take advantage of the initial erroneous discovery ruling, the court directs that, to the extent possible, the regression studies including officers replicate those based solely upon non-officers. In particular, neither side is to use this reopening of the evidence as an opportunity to add or delete contested independent variables or to alter questioned data used for non-officers. The new studies, moreover, will pertain solely to the issue of classwide salary discrimination, because there is an insufficient showing that the officer data significantly could affect the evidence submitted at trial with regard to the other issues in this case. After the revised regressions are submitted, the court will, if necessary, hear additional expert testimony with regard to the significance of the new studies.

B. Sex Segregation/Placement

Plaintiff Class II, as defined above, has alleged that O&M discriminatorily maintains certain departments as sex-segregated. In support of that claim, plaintiffs have introduced evidence indicating that in a significant number of job families, 80% or more of the employees are female. This claim is also one of disparate treatment; members of Class II allege that O&M intentionally has assigned them to, and prohibited them and other women from, transferring out of certain segregated job families.

To prove a prima facie case, from which the court can infer a likelihood that O&M discriminatorily maintained sex-segregated departments, plaintiffs must show more than simply the existence of largely female departments. By analogy with an individual disparate treatment case, plaintiffs must show that O & M’s acts, “if otherwise unexplained, are more likely than not based on the-consideration of impermissible factors” such as gender. Furnco Construction Corp. v. Waters, 438 U.S. 567, 577, 98 S.Ct. 2943, 2949, 57 L.Ed.2d 957 (1978).

Plaintiffs have failed to make such a showing. They have offered no testimony of women other than Zukofsky assigned to the predominantly female job families. Although Zukofsky herself was not permitted to transfer from broadcast forwarding to a purportedly more desirable department having a larger number of male employees, she named several other females who did receive such transfers. Plaintiffs offered no description of the skills or background of the women in the largely female job families, thereby failing to raise any implication that the distribution of women in those departments is anything other than a reflection of the abilities and interests of those women. Cf. Sainte Marie v. Eastern Railroad Association, supra, 650 F.2d at 401 (disproportionate assignment of women to lower level jobs not probative of discrimination in absence of evidence that they possessed skills relevant to other positions).

Plaintiffs introduced no data measuring the approval rate for requests for transfers from the largely female departments or indicating a disproportionate number of females hired after the limitations date and assigned to job families at issue. Cf. Lewis v. Tobacco Workers, supra, 577 F.2d at 1143 (significant number of blacks hired into permanent positions indicative that black employees not limited to largely minority seasonal positions). O&M did introduce studies indicating that for the period from 1976 to 1979 women at 0 & M were promoted or transferred in numbers not disproportionate to their overall percentage at 0 & M. 47 0 & M’s data also indicate proportionate assignment of women to job titles and departments which undisputedly were not sex segregated. 48 Although 0 & M’s studies are flawed by their inclusion of transfers or promotions from clerical to professional positions, those studies provide at least some indication that women employees at 0 & M were not routinely stranded in the job family or job title in which they entered. Plaintiffs have introduced no statistical evidence to suggest the opposite conclusion. The court accordingly finds that the members of Class II have failed to establish a prima facie case of impermissible segregation of job families by gender.

IV. Zukofsky’s Individual Claims

As discussed above, Zukofsky has failed even to allege facts which would state a claim for discriminatory denial of promotion or training opportunities. As indicated at p. 1168, supra, members of Class II have failed to establish their claim of sex-segregated job families. For the reasons stated in that section, Zukofsky has failed to establish that her department broadcast operations, discriminatorily was maintained as a largely female unit. The court will reserve decision on Zukofsky’s individual salary claim, which will be analyzed in conjunction with the new data to be submitted for the classwide salary claim. Accordingly, the court now considers Zukofsky’s remaining claims: discriminatory denial of transfer and retaliation for her allegations of sex discrimination.

A. Transfers

Zukofsky argues that O&M discriminatorily denied her repeated requests for transfer out of broadcast operations. 0 & M argues that Zukofsky, as part of her prima facie case, must prove that she submitted timely applications for actual openings in other departments. As discussed at pp. 1145-1146, the court finds that such a requirement is inappropriate in view of 0& M’s failure to post job openings or other-' wise to keep formal records regarding opportunities and requests for transfer. However, the court does not agree with Zukofsky that it must now consider every entry level assignment in the departments for which Zukofsky expressed an interest. The evidence establishes that large numbers of entry level positions were filled in several of the departments at issue during the relevant period. The court does not find it reasonable to expect 0 & M to have kept Zukofsky in mind for all of those jobs, particularly since 0 & M’s statistics indicate a proportionate assignment of women assigned to those entry level positions. The court therefore will examine those denials occurring close to the occasions when Zukofsky made a specific request for transfer.

Zukofsky testified regarding- a number of attempts to obtain a transfer, and the denial of any positions which actually became available within a reasonable period of time after Zukofsky’s requests. The court will consider those efforts under the same modified McDonnell Douglas standard which was applied in Rossini’s individual case. See pp. 1139-1141 and 1145-1146, supra.

In or about 1974, Zukofsky spoke to Jack Silverman, a high-ranking employee in the creative department, in an effort to become an assistant producer, an entry level position in the sub-department responsible for the creation of television commercials. Although Zukofsky thereafter made some effort to contact Reva Korda, the woman who headed 0 & M’s creative department at that time, Zukofsky acknowledged that she, in effect, sabotaged her own effort to be transferred to a job as assistant producer.

Q After you spoke to Jack Silverman, is it correct that he called you a number of times?

A Yes.

Q He called you about three or four times; isn’t that true?

A Yes.

Q And when he asked you questions, is it correct that you told him you didn’t know [the answers] when you did know [the answers] because you thought his questions were silly?

A Yes.

Even assuming that Zukofsky has established a prima facie case of her eligibility for an assistant producer job, 0 & M has met its rebuttal burden of articulating a legitimate reason for denying her transfer. In the absence of a showing that Silver-man’s questions were discriminatory or otherwise provocative, the court concludes that Zukofsky’s failure to answer them constituted a non-discriminatory basis for denying her that job.

Zukofsky also testified that at some time, apparently in the late 1970’s, she met with Bob White, Rossini’s supervisor, to discuss an assistant account executive position in the account management department. According to Zukofsky, White had suggested that a position in account management would not meet her interests and that a job. in the creative department would. Zukofsky concurred and did not make any other efforts to obtain that assistant account executive position. Zukofsky, therefore, has failed to establish a prima facie case with regard to her attempt to transfer into the account management department.

Zukofsky also testified regarding her effort to get a job as a copywriter by seeking enrollment in 0 & M’s creative training program. In 1974 and 1975 she submitted a copy test to John Rand, a veteran copywriter at 0 & M who voluntarily met with her and gave her practice writing assignments over a number of months. Zukofsky contends that Rand received her work favorably and incorporated some of her ideas into some of 0 & M’s advertising campaigns. Rand hotly disputed those contentions, stating that, in fact, the ideas he used in the advertising campaign had been his own, fed to Zukofsky in an effort to prod her thinking.

Rand, undisputedly an expert with regard to writing advertising copy, gave a lengthy criticism of Zukofsky’s copy test. He noted that her storyboard, or proposed television commercial, did not immediately draw the product’s name to the audience’s attention, and was otherwise “plodding” and “uninteresting,” failing to take advantage of the visual aspects of the medium. Rand testified that the language of some of her proposed ads was boring or trite while that of other ads was insulting to the reader’s intelligence. The court finds Rand’s testimony to be convincing, particularly since Rand left 0 & M a number of years ago to form his own agency, which has no business dealings with 0 & M.

Mason Clarke, another thirty year advertising veteran, evaluated the copy tests on behalf of Zukofsky. The value of Clarke’s testimony was limited, however, since his expertise admittedly is in the area of advertising art, rather than copy. Although Clarke noted that Zukofsky had suggested some imaginative new products as part of her copy test, he stated that some of her proposed ads “did not thrill [him] all that much,” were “too far-reaching,” contained superfluous language, or were simply “adequate.” Lori Martin, a former supervisor of copywriters at 0 & M, briefly stated that she would have offered a training position in copywriting. The court finds Martin’s testimony to.be of limited weight, both because of its lack of details, and because it may be tainted by animosity resulting from 0 & M’s termination of Martin’s employment. Michael Lipton, the creative director of an advertising agency, gave the most convincing evidence on Zukofsky’s behalf, particularly with regard to her suggested new products. Nevertheless, Lipton had never worked at 0 & M and acknowledged that standards for good copy vary from agency to agency. Moreover, he admitted that some of Zukofsky’s work was “overwritten.”

On balance the court finds persuasive the detailed criticism of John Rand, an experienced copywriter familiar with 0 & M’s standards but with no continuing relationship with that agency. Zukofsky correctly contends that she was applying only for an entry level position, and that the evidence establishes that many of the skills relevant to copywriting are learned on the job. Nevertheless, it is similarly undisputed that competition was keen, even for a beginning position in that field. A minimal level of competence in writing may not have rendered her qualified for the position. Cf. Leiberman v. Grant, 630 F.2d 60 (2d Cir.1980) (showing of minimal level of competence insufficient to establish prima facie case of discriminatory denial of tenure, given highly competitive nature of that employment decision). Although Zukofsky offered into evidence copy tests submitted by a number of individuals, including several females, who did get entry level creative jobs at 0 & M, none of her expert witnesses compared her work with those of the individuals hired as copy writers. The court concludes that the preponderance of the evidence does not establish that 0 & M discriminatorily denied Zukofsky admission to the creative training program or work as a copywriter.

Finally, Zukofsky contends that she discriminatorily was denied an entry level position in the print traffic department which is responsible for forwarding advertising to the printed media on schedule. Elbe Watrous, the director of print traffic, told Zukofsky that she would “have to start as a secretary.” Zukofsky asserts that she did not submit a formal application since “[t]here is no application for a job at 0 & M,” but that she told Watrous to “bear her in mind.”

With respect to the print traffic job, Zukofsky has established a number of elements of the McDonnell Douglas prima facie case. 0 & M does not dispute that she was qualified for that entry level position or for a secretarial position in that department. Under the analysis supra, Zukofsky’s expression of interest to Watrous suffices as an application for a print traffic job. Zukofsky, however, has failed to establish exactly when she inquired about the print traffic job and whether, within a reasonable period of her inquiry, print traffic positions became available for which she should have been considered. Under the McDonnell Douglas analysis, there can be no inference of discrimination where no job opening is shown to exist. The court concludes that Zukofsky has failed to establish that 0 & M, on the basis of her gender, denied her the opportunity to transfer to other departments.

B. Retaliation

Zukofsky also contends that 0 & M has retaliated against her by failing to give her a raise in 1980. 49 Zukofsky contends that 0 & M’s reference to her absenteeism in seeking to justify the raise is a mere pretext for discrimination because Zukofsky had asserted her rights under Title VII.

0 & M asserts that Zukofsky failed to receive a raise because of her excessive absenteeism. Zukofsky’s attendance record for 1979, in fact, was terrible. For example, from January 1, 1979 to March 8, 1979, she missed a .total of eight days for illness, 4V2 days of personal leave and one snow day. She averaged at least one day of sick leave or excused time in each of the two week attendance record periods from March through December 1979.

In a memorandum dated October 1980, protesting her failure to receive a raise, Zukofsky alluded to others in her department “working a four day week” or “arriving at 5 p.m. or later to start their work day.” At trial, Zukofsky did not name any of those individuals. Zukofsky also cites Bill Williams’ poor attendance, but has not introduced his attendance records. Williams received a $1,000 raise early in 1980, based on the recommendation of Evelyn Ward, a supervisor with authority over both Williams and Zukofsky. Zukofsky received a raise in March 1981; Williams received no raise that year and was terminated in March 1982. It is unclear whether Williams was terminated, because of his. absenteeism or as a result of staff cuts, but, in any event, it is apparent that eventually he was treated more severely than Zukofsky.

The court finds that Zukofsky has not established by the preponderance of the evidence that Williams’ attendance was worse than hers during 1979. Accordingly, the court cannot conclude that Zukofsky was denied a raise in 1980 for retaliatory reasons.

V. O & M’s Counterclaim

O & M has filed a counterclaim against plaintiffs for its attorneys’ fees and costs in defending this action. A Title VII defendant is entitled to such an award only if the plaintiff’s claim was “frivolous, unreasonable, or groundless, or that the plaintiff continued to litigate after it clearly became so.” Christiansburg Garment Co. v. EEOC, 434 U.S. 412, 422, 98 S.Ct. 694, 700, 54 L.Ed.2d 648 (1978). The court finds that the claims on which O & M prevailed do not meet that standard, and its counterclaim accordingly must be denied.

Conclusion

For the foregoing reasons,

1) O & M’s motion for class decertification is granted with respect to the claims of transfer, promotion and training discrimination;

2) 0 & M’s motion for class decertification is denied with respect to the class which raises salary claims and which comprises all females (excluding corporate officers) who are, have been, or will be, or were since May 30, 1975 employed by 0 & M as managers and professionals and who have been, are, or continue to be affected by 0 & M’s alleged salary discrimination (“Class I”);

3) the class raising placement/sex segregation claims is redefined as Class II, comprised of all females (excluding corporate officers) who are, have been, or will be or were since May 30, 1975 employed by O & M as managers and professionals in job families having five or more employees, which for the years from 1975 to 1979, inclusive, were more than 80% female and who are, have been, or will be or were since May 30,1975 affected by O & M’s allegedly discriminatory placement decisions and sex segregation of job families;

4) the portion of the complaint alleging • the class placement/sex segregation claim described in paragraph three is dismissed;

5) the portion of the complaint alleging Rossini’s individual claims is dismissed in its entirety;

6) the portion of the complaint alleging Zukofsky’s promotion, training, transfer, retaliation and placement/sex-segregation claims is dismissed;

7) decision is reserved with respect to Zukofsky’s individual salary claim and the class saláry claim, pending the submission of additional evidence as provided herein.

So Ordered.

1 . The defendant, Ogilvy & Mather, Inc., until on or about December 14, 1980, was a New York corporation and a wholly-owned subsidiary of O & M International. Since that date, the defendant has been called Ogilvy & Mather Advertising and has existed as a division of O & M International. Defendant will be referred to throughout as "O & M.”

2 . Within the account management department, O & M recently created the title of "account group directors," for employees assigned to a higher level position than management supervisor. The management supervisors formerly had held the top-ranking supervisory job in account management. The other significant positions in account management are, in descending order of responsibility: account supervisor, account executive, and assistant account executive or staff assistant.

3 . The account coordination and services department also has been called the broadcast operations department or the broadcast production services department. In this decision, the smaller unit in which Zukofsky worked will be referred to as broadcast operations. That department is generally responsible for arranging for the airing of commercials on radio or television.

4 . This action initially was assigned to Judge Brieant and on May 13, 1981 was reassigned to me after Judge Brieant. had recused himself.

5 . At the commencement of trial, plaintiffs moved for reconsideration of the denial of class representative status to Rossini and of the exclusion from the class of female O & M officers. That motion was denied during the trial and plaintiffs have not renewed it in their post-trial submissions. Accordingly, except where otherwise specified, the class at issue in this decision is the class as defined in the April 18, 1979 order.

6 . Rule 23(a), Fed.R.Civ.P., provides:

One or more members of a class may sue or be sued as representative parties on behalf of all only if (1) the class is so numerous that joinder of all members is impracticable, (2) there are questions of law or fact common to the class, (3) the claims or defenses of the representative parties áre typical of the claims or defenses of the class, and (4) the representative parties will fairly and adequately protect the interests of the class.

The Supreme Court noted in General Telephone v. Falcon, supra, 457 U.S. at 157 n. 13, 102 S.Ct. at 2371 n. 13, that the commonality and typicality requirements of Rule 23(a) "tend to merge” with each other and with "the adequacy-of-representation requirement, although the latter requirement also raises concerns about the competency of class counsel and conflicts of interest.” The evidence at trial suggested no reason for reversing Judge Brieant's findings that plaintiffs’ counsel would competently litigate on behalf of the class and that Zukofsky’s individual case posed no conflicts of interest which would preclude her from representing the class.

7 . O & M argued at trial that plaintiffs’ amended complaint failed to allege an individual salary claim on behalf of Zukofsky. The court found that issue to be fairly encompassed by the pretrial order and permitted an amendment to conform the pleadings to the evidence offered at trial with respect to Zukofsky’s individual salary claim. See Fed.R.Civ.P. 15.

8 . O & M argues that the Supreme Court in General Telephone v. Falcon, supra, 457 U.S. at 158 & n. 14, 102 S.Ct. at 2371 & n. 14, found that the commonality requirement was not met simply because the named plaintiff and the unnamed class members all were affected by the defendant’s use of subjective employment criteria. In that case, however, the subjective criteria at issue in the named plaintiffs case pertained to promotion while those at issue in the class case concerned hiring, and the plaintiffs did not establish any overlap between the criteria applied to those distinct employment decisions. See pp. 1135-1136 infra. With regard to the salary claims at issue in the instant case, the same subjective criteria relevant to Zukofsky’s salary increases are also generally applicable to those of the unnamed class members.

9 . With regard to class certification, the salary claims here are distinguishable from the putative class claims in the decisions cited by O & M. In those cases, class certification was denied or limited because the named plaintiff had asserted no common issues more specific than the question of whether the defendant had discriminated against all employees and/or applicants of the named plaintiff’s gender or race. See Eastland v. Tennessee Valley Authority, 704 F.2d 613, 617-18 (11th Cir.1983), cert. denied, — U.S. -, 104 S.Ct. 1415, 79 L.Ed.2d 741 (1984); Nelson v. United States Steel Corp., 709 F.2d 675 (11th Cir.1983); O’Neal v. Riceland Foods, 684 F.2d 577, 581 n. 2 (8th Cir.1982); Grant v. Morgan Guarantee Trust Co. of New York, 548 F.Supp. 1189 (S.D.N.Y.1982).

10 . O & M objected to admission of plaintiffs’ initial statistical report on a number of grounds. The court finds that O & M’s objections are relevant to the weight to be accorded plaintiffs’ initial report, but do not preclude its admission. See pp. 1162-1163 infra.

11 . "‘Plaintiff’s burden of proof to demonstrate the existence of this common question entails more than the simple assertion of its existence, but less than a prima facie showing of liability.’ ’’ Nelson v. United States Steel, supra, 709 F.2d at 680, (quoting Gilchrist v. Bolger, 89 F.R.D. 402, 406 (S.D.Ga.1981)). Because Zukofsky has made such a showing, the class salary claim in the instant case is distinguishable from the class claims in Hawkins v. Fulton County, 95 F.R.D. 88, 93 (N.D.Ga.1982) and Nation v. Winn-Dixie Stores, Inc., 95 F.R.D. 82, 87-88, (N.D.Ga.1982), cited by O & M. In those cases, the named plaintiffs offered no evidence suggesting that unnamed class members might have been affected by the challenged subjective decision-making procedures.

12 . Rule 23(b) provides in relevant part:

An action may be maintained as a class action if the prerequisites of subdivision (a) are satisfied, and in addition:

(1) the prosecution of separate actions by or against individual members of the class would create a risk of

(A) inconsistent or varying adjudications with respect to individual members of the class which would establish incompatible standards of conduct for the party opposing the class, or

(B) adjudications with respect to individual members of the class which would as a practical matter be dispositive of the interests of the other members not parties to the adjudications or substantially impair or impede their ability to protect their interests; or

(2) the party opposing the class has acted or refused to act on grounds generally applicable to the class, thereby making appropriate final injunctive relief or corresponding declaratory relief with respect to the class as a whole; or

(3) the court finds that the questions of law or fact common to the members of the class predominate over any questions affecting only individual members, and that a class action is superior to other available methods for the fair and efficient adjudication of the controversy____

13 . Rule 23(c)(2) provides:

In any class action maintained under subdivision (b)(3), the court shall direct to the members of the class the best notice practicable under the circumstances, including individual notice to all members who can be identified through reasonable effort. The notice shall advise each member that (A) the court will exclude him from the class if he so requests by a specified date; (B) the judgment, whether favorable or not, will include all members who do not request exclusion; and (C) any member who does not request exclusion may, if he desires, enter an appearance through his counsel.

14 . O & M, on the other hand, introduced statistical evidence indicating that since 1976, women' professionals at O & M were transferred in numbers generally proportionate to the percentage of women in O & M’s professional staff.

15 . Those job families or groupings of job titles were created by plaintiffs’ counsel and do not exist independently at O & M. The job families - correspond generally, but not completely, to the departments at O & M, and group together certain job titles sometimes categorized as independent units by O & M. Nevertheless, on the basis of the testimony describing O & M’s operations, the court concludes that the job families, as grouped by plaintiffs, represent a reasonable indicator of the relationships among various positions at O & M.

16 . The evidence submitted by both parties generally indicates O & M employment patterns no later than 1980, when discovery of class data in this action, for the most part, terminated. O & M contends that the relevant period for analysis of plaintiffs’ claims begins no earlier than May 30, 1975. See footnote 23. In Association Against Discrimination in Employment, Inc. v. City of Bridgeport, 647 F.2d 256, 274 (2d Cir.1981), ce rt. denied, 455 U.S. 988, 102 S.Ct. 1611, 71 L.Ed.2d 847 (1982), the Second Circuit stated (emphasis in original; citations omitted):

Where, however, the defendant has engaged in a continuous policy of discrimination, acts in furtherance of that policy are not viewed in isolation. In such circumstances if the [Equal Employment Opportunity Commission] charge has been filed no later than 300 days after the last act by the defendant pursuant to its policy, the plaintiff may recover for earlier acts of discrimination as well____ In Acha v. Beame, we stated this principle as follows:

To succeed at trial, the appellants must be able to demonstrate a Title VII violation occurring after the effective date of [Title VII] and within the period for the statute of limitations, or 300 day charge-filing period...:.

A continuously maintained illegal employment policy may be the subject of a valid complaint until a specified number of days after the last occurrence of an instance of that policy____ Furthermore, where an illegal policy is so maintained, relief for injuries sustained even before the beginning of the limitations period is appropriate.

570 F.2d [57], 65 [ (2d Cir.1978) ].

Thus if plaintiffs can prove their allegation that after May 30, 1975, O & M engaged in acts in furtherance of a continuous policy of discrimination, they may also be able to establish liability for discriminatory acts prior to that date. See Association v. Bridgeport, supra, 647 F.2d at 275.

17 . For example, for 1978, plaintiffs’ statistics indicate the following breakdown for- the account management and the creative departments:

Average Salaries by Job Family and Title

Female Male

Job Family Job Title Count Count Per Cent Female

Account Staff Assistant 11 4 73.33

Management Account Management Trainee 0 1 .00

Art/Contract 0 1 .00

Asst. Acct. Exec. 22 8 73.33

Acct. Executive 24 25 48.98

Acct. Supervisor 1 17 5.56

Copy/Contract 0 2 .00

Creative-Copy Copywriter 7 15 31.82

Copy Supervisor 2 0 100.00

Sr. Copywriter 0 4 .00

Creative-Other Asst. Photo Research 1 0 100.00

Casting Ass't 1 0 100.00

Asst. Music Producer 1 0 100.00

Photo Researcher 1 0 100.00

Copy Researcher 1 0 100.00

Stylist 1 0 100.00

Styling Supervi1 0 100.00

Typographer 0 1 .00

Type Director 0 1 .00

Casting Director 0 2 .00

Sr. Type Director 1 0 100.00

Admin. Manager 1 0 100.00 •

Creative-TV .Production Asst. Producer 4 1 80.00

Production Asst. 1 0 100.00

Producer 3 6 33.00

Creative-Art Art Trainee 1 2 33.33

Asst. Sketchman 1 0 100.00

Asst. Art Director 6 6 50.00

Art Director 6 23 20.69

Sr. Art Director 0 4 .00

Sketchman 0 4 .00

Art Supervisor 0 2 .00

ManagerSketchman 0 1 .00

Art Group Head 0 1 .00

Creative-Copy Jr. Copywriter 3 5 37.50

18 . The court does not disagree with plaintiffs’ assertion that any unnamed class member may well find it personally or professionally difficult to testify against her employer. That situation, however, does not relieve the named plaintiff of the burden of meeting the typicality requirement of Rule 23. Cf. General Telephone v. Falcon, supra, 457 U.S. at 161, 102 S.Ct. at 2373 (’’[A] Title VII class action, like any other class action, may only be certified if the trial court is satisfied, after a rigorous analysis, that the prerequisites of Rule 23(a) have been satisfied”). The court notes, moreover, Title VII’s prohibition of employer retaliation, 42 U.S.C. § 2000e-3(a), invoked by both Zukofsky and Rossini on their own behalf.

19 . In East Texas Motor Freight System, Inc. v. Rodriguez, 431 U.S. 395, 406 n. 12, 97 S.Ct. 1891, 1898 n. 12, 52 L.Ed.2d 453 (1977), the Supreme Court stated:

Obviously, a different case would be presented if the District Court had certified a class and only later had it appeared that the named plaintiffs were not class members or were otherwise inappropriate class representatives. In such a case, the class claims would have already been tried, and, provided the initial certification was proper and decertification not appropriate, the claims of the class members would not need to be mooted or destroyed because subsequent events or the proof at trial had undermined the named plaintiffs’ individual claims.

See Sirota v. Solitron Devices, Inc., 673 F.2d 566 (2d Cir.1982), cert. denied, 459 U.S. 838, 103 S.Ct. 86, 74 L.Ed.2d 80 (1983). In the instant case, the class transfer claim is decertified not as a result of any undermining of Zukofsky’s individual transfer claim. Rather, fatal to the class transfer claims is plaintiffs’ failure at any stage of these proceedings to point to evidence suggesting the existence of unnamed class members with transfer claims.

20 . Those two broad Title VII doctrines were described by the Supreme Court in International Brotherhood of Teamsters v. United States, 431 U.S. 324, 335 n. 15, 97 S.Ct. 1843, 1854 n. 15, 52 L.Ed.2d 396 (1977):

“Disparate treatment" ... is the most easily understood type of discrimination. The employer simply treats some people less favorably than others because of their race, color, religion, sex, or national origin. Proof of discriminatory motive is critical, although it can in some situations be inferred from the mere fact of differences in treatment____

Claims of disparate treatment may be distinguished from claims that stress “disparate impact." The latter involve employment practices that are facially neutral in their treatment of different groups but that in fact fall more harshly on one group than another and cannot be justified by business necessity____ Proof of discriminatory motive, we have held, is not required under a disparate-impact theory-

21 . A similar analysis indicates that Zukofsky's individual salary claim, see p. 1130, supra, and her individual placement/sex segregation claim, see p. 1137, infra, also do not present predominating common issues with the class promotion claims. Neither of those individual claims therefore can serve as a basis for Zukofsky’s representation of the promotion class.

22 . Decertification of the promotion class is appropriate because at no stage in these proceedings has Zukofsky met the Rule 23 requirements for serving as representative for that class. See footnote 17.

23 . On April 14, 1978, Zukofsky filed with the Equal Employment Opportunity Commission (“EEOC”) a charge alleging sex discrimination by O & M. Under 42 U.S.C. § 2000e-5(e), in jurisdictions like New York having agencies charged with enforcing local or state fair employment laws, an EEOC charge must be filed within 300 days of the alleged unlawful discrimination practice. The limitations date relevant to Zukofsky’s claim, however, is earlier than 300 days before her EEOC charge. Where a class action complaint has been filed, but class certification is denied, the complaint tolls the statute of limitations for members of the purported class who act in a timely fashion to protect their rights. See Crown Cork & Seal Co. v. Parker, 462 U.S. 345, 103 S.Ct. 2392, 76 L.Ed.2d 628 (1983). Thus the filing of the complaint by Rossini tolled the statute of limitations for Zukofsky, a member of the class Rossini unsuccessfully sought to represent. Accordingly, the period relevant to Zukofsky’s claims begins, at the latest, on May 30, 1975, rather than on the date 300 days before Zukofsky’s April 1978 EEOC charge.

24 . 42 U.S.C. § 2000e-2(a)(2) specifies as an unlawful employment practice an employer’s decision

to limit, segregate, or classify his employees or applicants for employment in any way which would deprive or tend to deprive any individual of employment opportunities or otherwise adversely affect his status as an employee, because of such individual’s race, color, religion, sex, or national origin.

25 . For example, in 1978, certain of O & M’s job families, as grouped by plaintiffs, were staffed as follows:

Number of Females Number of Males Per Cent Female

Accounting-EST-Billing Print EST/Bill-er 2 2 100.00

Spot EST/Biller 2 0 100.00

Production Biller 5 0 100.00

Asst. Supvsr. Spot Billing 2 0 100.00

Supvsr. Print EST Billing 2 0 100.00

Mgr. Media EST & Billing 1 1 50.00

Supvsr. Print Prod. Billing 0 1 .00

Accounting-Gen. Acctg. ‘ Payroll Clerk 1 0 100.00

Jr. Accountant 0 1 .00

Cost Accountant 1 0 100.00

Benefits Asst. 1 1 50.00

Bookkeeper 1 0 100.00

Supvsr. Accts. Payable 1 0 100.00

Prod. Accts. Payable Supr. 1 0 100.00

Auditor 0 1 .00

Payroll Supvsr. 1 0 100.00

Chief Accountant 0 1 .00

Supvsr, Accts. Receiv. 1 0 100.00

Accountant 0 1 .00

Controller 1 0 100.00

Personnel EEO Coordinator 0 1 .00

Dining Room Manager1 ,0 100.00

Asst. SupvsrTyping Center 1 0 100.00

Personnel . Work/Educ. Trainee 2 0 100.00

Personnel Asst. 3 0 100.00

Switchboard Supvsr. 1 0 100.00

Supvsr/Pers. Admin. 1 0 100.00

Personnel Recruiter 1 0 100.00

Manager-Benefits Admin. 0 1 .00

Creative-Other Asst. Photo Research 1 0 100.00

Casting Ass't 1 0 100.00

Asst Music Producer 1 0 100.00

Photo Research-0 100.00

Copy Researcher • 1 0 100.00

Stylist 1 0 100.00

Styling Supvsr. 1 0 100.00

Typographer 0 1 .00

Type Director 0 1 .00

Casting Director 0 2 .00

Sr. Type Director 1 0 100.00

Admin. Manager 1 0 100.00

Broadcast Prod. Serv-Brdct Ford Asst. Dir. Brdct-Oper. 1 0 100.00

Bdct. Forward 10 1 90.91

Supvsr-Bdct Estim. 1 0 .00

Mgr. Brocst Forwarding 1 0 100.00

26 . 0 & M’s expert studied hiring, promotions, and transfers with respect to job titles having more than ten employees. He expressed the opinion that smaller groups could not be the subject of meaningful analysis of allegations of discrimination with respect to those practices. With regard to the placement/sex segregation claim, the court will not initially limit the class definition to the extent suggested by O & M's expert, but will set a cutoff at job families having five or more employees.

27 . Decertification of the training class is appropriate because at no stage in these proceedings has Zukofsky met the Rule 23 requirements as interpreted in General Telephone v. Falcon, supra, for serving as representative of that class. Moreover, although more than a year before trial O & M requested decertification in light of General Telephone v. Falcon, supra, plaintiffs have not proposed intervention of another named plaintiff competent to represent the putative class alleging training or promotion discrimination. Cf. Rivers v. Califano, 86 F.R.D. 41 (S.D.N.Y.1980) (where intervenors had live individual claims, class certification permitted despite mootness of claims of original named plaintiffs).

28 . In EEOC v. TWA, supra, the plaintiffs alleged age discrimination in violation of the Age Discrimination in Employment Act, 29 U.S.C. § 621 et seq. (“ADEA”). The Title VII McDonnell Douglas-Burdine formula of burdens of proof and production repeatedly has been applied by analogy to cases arising under the ADEA. See Pena v. Brattleboro Retreat, 702 F.2d 322, 323 (2d Cir.1983) and cases cited therein.

29 . Rossini contends that her well-evaluated performance on the Amex service establishment account shortly after she received the “4" category rating on Drain Power, largely negates any inferences to be drawn from that critical evaluation. On the contrary, the dramatic improvement in Rossini’s evaluation corroborates 0 & M’s position that Rossini lacked the skills relevant to accounts like Drain Power or the card division of Amex, but that she possessed other strengths which permitted her to function effectively on the Amex service establishment division.

30 . Rossini has argued that her ability to cite specific instances of discrimination or to show the pretextual nature of O & M’s defenses, unfairly has been limited by a protective order, signed by Judge Brieant on June 6, 1980. That order granted only Rossini’s counsel the ability to review personnel files of males at O & M allegedly comparable to . Rossini, and barred Rossini herself from reviewing the files. - The court finds, however, that despite Rossini’s protestations to the contrary, the protective order was entered on consent, as was indicated by Judge Brieant’s handwritten amendments to the order. Moreover, even if that notation as to consent were erroneous, Rossini’s failure to move to amend that order prior to trial, precludes her from disputing it at this time. Cf. Zenith Radio Corp. v. Matsushita Electric Indus trial Co., 529 F.Supp. 866 (E.D.Pa.1981) (by proceeding under blanket protective order throughout litigation, party waived right to seek wholesale declassification of documents protected thereunder).

31 . Where an employer has a “promote from within policy," as generally is the case at O & M, the percentage of women or minorities in the upper.level positions often may be compared, in proving claims of promotion discrimination, against the percentage of those groups in the employer’s workforce as a whole. See Carroll v. Sears, Roebuck & Co., 708 F.2d 183 (5th Cir.1983).

32 . Rossini did provide evidence indicating the expected numbers of female hires for account supervisor and account executives. Because O & M in many areas observed a promote from within policy, it is likely that many account executive jobs were filled by promotions or transfers from within O & M and that many account supervisor jobs were filled by promotion from within the O & M account management department. Accordingly, the figures regarding hires are of limited utility to the court in its effort to discern any pattern of discriminatory treatment.

33 . O & M failed to introduce Newbrand’s personnel file or other evidence indicating his account management experience. Rossini, however, does not dispute O & M's allegations with regard to that experience.

34 . Rossini cites other criticism of Berkowitz written as part of a generally favorable evaluation after his appointment to TWA. Those post-appointment comments are of limited relevance to the question of whether the assignment of Berkowitz to that account was discriminatory. See pp. 1149-1150, supra.

35 . Rossini challenges the discovery order denying her access to all officer personnel files for purposes of her individual case. Rossini's counsel were provided with personnel files of more than twenty male O & M officers regarded by Rossini as comparable to her, including files of any male officer who received one of the disputed assignments or promotions. The court finds that Rossini has had access to those files likely to contain or lead to the discovery of descriptive evidence relevant to her case. In view of 0 & M’s strong showing of non-discriminatory reasons for each of the challenged decisions, the court finds that Rossini’s case would not be significantly advanced by any statistical evidence that could be gleaned from broader access to the officer files. See pp. 1147-1148, supra. Accordingly, the court adheres to the decision to deny plaintiffs access to additional files of O & M officers for purposes of Rossini’s individual case.

36 . Although evidence was introduced with regard to the numbers of male and female officers at O & M, neither the complaint nor the pre-trial order alleges that Rossini’s election to vice-president was discriminatorily delayed. Accordingly, the court does not consider that issue.

37 . The Equal Pay Act provides in relevant part:

(1) No employer having employees subject to any provisions of this section shall discriminate, within any establishment in which such employees are employed, between employees on the basis of sex by paying wages to employees in such establishment at a rate less than the rate at which he pays wages to employees of the opposite sex in such establishment for equal work on jobs the performance of which requires equal skill, effort, and responsibility, and which are performed under similar working conditions, except where such payment is made pursuant to (i) a seniority system; (ii) a merit system; (iii) a system which measures earnings by quantity or quality of production; or'(iv) a differential based on any other factor other than sex____

38 . For examples of the detailed job content evaluations relevant to an EPA prima facie case, see Usery v. Columbia University, supra, 568 F.2d at 956-961; Brennan v. Owensboro-Daviess County Hospital, 523 F.2d 1013 (6th Cir.1975).

39 . Rossini introduced the following data with regard to the salaries of the allegedly comparable males, identified by number for purposes of confidentiality.

Salaries in Thousands of Dollars

Date #87 #94 #88 #89 #93 #92 #76 Rossini

12/70 20.5 22.0 16.0 22.0 16.0 14.0 — 13.0

12/71 20.5 24.0 16.0 24.0 18.0 16.0 14.0 15.0

12/72 22.5 28.0 20.0 29.0 20.5 20.0 19.0 17.0

12/73 25.0 28.0 22.0 29.0 — 24.0 21.0 19.5

12/74 28.0 31.0 25.0 32.0 — 26.0 27.0 22.0

12/75 30.5 33.0 27.5 35.0 35.0 32.0 30.0 24.5

12/76 33.5 35.0 30.0 38.0 — 45.0 35.0 27.5

12/77 36.5 35.0 33.0 43.0 — 40.0 48.0 30.5

12/78 40.0 40.0 38.5 48.0 44.0 — 48.0 35.0

12/79 45.0 45.0 45.0 48.0 48.0 — 54.0 45.0

40 . Although the 1976 EEOC charge was filed under the name of Rossini’s attorney, there is evidence that O & M prior to the institution of this action learned that Rossini ^ad filed the EEOC charge.

41 . The court therefore does not reach the issue of whether a Title VII challenge to subjective employment criteria can be evaluated under the disparate impact model. Compare Pouncy v. Prudential Insurance Co., 668 F.2d 795 (5th Cir.1982) with Rowe v. Cleveland Pneumatic Co. Numerical Control, 690 F.2d 88 (6th Cir.1982). However, even assuming that a disparate impact analysis appropriately may be applied here and that plaintiffs will establish the adverse impact of the subjective salary criteria, the court finds that O & M has met its burden in defending such a claim. When it is shown that a facially neutral employment practice has a disproportionate impact on women or minorities, the employer must demonstrate the job relatedness of that practice. See Abermarle Paper Co. v. Moody, 422 U.S. 405, 95 S.Ct. 2362, 45 L.Ed.2d 280 (1975); Rowe v. Cleveland Pneumatic, supra, 690 F.2d at 93-94. In the instant case, O & M has amply demonstrated that in setting salaries it relies to some extent on objective factors such as experience, supervisory responsibilities, and understanding of various relevant areas of advertising (e.g. numerical ability in account management, see pp. 1141-1143, supra). O & M's reliance in setting salaries on evaluations of certain subjective factors — creativity, initiative, leadership — is job-related given the nature of the work at issue. See Coser v. Moore, 587 F.Supp. 572 (E.D.N.Y.1983), aff’d, 739 F.2d 746 (2d Cir.1984); cf. Rowe v. Cleveland Pneumatic, supra, 690 F.2d at 95 (no showing of job relatedness where foremen did not indicate factors relied upon in their subjective determination of whether to hire a former employee). But see Bartholet, Application of Title VII to Jobs in High Places, 95 Harv.L.Rev. 945 (1982).

42 . The analyses in Melani and Sobel therefore did not require the detailed analysis of job content appropriate to an EPA claim. See pp. 1154-1155, supra.

43 . Accordingly, the court finds it appropriate to cite the Robbins report’s description of serial correlation even though that report is not in evidence.

44 . The plaintiff class asserting salary claims comprised women professionals and managers at O & M during the relevant period. The designations "professionals” and "managers” are general categories taken from the Title VII recordkeeping system and are not to be confused with particular job titles or departments as designated by O & M.

45 . The green report appears to have been admitted initially, prior to testimony indicating that newly gathered education data was used therein. That oral order therefore is vacated to the’ extent that it is inconsistent with this decision.

46 . At trial, the court excluded Pl. Ex. 285, which modified one of plaintiffs’ regression studies. Because Pl. Ex. 285, like the green report was based upon educational information plaintiffs gathered during trial and not included in the data bases exchanged before trial, the court adheres to its ruling excluding that study.

47 . Table VIII-2

Promotions or Transfers of Female Professionals at Ogilvy & Mather, 1976-79

Year Promoted or Transferred at O&M Total Females Expected Number of Females Promoted or Transferred

1976 69 • 43 36

1977' 67 43 35

1978 100 63 53*

1979 110 64 56

Total 1976-79 346 213 180*

An * indicates a difference between expected and actual promotions and transfers that is statistically significant at the .05 probability level.

The expected number of female promotions and transfers is estimated as the fraction that women professionals were of total professionals on December 31 of the preceeding year multiplied times the total number of promotions of men and women during the year.

Table VI-3

Accessions by Job Title of Female Professionals at Ogilvy & Mather, 1975-79

Job Title Accessions at O&M Total Females Expected Number of Female Accessions

Art Director 37 12 16

Assistant Art Director 31 17 13

Art Trainee 13 6 5

Account Executive 114 59 55

Assistant Account Executive 100 64 45*

Copy Writer 30 10 13

Junior Copy Writer 22 9 9

Media Planner 36 15 18

Assistant Media Planner 47 23 19

Job titles with more than ten hires.

An * indicates a difference between actual and expected accessions that is statistically significant at the .05 probability level.

Expected female accessions is estimated as follows:

The percentage that women were of all professional employees at O&M in 1978, 52.8 percent, is first multiplied by the fraction that transfers and promotions were of total accessions. To this is added the percentage that women were of all professionals in the New York City Advertising Agencies in 1978, 37.5 percent, which in turn has been multiplied by the portion that hires were of total accessions.

48 . Table V-3

Employment of Female Professionals at Ogilvy & Mather by Department, 1979

Department Actual Number Employed at O&M Total Females Expected Number Female Employees (a) (b) (c)

Account Management 128 73 48* 53* 68

Accounting 31 20 13* 13* 16

Broadcast Production Ser30 20 11* 12* 16

Creative 39 33 36 46

Media 51 28 19* 21* 27

Print Production Services 13 7 5 5 7

Departments with more than ten employees.

An * indicates a difference between actual and expected employment that is statistically significant at the .05 probability

Expected female employment is estimated as follows:

(a) from the percentage that women were of professionals in New York City Advertising Agencies in 1978, 37.5 percent.

(b) from the percentage that women were of professionals in advertising in the United States. From the Equal Employment Opportunity Commission, [Employment Analysis Report Program 1978], 41.3 percent.

(c) from the percentage that women were of all professional employees at O&M in 1978, 52.8 percent.

Sources: For (a): Minority Employment in the Advertising Business. January 1979. For (b): Employment Analysis Report Program, EEOC, 1978. For (c): O&M personnel records.

49 . In her post-trial memoranda, Zukofsky suggests several other allegedly retaliatory actions by O & M. At trial, however, Zukofsky sought leave to amend her complaint solely to add the claim of retaliatory denial of a salary increase in 1980. Accordingly, the court will consider only that claim of recrimination.