Opinion
2:10-cv-1283
07-13-2015
RUDOLPH A. KARLO, MARK K. MCLURE, WILLIAM S. CUNNINGHAM, JEFFREY MARIETTI, and DAVID MEIXELSBERGER, Plaintiffs, v. PITTSBURGH GLASS WORKS, LLC, Defendant.
MEMORANDUM OPINION AND ORDER OF COURT
Pending before the Court are the (1) DEFENDANT'S RENEWED MOTION TO BAR PROPOSED EXPERT OPINION OF ANTHONY G. GREENWALD RELATED TO PURPORTED IMPLICIT SOCIAL BIAS (ECF No. 380); (2) DEFENDANT'S RENEWED POST-DECERTIFICATION MOTION TO BAR DR. MICHAEL CAMPION'S STATISTICAL ANALYSIS (ECF No. 381); (3) DEFENDANT'S RENEWED MOTION TO BAR DR. CAMPION'S EXPERT OPINION ON REASONABLE HUMAN RESOURCE PRACTICES (ECF No. 382); and (4) DEFENDANT'S MOTION TO BAR PURPORTED REBUTTAL EXPERT OPINION OF DAVID DUFFUS (ECF No. 383), all of which were filed on behalf of Pittsburgh Glass Works, LLC ("PGW"). The issues have been fully briefed and well-argued by the parties in their memoranda and attached exhibits (ECF Nos. 392, 393, 394, 395, 413, 414, 415, 416, 423, 427, 428, 429). The Court heard oral argument on January 13, 2015 at which counsel for Plaintiffs Rudolph A. Karlo, Mark K. McLure, William S. Cunningham, Jeffrey Marietti, and David Meixelsberger presented additional authority in support of their position; PGW has since filled a RESPONSE TO SUPPLEMENTAL AUTHORITY PRESENTED BY PLAINTIFFS AT ORAL ARGUMENT (ECF No. 432-1). The motions are ripe for disposition.
I. Background
The Court has previously detailed the background of this action in its forty-four page Memorandum Opinion and Order issued on March 31, 2014, and it need not be reprised here. See Mem. Op., ECF No. 343 at 3-26. Since that time, PGW has filed three motions for summary judgment relating to the individual disparate impact and disparate treatment claims of Karlo, McLure, Cunningham, Marietti and Meixelsberger as well as the individual retaliation claims of Karlo and McLure; and four renewed motions to bar Plaintiffs' experts, Anthony G. Greenwald, Ph.D., Michael Campion, Ph.D., and David Duffus, CPA/ABV/CFF, CFE. This Memorandum Opinion will address only the renewed expert-related motions, which the Court will now address.
II. Legal Standard
"Under the Federal Rules of Evidence, a trial judge acts as a 'gatekeeper' to ensure that 'any and all expert testimony or evidence is not only relevant, but also reliable." Pineda v. Ford Motor Co., 520 F.3d 237, 244 (3d Cir. 2008) (citing Kannankeril v. Terminix Int'l, Inc., 128 F.3d 802, 806 (3d Cir. 1997)). Thus, whenever a party seeks to admit expert testimony at trial, the district court must make an initial preliminary determination "that the requirements of Fed. R. Evid. 702 have been met." Magistrini v. One Hour Martinizing Dry Cleaning, 68 F. App'x 356, 356 (3d Cir. 2003) (citing Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 592 (1993)).
Federal Rule of Evidence 702 provides that "[a] witness who is qualified as an expert by knowledge, skill, experience, training, or education may testify in the form of an opinion or otherwise if: (a) the expert's scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue; (b) the testimony is based on sufficient facts or data; (c) the testimony is the product of reliable principles and methods; and (d) the expert has reliably applied the principles and methods to the facts of the case." Moreover, the United States Court of Appeals for the Third Circuit has interpreted Rule 702 as having "three major requirements: (1) the proffered witness must be an expert, i.e., must be qualified; (2) the expert must testify about matters requiring scientific, technical or specialized knowledge; and (3) the expert's testimony must assist the trier of fact." Pineda, 520 F.3d at 244 (citing Kannankeril, 128 F.3d at 806). Our Court of Appeals has also "interpreted the second requirement to mean that 'an expert's testimony is admissible so long as the process or technique the expert used in formulating the opinion is reliable.'" Id. (citing Kannankeril, 128 F.3d at 806 (quoting In re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 742 (3d Cir. 1994)).
The qualification prong of Rule 702 "requires 'that the witness possess specialized expertise.'" Id. (citing Schneider ex rel. Estate of Schneider v. Fried, 320 F.3d 396, 404 (3d Cir. 2003)). The Court of Appeals has "interpreted [this] requirement liberally." Id. (citing Schneider, 320 F.3d at 404; Paoli, 35 F.3d at 741). It has explained that "a 'broad range of knowledge, skills, and training'" can suffice to "'qualify an expert.'" Id. (quoting Paoli, 35 F.3d at 741). "This liberal policy of admissibility extends to the substantive as well as the formal qualifications of experts." Pineda v. Ford Motor Co., 520 F.3d 237, 244 (3d Cir. 2008) (citing Paoli, 35 F.3d at 741). The Court of Appeals has repeatedly "stated that 'it is an abuse of discretion to exclude testimony simply because the trial court does not deem the proposed expert to be the best qualified or because the proposed expert does not have the specialization that the court considers most appropriate.'" Kannankeril, 128 F.3d at 809 (quoting Holbrook v. Lykes Bros. S.S. Co., 80 F.3d 777, 782 (3d Cir. 1996)).
"The second requirement is that of reliability." Meadows v. Anchor Longwall & Rebuild, Inc., 306 F. App'x 781, 788 (3d Cir. 2009). In evaluating whether a particular methodology is reliable, the Court of Appeals has identified several factors district courts should consider:
(1) whether a method consists of a testable hypothesis; (2) whether the method has been subject to peer review; (3) the known or potential rate of error; (4) the existence and maintenance of standards controlling the technique's operation; (5) whether the method is generally accepted; (6) the relationship of the technique to methods which have been established to be reliable; (7) the qualifications of the expert witness testifying based on the methodology; and (8) the non-judicial uses to which the method has been put.Id. (citing Paoli, 35 F.3d at 742 n.8). However, these factors "'are neither exhaustive nor applicable in every case.'" Id. (quoting Kannankeril, 128 F.3d at 806-07). Whether they are relevant "depend[s] on the nature of the issue, the expert's particular expertise, and the subject of his testimony." Kumho Tire Co. v. Carmichael, 526 U.S. 137, 150 (1999). This inquiry remains inherently "flexible." Pineda, 520 F.3d at 247 (quoting Daubert, 509 U.S. at 594). The question "is not whether a particular scientific opinion has the best foundation, or even whether the opinion is supported by the best methodology or unassailable research." In re TMI Litig., 193 F.3d 613, 665 (3d Cir. 1999), amended, 199 F.3d 158 (3d Cir. 2000). The goal is simply "to make certain that an expert, whether basing testimony upon professional studies or personal experience, employs in the courtroom the same level of intellectual rigor that characterizes the practice of an expert in the relevant field." Kumho Tire, 526 U.S. at 152. Therefore, the focus must remain "on principles and methodology, not on the conclusions generated by the principles and methodology." In re TMI Litig., 193 F.3d at 665 (citing Kannankeril, 128 F.3d at 806). "'The analysis of the conclusions themselves is for the trier of fact when the expert is subjected to cross-examination.'" Id. (quoting Kannankeril, 128 F.3d at 806).
Even so, "'conclusions and methodology are not entirely distinct from one another.'" In re TMI Litig., 193 F.3d 613, 665 (3d Cir. 1999), amended, 199 F.3d 158 (3d Cir. 2000) (quoting General Elec. Co. v. Joiner, 522 U.S. 136, 146 (1997)). On the contrary, when assessing reliability, an expert's conclusions must be examined to decide "'whether they could reliably flow from the facts known to the expert and the methodology used.'" Id. (quoting Heller v. Shaw Indus., Inc., 167 F.3d 146, 153 (3d Cir. 1999)). If there is a "'too great a gap between the data and the opinion proffered,'" the opinion should be excluded. Id. (quoting Joiner, 522 U.S. at 519). The Court is thus not required to "admit opinion evidence that is connected to existing data only by the ipse dixit of the expert." Joiner, 522 U.S. at 146.
"The third element under Rule 702, namely, whether the expert testimony would assist the trier of fact, "goes primarily to relevance.'" Meadow, 306 F. App'x at 790 (quoting Lauria v. Amtrak, 145 F.3d 593, 599 (3d Cir. 1998). This element requires that "[t]he expert's testimony must 'fit' under the facts of the case so that 'it will aid the jury in resolving a factual dispute.'" Id. And "[t]he standard for the factor is not high; it is met when there is a clear 'fit' connecting the issue in the case with the expert's opinion that will aid the jury in determining an issue in the case." Id. (citing Lauria, 145 F.3d at 600 (quoting Paoli, 35 F.3d at 745)); see also In re TMI Litig., 193 F.3d at 670 ("[A]dmissibility depends, in part, on a connection between the expert opinion offered and the particular disputed factual issues in the case.") (citation omitted). Therefore, "expert testimony based on assumptions lacking factual foundation in the record is properly excluded" under the fit requirement in addition to the reliability requirement. Meadows, 306 F. App'x at 790 (citing Stecyk v. Bell Helicopter Textron, Inc., 295 F.3d 408, 414 (3d Cir. 2002)).
III. Discussion
A. Dr. Anthony G. Greenwald (Implicit Social Bias)
Dr. Anthony G. Greenwald, Ph.D is a tenured faculty member at the University of Washington in its Department of Psychology, where he has been an active member of its teaching and research faculty since 1986. Previously, Dr. Greenwald served as a tenured faculty member at The Ohio State University in its Department of Psychology from 1965 to 1986. Among the many accomplishments throughout his career, Dr. Greenwald has published more than one-hundred-and-eighty refereed journal articles and book chapters in the areas of social psychology, cognitive psychology, and research methodology; received six awards for career research achievements; and served on journal editorial boards of prominent publications.
Dr. Greenwald has been retained by Plaintiffs' counsel to provide his opinion "in the area of social psychological research on attitudes, prejudices, and stereotypes," which includes the topic of implicit bias—i.e., "a lay designation for mental processes that function outside of conscious awareness." Greenwald Exp. Rep. at 2, ECF No. 380-5. In that capacity, Dr. Greenwald has authored an expert report in which he uses the term "implicit bias" as "an informal reference to the relevant scientific work on attitudes and stereotypes." Id. at 4. Citing to that report, Plaintiffs submit that they offer Dr. Greenwald's opinion to "'provide a framework that can aid a judge or jury in evaluating the facts of this case to better understand the evidence as it relates to discriminatory intent, to counteract common misconceptions concerning the character of discriminatory intent, and to determine whether the Plaintiffs' ages substantially motivated the defendants' [sic] actions outlined in the Complaint.'" Pls.' Br. in Opp. at 4 (quoting Greenwald Exp. Rep. at 6, ECF No. 395-2).
Moreover, in his report, Dr. Greenwald explains that his original research in the area of implicit social cognition includes the "invention and development of a specific research method—the Implicit Association Test ('IAT")." Greenwald Exp. Rep. at 2, ECF No. 380-5. One district court has described the IAT in a race-based employment discrimination action as follows:
[I]t is a computerized exercise based on automatic word associations that test subjects make when shown pictures of individuals of various genders, races, and ethnicities. The photos are displayed for only milliseconds; then the test subjects are asked to make an association. If a test-taker responds more quickly, say, to the pairing of photographs of African-American faces with negative character trait words than to the pairing of European-American faces with the sameJones v. Nat'l Council of Young Men's Christian Associations of the United States of Am., 34 F. Supp. 3d 896, 899 (N.D. Ill. 2014) (internal citation and quotation marks omitted). The IAT is apparently "considered an 'implicit' measure because it infers the strength of a mental association that links a social category (such as race, gender, or age group) with a trait (i.e., a stereotype) from testing procedures that are influenced by those associations in a manner not discerned by the respondents." Greenwald Exp. Rep. at 2, ECF No. 380-5. According to Dr. Greenwald, the IAT has "been successfully used as an implicit measure for a wide variety of mental associations that underlie stereotypes and social attitudes," with his own research including the "study of implicit biases involved in age attitudes." Id.
negative traits, the test-taker is said to exhibit an implicit negative stereotype toward African-Americans.
Dr. Greenwald also describes the IAT as a well-known and well-respected research method:
The psychometric properties of IAT measures have been validated with tens of thousands of participants in laboratory research results. Variations of the IAT have been taken more than 14 million times at the on-line educational site, www.implicit.harvard.edu/implicit. Many social cognition experts have used the IAT as a method in their own research. No method for measuring implicit bias is more widely used than the IAT. IAT measures have been subjected to repeated empirical testing and peer review. There exists near unanimous agreement among social psychologists as to the validity of the IAT as a method for implicit measurements of attitudes and stereotypes.Greenwald Exp. Rep. at 4, ECF No. 380-5. Dr. Greenwald has apparently extrapolated that the "strongest bias that has been demonstrated in extensive Internet data collections using the [IAT]" is "that substantial majorities of Americans associate young (more than old) with positive characteristics." Id. at 13.
After outlining his credentials, Dr. Greenwald discloses that he based his "opinions rendered in this case on the results of [his] own research as well as on [his] knowledge of published works of many others who have conducted research relevant to the conditions of this case." Id. at 5. Dr. Greenwald also has "become acquainted with the conditions of this case by reading the Complaint statement prepared by Plaintiffs' attorneys" and has "read the depositions of Kevin Cooney in full, as well as a collection of excerpts from the depositions of James Wiggins, Gary Cannon, Diana Jarden, Paul McFarland, Sarah Leider, Tom Casey, Craig Barnette, John Wysseier, Mark Bulger, Anthony Canil, Myrtle Smith, David V. King, Ed Dunn, Peter Dishart, and Gary Eilers." Id. at 5-6. In sum, Dr. Greenwald opines "that there are a number of research findings regarding implicit biases that bear on this case." Id. at 6.
Notably, many of these individuals had nothing to do with the employment situations of the remaining Plaintiffs at PGW's facility in Harmarville, Pennsylvania at which they worked. As this Court has previously detailed at length, Paul McFarland was the direct supervisor of Opt-in Plaintiff Stephen Shaw who worked as a Marketing Manager for LYNX from PGW's headquarters in Pittsburgh; Sarah Leider was the HR Director at the Evansville, IN facility where Opt-in Plaintiff Matthew Clawson worked; Tom Casey was the Manager of Advanced Production at a Satellite manufacturing facility located near PGW's Crestline, OH plant; Craig Barnette was the Plant Manager at the Creighton, PA facility; John Wysseier served as the Director of Service Solutions for LYNX and was responsible for Shaw's termination along with Gary Eilers, the General Manager of ARS & Services with a supervisory role over the entire LYNX business; and David King was the Director of Enterprise Excellence and Quality in Evansville, IN.
Dr. Greenwald's report then "summarizes [his] opinions based on research-established findings that bear on the case." Id. Those opinions include the following:
¶ 13. Implicit biases are pervasive, often observed in more than 70% of Americans, most of whom regard themselves as unprejudiced . . .Id. at 6-8. Dr. Greenwald continues: "[t]he next five paragraphs (¶¶ 17-21) describe prominent recent articles by economists, organizational psychologists, and legal scholars, based on their consideration of research on the roles of subjectivity and discretion in personnel decision making" and "[t]he seven paragraphs after those (¶¶ 22-28) summarize the articles' conclusions in language intended to be more accessible to non-scientists." Id.
¶ 14. In contrast to the small percentages of survey respondents who express self-reported ("explicit") bias based on age, approximately 80% of all research participants hold implicit (sometimes called "unconscious") bias based on age . . .
¶ 15. Implicit bias is scientifically established as a source of discriminatory behavior in employment . . .
¶ 16. Discretion-affording personnel evaluations that permit subjectivity in decision making are open to influence by implicit bias . . .
Following his review of the literature, Dr. Greenwald attempts to apply his research to the facts of this case. Paragraph twenty-nine states:
Plaintiffs' counsel asked me to examine the deposition of Kevin Cooney, and to review portions of several other depositions in which deponents were queried about procedures used in termination decisions affecting Plaintiffs. My objective in examining this material was to form an opinion as to whether the procedures could qualify as applying objective criteria in making the decisions or, alternately, whether the procedures that were used permitted substantial subjectivity in termination decisions. Most remarkable in the deposition material I examined were a collection of absences: First, absence of existing performance appraisals that could be used to inform termination decisions[footnote]; second, absence of systematic procedures for obtaining new data that might provide justifiable bases for those decisions; third, absence of procedures for reviewing decisions before they became final, and if any, they were arbitrary and not systematic; and fourth, absence of procedures for monitoring whether termination decisions had adverse impacts on protected classes of employees.[footnote]Id. at 15-16. The footnotes further qualify his conclusion(s): first, Dr. Greenwald asserts, without explanation, that "[a]lthough there was some conflicting deposition testimony on whether performance appraisals were available to be used in the decision-making process, [he] believe[s] the evidence established that there was an absence of recent performance appraisals and that most managers did not even consult whether performance reviews were available;" and, second, he notes that "the conclusions of [his] present report draw on Dr. [Michael] Campion's expert appraisal of the deposition material he examined." Id. at n.7.
In its entirety, Dr. Greenwald's latter footnote states as follows:
Plaintiffs' counsel made available to me Dr. Michael Campion's expert evaluation of the Human Resources (HR) practices employed by Defendant in the reduction in force decisions that could potentially affect Plaintiffs. I read in entirety the main section of Dr. Campion's report, titled "Review of Human Resources (HR) Practices". Dr. Campion was able to give considerably greater attention than I did to the available deposition material. His opinion also was based on greater expertise than I have in appraising the details of management practices described by the deponents. Accordingly, the conclusions of my present report draw on Dr. Campion's expert appraisal of the deposition material he examined.Greenwald Exp. Rep. at 16 n.7, ECF No. 380-5. PGW characterizes Dr. Greenwald's use of Dr. Campion's analysis as a proverbial game of telephone between plaintiffs' purported experts, sufficiently flawed on its own to bar his testimony. See Def.'s Br. at 6, ECF No. 380 ("Dr. Campion never made any findings about discrimination, did not offer any opinions on causation, and did not have complete data. Like [Dr.] Greenwald, he did not even know the Plaintiffs' names. So, for [Dr.] Greenwald to rely on [Dr.] Campion is like adding 0+0 and then proclaiming the total is greater than zero.").
PGW now seeks to exclude Dr. Greenwald as an expert, arguing that his opinions "should be barred because they lack any relation to the facts of this case and because his methodology is unreliable." Def.'s Br. at 2, ECF No. 380. In addition, PGW contends that Dr. Greenwald lacks the requisite qualifications to offer any opinion about a reduction in force ("RIF") and that his proffered testimony regarding corporate culture and generalized stereotyping in the workplace is inadmissible under Federal Rule of Evidence Rule 404.
For their part, Plaintiffs maintain that PGW's criticism of Dr. Greenwald rings hollow. In doing so, Plaintiffs dispute each of PGW's positions and contend that Dr. Greenwald's testimony about implicit bias is relevant to both their disparate impact and disparate treatment claims by providing the jury with the "framework" referenced above. Pls.' Br. in Opp. at 4, ECF No. 395. Plaintiffs also submit that "[c]ourts throughout the country have routinely admitted similar expert testimony regarding implicit bias and stereotyping in intentional discrimination cases." Id. at 5. This issue is not, however, as straightforward as Plaintiffs suggest.
Plaintiffs cite several cases for this proposition: Price Waterhouse v. Hopkins, 490 U.S. 228 (1989); Peterson v. Seagate U.S. LLC, 809 F. Supp. 2d 996 (D. Minn. 2011); Tuli v. Brigham & Women's Hosp., Inc., 592 F. Supp. 2d 208 (D. Mass. 2009); Hnot v. Willis Grp. Holdings Ltd., No. 01 CIV 6558 GEL, 2007 WL 1599154 (S.D.N.Y. June 1, 2007); Int'l Healthcare Exch., Inc. v. Global Healthcare Exch., LLC, 470 F. Supp. 2d 345 (S.D.N.Y. 2007); Beck v. Boeing Co., No. C00-0301P, 2004 WL 5495670 (W.D. Wash. May 14, 2004); Butler v. Home Depot, Inc., 984 F. Supp. 1257 (N.D. Cal. 1997); Flavel v. Svedala Indus., Inc., 875 F. Supp. 550 (E.D. Wis. 1994); Robinson v. Jacksonville Shipyards, Inc., 760 F. Supp. 1486 (M.D. Fla. 1991). Of course, the only case binding on this Court is Price Waterhouse, which "held that a woman who was denied a promotion because she failed to conform to gender stereotypes had a claim cognizable under Title VII as she was discriminated against 'because of sex.'" Prowel v. Wise Bus. Forms, Inc., 579 F.3d 285, 290 (3d Cir. 2009).
This Court is not the first to decide a motion to bar Dr. Greenwald as an expert witness. For example, in Jones v. Nat'l Council of Young Men's Christian Associations of United States of Am., a United States Magistrate Judge reviewed a motion to strike the report and testimony of Dr. Greenwald in a Title VII disparate impact suit, challenging the defendants' practices for performance evaluations, compensation, and promotion/job assignments of African-American employees. 2013 WL 7046374 (N.D. Ill. Sept. 5, 2013). There, the plaintiffs offered Dr. Greenwald as an expert to "explain the scientific principles demonstrating a phenomenon many do not understand: people who are not overtly racist (and even many African-Americans) subconsciously consider race in decision-making to the detriment of African-Americans, particularly when subjective criteria are involved." Id. at *4. The court began by assessing whether Dr. Greenwald had met the requirements of Rule 702; it concluded that he had not. Id. In doing so, the court highlighted that Dr. Greenwald "did not visit the Y's offices, speak with a current or former employee of the Y, or read any of the employee declarations," that "[h]e did not have IAT data for any Y manager," that he d[id] not even recall seeing the Y's equal employment opportunity policy or its diversity policy," that" [h]e did not critique any Y policy," and that "[h]e has no opinion as to what sort of performance evaluation criteria are used at the Y." In fact, "only one paragraph in [Dr. Greenwald's] report discusse[d] the Y, and that paragraph came from his review of Plaintiffs' statistical expert . . . ." Id. To the court, "Dr. Greenwald ha[d] done nothing that would reliably equip an expert to say something meaningful about employment practices at the Y." Id. at *9. Accordingly, "[i]n light of the fact that Dr. Greenwald did not consider the facts of th[e] case or give a scientific basis to apply his general theories based on the IAT testing to the decisions managers make in a workplace setting, th[e] [c]ourt recommend[ed] that the report and testimony of Dr. Greenwald be stricken . . . ." Id.
The district court adopted the recommendation of the magistrate judge, but it also wrote separately on this issue. Jones, 34 F. Supp. 3d at 896. As an initial matter, the district court rejected the plaintiffs' suggestion that they offered Dr. Greenwald for the limited purpose of educating the jury and found that they sought to rely on his opinion and testimony as evidence of causation—an endeavor which was rejected because his "six-page report [fell] far short of providing a reliable basis to support an opinion that implicit bias of the Y's managers caused any disparity in performance evaluations, pay, or promotions at the Y." Id. at 899. Yet "[e]ven at the level of general principles," the district court also "[wa]s not persuaded that Dr. Greenwald's testimony and opinions are adequately tied to the facts of this case to be useful to a jury" because those general principles did not "fit" the case. Id. at 900. Instead, his opinions were "derived solely from laboratory testing that d[id] not remotely approximate the conditions that apply in th[e] case specifically or more generally in the context of an employer's decisions about employee compensation and work assignments." Id. And "[n]either Dr. Greenwald nor the plaintiffs establish[ed] a logical connection between the principle that hidden bias may be manifested in the absence of any other information and the premise that hidden bias says anything about the results of employment decisions made by supervisors and managers who are armed with abundant data and are personally invested in the results of the process." Id. Accordingly, the district court overruled the objections to the magistrate judge's recommendation. Id. at 901.
At least one state court has also discussed the opinion(s) of Dr. Greenwald, in a disparate impact case brought under Title VII of the Civil Rights Act of 1964, as well as the Iowa Civil Rights Act of 1965. See Pippen v. State, No. LACL107038, 2012 WL 1388902 (Iowa Dist. Ct. April 17, 2012). In Pippen, the trial court considered and rejected Dr. Greenwald's opinion regarding implicit social bias following a non-jury trial. See id. ("Even more significant is the fact that neither [Dr. Greenwald] nor Dr. Kaiser offered a reliable opinion as to how many, or what percentage, of the discretionary subjective employment decisions made by managers or supervisors in the State employment system were the result of 'stereotyped thinking' adverse to the protected class. The closest Dr. Greenwald came to such an opinion was extrapolating data from an internet based site relating to the IAT."). Moreover, the Pippen Court specifically found that the implicit bias evidence offered in that case could not prove causation. See id. ("Both social scientists seem to operate from the assumption that every three out of four subjective discretionary employment decisions made in the State's hiring process were the result of, or tainted by, an unconscious state of mind adverse to African-Americans. The Supreme Court has noted this is a fatal flaw in the proof of a social scientist in a case of this nature and is 'worlds away from "significant proof'"' that an employer 'operated under a general policy of discrimination.' In legal parlance, this is an opinion of conjecture, not proof of causation.") (quoting Wal-Mart Stores, Inc. v. Dukes, -- U.S. --, 131 S. Ct. 2541, 2554 (2011)).
One district court has, however, reached a contrary conclusion. See Samaha v. Washington State Dep't of Transp., No. CV-10-175-RMP, 2012 WL 11091843 (E.D. Wash. Jan. 3, 2012). In Samaha, the district court denied the defendants' motion to exclude Dr. Greenwald from offering expert testimony about implicit bias, rejecting their challenges to the reliability and fit prongs in an employment discrimination case in which the plaintiff (an individual of Arab descent) alleged a disparate treatment claim. 2012 WL 11091843, at *4. The district court first disagreed that "the [IAT] on which Dr. Greenwald bases his testimony amount[ed] to mere 'statistical generalizations about segments of the population,'" by reiterating that the IAT has been validated, peer-reviewed, and subjected to repeated empirical testing, and therefore, it was "satisfied that Dr. Greenwald's opinions [were] sufficiently 'ground[ed] in the methods and procedures of science.'" Id. at *3 (quoting Daubert, 509 U.S. at 597); see also supra n.1.
The Samaha Court also disagreed that Dr. Greenwald's testimony would be neither relevant nor helpful to the jury, finding that the Advisory Committee Notes to the 2000 Amendments to Rule 702 contemplated an expert such as Dr. Greenwald—i.e., one who "conclude[d] that his 'research findings regarding implicit bias . . . bear on this case' even though he d[id] not provide a conclusion as to whether his findings are consistent with the alleged actions of [the] [d]efendants." Id. at *4. Similarly, the district court found that Dr. Greenwald satisfied each factor of the "four-step test to determine the admissibility of expert testimony that does not apply the principles and methods to the facts of the case," set forth in The Advisory Committee Notes for the 2000 Amendments to Rule 702: "that: (1) the expert be qualified; (2) the testimony address a subject matter on which the factfinder can be assisted by an expert; (3) the testimony be reliable; and (4) the testimony 'fit' the facts of the case." Id.
The passage of the Advisory Committee Notes for the 2000 Amendments to Rule 702 cited by the Samaha Court is as follows:
Yet it might also be important in some cases for an expert to educate the factfinder about general principles, without ever attempting to apply these principles to the specific facts of the case. For example, experts might instruct the factfinder on the principles of thermodynamics, or bloodclotting, or on how financial markets respond to corporate reports, without ever knowing about or trying to tie their testimony into the facts of the case. The amendment does not alter the venerable practice of using expert testimony to educate the factfinder on general principles.2012 WL 11091843, at *4.
Unlike the Samaha Court, the undersigned cannot conclude that Dr. Greenwald satisfies the requirements of Rule 702. Dr. Greenwald's opinion is not based on sufficient facts or data. It is not the product of reliable methods. And it would not assist the factfinder in resolving an issue in this case. To be sure, the Court views Jones as instructive.
As in Jones, Dr. Greenwald did not visit PGW at its Harmarville plant or speak with any current or former employee, interview the managers who took part in the March 2009 RIF, or subject any of those individuals to his self-invented IAT—hardly sufficient facts and data on which to base an opinion. Nor did Dr. Greenwald perform any independent, objective analysis on whether implicit biases played any role in the decisions to terminate the remaining Plaintiffs. Instead, Dr. Greenwald recites his credentials, reviews the literature, and attempts to highlight flaws in the employment practices of PGW (its so-called "collection of absences") which he gathered after reviewing one deposition in full and excerpts of others—all of which were selected and supplied to him by counsel for Plaintiffs. See Dep. of Greenwald at 9, ECF No. 413-1; see also E.E.O.C. v. Bloomberg L.P., No. 07 CIV. 8383 LAP, 2010 WL 3466370, at *14 (S.D.N.Y. Aug. 31, 2010) ("Relying solely on the information fed to him by the EEOC without independently verifying whether the information is representative undermines the reliability of his analysis."); Rowe Entm't, Inc. v. William Morris Agency, Inc., No. 98 CIV, 8272 (RPP), 2003 WL 22124991, at *3 (S.D.N.Y. Sept. 15, 2003) ("[A]ny expert should be aware that a party and counsel in a litigation have an interest in the outcome and that an expert study should not be dependent on the information they supply."). Worse yet, Dr. Greenwald filtered his analysis through the lenses of Dr. Campion's purported "expert appraisal of the deposition material he examined." This sort of superficial analysis of the data underlying his opinion is not expert material; it is the say-so of an academic who assumes that his general conclusions from the IAT would also apply to PGW. Simply put, the data underlying his opinion is unreliable and cannot withstand scrutiny in this Court's function as a gatekeeper.
As Dr. Greenwald testified: "Q: Did you try and do anything in this case to determine whether or not any of the decisions made at PGW were actually based on unrecognized mental association? A. Yeah. As I explained earlier, I was not asked to do that. I did not try to do it." Def.'s Br., Ex. 1 at 27, ECF No. 380-3.
The Court also finds that Dr. Greenwald's methodology is unreliable, to the extent that the IAT informed his analysis and provided a basis for his opinion that most people experience implicit bias. Although taken more than fourteen million times, Dr. Greenwald cannot establish that his publicly available test was taken by a representative sample of the population—let alone any person or the relevant decision-maker(s) at PGW. Dr. Greenwald also fails to show that the data is not skewed by those who self-select to participate, without any controls in place to, for example, exclude multiple retakes or account for any external factors on the test-taker. Perhaps to compensate for these shortcomings, Dr. Greenwald explains that his test is widely-used by "[m]any social cognition experts as a method in their own research" and that "[t]here exists near unanimous agreement among social psychologists as to the validity of the IAT as a method for implicit measurement of attitudes and stereotypes." Be that as it may, the IAT still says nothing about those who work(ed) at PGW.
Additionally, the Court finds that Dr. Greenwald's opinion does not "fit." Although Plaintiffs submit that "Dr. Greenwald does not claim his opinions prove causation," Pls.' Br. in Opp. at 6, ECF No. 395, his own report calls this suggestion into question, in which he states that "[t]hese findings [regarding implicit bias] provide a framework that can aid a judge or jury in evaluating the facts of this case . . . . to determine whether the Plaintiffs' ages substantially motivated the defendants' [sic] actions outlined in the Complaint." Greenwald Exp. Rep. at 2, ECF No. 380-5. This is the sort of "substantial disconnect" between the abstract principle from which his general principle is derived and the facts of this case, which was fatal to his opinion in Jones. Even so, it is the wrong standard; a disparate impact claim under the ADEA "requires a showing that 'age was the "but-for" cause of the employer's adverse action,'" while Title VII claims can be maintained with the lesser showing that an improper consideration was a motivating factor for the employer's action." Gladden v. Vilsack, 483 F. App'x 664, 665 (3d Cir. 2012) (quoting Gross v. FBL Financial Servs., Inc., 557 U.S. 167 (2009)); c.f. Smith v. CH2M Hill, Inc., 521 F. App'x 773, 774-75 (11th Cir. 2013) (per curiam) (in alleging that the plaintiff's termination was "substantially motivated" by age, the plaintiff had not "allege[d] sufficient facts" to allow the court to "reasonably infer that [his employers] violated the but-for standard set forth in the ADEA"). If anything, Dr. Greenwald's opinion is more likely to confuse a jury rather than elucidate the issue(s) for the factfinder. Dr. Greenwald may be renowned in his field, "but if he cannot explain how his conclusions satisfy Rule 702's requirements, then he is not entitled to give expert testimony." E.E.O.C., 2010 WL 3466370, at *15.
One final point bears mentioning: the Court doubts that Dr. Greenwald's testimony regarding implicit bias is even relevant in deciding ADEA disparate impact or disparate treatment claims, which are analytically distinct from each other. See generally Jensen v. Solvay Chemicals, Inc., 625 F.3d 641, 660 (10th Cir. 2010) ("In general, disparate treatment occurs when an 'employer simply treats some people less favorably than others' because of a certain characteristic, such as race or age; disparate impact, on the other hand, 'involve[s] 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.'") (quoting Int'l Bhd. of Teamsters v. United States, 431 U.S. 324, 335 n.15 (1977)). Where, as here, a plaintiff asserts a disparate treatment claim, he or she must "prove that intentional discrimination occurred at th[e] particular [employer], not just that gender stereotyping or intentional discrimination is prevalent in the world." E.E.O.C. v. Wal-Mart Stores, Inc., No. CIVA 6:01-CV-339-KKC, 2010 WL 583681, at *4 (E.D. Ky. Feb. 16, 2010). Moreover, disparate treatment claims require proof of a discriminatory motive, which seems incompatible with a theory in which bias may play an unconscious role in decision-making. In a disparate impact claim, evidence of implicit bias makes even less sense, particularly because a plaintiff need not show motive. See generally Rivera-Andreu v. Pall Life Sciences PR, LLC, No. CIV. 14-1029 MEL, 2014 WL 5488409, at *3 (D.P.R. Oct. 29, 2014) ("'The linchpin of a disparate treatment claim is proof of the employer's discriminatory motive. Not so in a claim of disparate impact: that type of claim is predicated not on proof of intentional discrimination, but, rather, on proof that the employer utilizes employment practices that are facially neutral in their treatment of different groups but . . . in fact fall more harshly on one group than another and cannot be justified by business necessity.'") (quoting Mullin v. Raytheon Co., 164 F.3d 696, 699-700 (1st Cir. 1999)). Accordingly, the Court finds that Dr. Greenwald's opinion does not meet the requirements of Rule 702, and therefore, it will bar his testimony at the trial of this action.
B. Dr. Michael Campion
Dr. Michael Campion, Ph.D. holds a doctoral degree in Industrial and Organizational Psychology, a "discipline that conducts most of the scientific research on employee staffing and other Human Resources ("HR") topics." Campion Revised Statistical Analysis at 2, ECF No. 381-4. Dr. Campion presently serves as the Herman C. Krannert Chaired Professor of Management at Purdue University, where he has been a faculty member since 1986. In addition, Dr. Campion "consults regularly to industry and government on staffing and related HR projects," conducting over eight-hundred projects for more than one-hundred-and-ten private and public sector clients, with half of those projects focusing on topics related to employee staffing and even more involving statistical analyses. Id. at 2-3. Before joining the Purdue faculty, Dr. Campion worked for four years at IBM Corporation and another four years at Weyerhaeuser Company in HR analyst and management positions "researching, developing, statistically analyzing, and administering a wide range of personnel systems especially hiring and staffing systems." Id.
Throughout his career, Dr. Campion "has published over 120 articles in scientific and professional journals and given over 200 presentations at professional meetings," half of which "have been on employee staffing-related topics such as testing, interviewing, validation, job analysis, fairness, adverse impact and turnover." Id. Among his many other academic achievements, Dr. Campion served for six years (1990-1996) as editor of the journal Personnel Psychology, a scientific publication that is the primary outlet for research on employee staffing. Dr. Campion also served as the President of the Society for Industrial and Organizational Psychology ("SIOP") from 1995-1996, which conferred upon him its most prestigious lifetime contribution award in 2010. According to Dr. Campion, approximately only one-percent of Industrial/Organizational Psychologists "have the honor of editing an 'A-Class' journal or serving as president of SOIP, let alone both honors." Id. at 3.
Dr. Campion has been retained by Plaintiffs' counsel to provide: (1) a (revised) statistical analysis "to examine whether there is any evidence of adverse impact by age in the reduction in force ('RIF') conducted at [PWG] on March 31, 2009," ECF No. 381-4 at 2; and (2) a report detailing human resource practices during a RIF, the purpose of which was "to review the RIF procedures followed, the relevant documents in the case [citation omitted] and the deposition testimony of the [HR] Managers and line managers who made the termination decisions to evaluate whether the procedures conformed to reasonable HR principles and procedures for conducting RIFs based on the research and practice literature in HR," ECF No. 382-3 at 3.
Dr. Greenwald conducted a "preliminary" statistical analysis on December 2, 2011 and submitted a revised report on May 7, 2013 "based on an updated dataset provided by PGW." Campion Revised Statistical Analysis at 2, ECF No. 381-4.
PGW now seeks to bar Dr. Campion's statistical analyses as well as his opinion on reasonable human resources practices. The Court will address the expert challenges seriatim.
a. Statistical Analysis
The motion to bar Dr. Campion's statistical analysis is complex to say the least. By way of background, the parties undertook fact discovery following the grant of conditional certification, during which they sought business records to establish what employees were selected for inclusion in the March 2009 RIF. At the time of the RIF, PGW distributed with its severance documents a decisional unit matrix ("DUM") (pursuant to the Older Workers Benefit Protection Act, 29 U.S.C. § 623, et seq.) that listed employees' job titles and birth date. In order to match employees to the DUM, the parties sought and received personnel records that were in control of non-party PPG, which continued to provide PGW with payroll system services under a then-existing transition agreement. The payroll data did, however, demonstrate errors in the DUM: certain individuals listed in the DUM were not in fact terminated; others were not included in the RIF but were subject to PGW's earlier decision to close a plant in Evart, Michigan (the "Evart Terminees"). Ultimately, the parties developed an agreed-upon dataset for the March 2009 RIF. The parties did not conduct discovery (other than incidentally) regarding other RIFS, retirements, or separations at PGW.
The errors in and data missing from the DUM form(ed) the basis for Plaintiffs' attempt(s) to invalidate the Separation Agreements and Releases.
PGW contends that the parties agree on two facts: (1) that the Evart plaint closing was not part of the March 2009 RIF; and (2) all but six Evart employees were terminated before March 31, 2009. Moreover, PGW asserts that "even though the payroll records and the DUM showed virtually the same number of employee 862 or 863 - 13 entries on the DUM could not be matched with payroll records" and that "[t]he Agreed Data Set reflects those uncontested facts and is the proper basis for any statistical analysis." Def.'s Br. at 3, ECF No. 381.
In his report, Dr. Campion conducted six analyses, some of which purport to use the Agreed Data Set (Analysis 1) while others go beyond the March 2009 RIF (Analyses 2-6). As Dr. Campion explains:
Although the employees terminated as part of the RIF were identified by PGW in the spreadsheet, several additional analyses were conducted to fully understand the data. Analyses were conducted with and without the December 2008 RIF, and analyses were conducted including retirements and other terminations with severance at the time of the RIF. Published research on employee turnover has shown that retirement is often coded as the reason for termination in HR databases, but employees will independently report different reasons. For example, previous research by the author found that the HR database in a large organization reported that nearly 22% of turnover was retirement, while employees themselves independently reported that only 13% of the turnover was retirement. Employees reported such reasons as dissatisfaction with working condition or supervision, poor health, or lack of promotion much more often than the "official" HR records (Campion, 1991). In the present context, it is possible that "retirement" was recorded, rather than "terminated," due to the RIF in some cases where the employee could qualify for retirement benefits in order to (intentionally or unintentionally) reduce the number of employees counted as terminations. The unusual spike in the number of retirements near the date of the RIF (March 31, 2009) seems coincidental without such an explanation.Campion Revised Statistical Analysis at 7-8, ECF No. 381-4. Dr. Campion also analyzed the potential of disparate impact on specific subgroups within the protected class—comparisons based on five-year increments (e.g., under forty versus forty and over, under forty-five versus forty-five and over, etc.) beginning at age forty and ending at fifty-five—because (as he explained) "[a]ge is a continuous variable." Id. And to determine whether there was in fact a disparate impact on older employees, Dr. Campion cited "[t]wo indices" that "have emerged from governmental guidelines and court cases on equal employment opportunity: Adverse Impact Ratio [(the 80% or four-fifths rule from the EEOC Guidelines)] and Standard Deviation Difference [(the z-score approach)]." Id. at 4; see generally Stagi v. Nat'l R.R. Passenger Corp., 391 F. App'x 133, 137-141 (3d Cir. 2010) (discussing various tests of statistical significance). Against that backdrop, the Court turns to Dr. Campion's six analyses.
The parties discussed Dr. Campion's reasoning for conducting multiple analyses at his June 25, 2013 deposition, during which the following exchange occurred:
Q. Okay. Your Analysis 1 analyzes the RIF on March 31[ ], 2009, right? I'm asking what does your Analysis 2, which has a much larger population, tell us about the March 31[ ], 2009 RIF.Dep. of Campion at 182-83, ECF No. 381-5.
A. Well, it - it would be relevant in the following way: That one could imagine thinking about this RIF as really a RIF that occurred over a six-month time period that began shortly after the acquisition of the company and includes both December and March. And if you conceive of the RIF as being this broader six-month thing, and I understand that you have come to some agreement that it's only the March RIF, but if you wanted to look at this broader context, that analysis would reflect that broader context within which the March RIF occurred.
In his report, Dr. Campion set forth his interpretation of the "80 percent rule" from the Equal Employment Opportunity Commission's ("EEOC") Uniform Guidelines on Employee Selection Procedures and the Federal Contract Compliance Manual, calling the former "[a] primary source of guidance on the analysis of adverse impact." Campion Revised Statistical Analysis at 4, ECF No. 381-4. According to Dr. Campion, for negative personnel action, "if the termination rate of the comparison group is less than 80% of the termination rate of the protected group, then adverse impact is presumed to have occurred thus establishing a prima facie case of discrimination." Id. at 5. As for the z-score approach—which indexes the difference in termination rates in terms of the number of standard deviations—Dr. Campion explains that "[w]hen the number of standard deviations is less than -2 (actually - 1.96), there is a 95% probability that the difference in termination rates of the subgroups is not due to chance alone" and "[w]hen the number of standard deviations exceeds about -3 (actually -2.58), there is a 99% probability that the difference in termination rates of the subgroups is not due to chance alone." Id. at 6. For his analyses, "the values are reversed so negative values indicate adverse impact against the older group." Id.
Analysis 1 "analyzes the base dataset produced by PGW; [t]hat is, it analyzes the adverse impact based on the terminations (n = 100) and the total number of employees (n = 876) as specified by PGW." Campion Revised Statistical Analysis at 8, ECF No. 381-4. In this Analysis, Dr. Campion does not examine the unit in which Plaintiffs worked—the Manufacturing Technology Group headed by Gary Cannon—but instead groups together all one-hundred terminated employees in his analysis. C.f. Karlo v. Pittsburgh Glass Works, LLC, No. 2:10-CV-1283, 2014 WL 1317595, at *3 (W.D. Pa. Mar. 31, 2014) ("Although this decision [to conduct the March 2009 RIF] was made by Wiggins and upper management, the employees were again selected for termination on a decentralized basis."). Be that as it may, Dr. Campion apparently starts with correct number of terminated employees when examining the RIF as a whole, adjusting downward from the 105 identified in the DUM in accordance with the Agreed Data Set. Dr. Campion did not, however, adjust the non-terminated employee data to exclude the Evart Terminees from the RIF. Analysis 1 reveals that "[a]ll adverse impact ratios are below .80," that the standard deviations exceed 2.0 (ranging from 2.15 to 2.46) at the 45, 50 and 55 levels," and that adverse impact ratio at the "40 level" is .69, with the standard deviation "fall[ing] just short of the 2.0 at 1.51, which is significant at the 13% level (two-tailed test)." Id. at 8-9. According to Dr. Campion, "[t]hese results suggest that there is evidence of disparate impact." Id. at 9.
PGW also faults Dr. Campion for ignoring the Agreed Data Set by using "the ghost '876' population" from the DUM—which apparently included thirteen "unmatched" entries—rather than the 863 employees identified on the verified payroll records. At his deposition, Dr. Campion conceded that he did not look at the payroll records maintained by PPG to determine or verify the population set, but instead relied upon the data provided by PGW, presumably in the DUM. Dep. of Campion at 308-310, ECF No. 381-5 ("A: The only data that I [Dr. Campion] have analyzed is that document we described earlier that was produced to us by defendants. And we presumed if you guys produced it, it was accurate. So, I did not go back and review it and try to verify it again.").
Unlike Analysis 1, Dr. Campion considers data fragments from before and after the March 2009 RIF in Analyses 2-6. Analysis 2 "includes the December 2008 RIF, which increases the number of terminations to 173 (additional 73) and the total number of employees to 1,677 (additional 801)"—"[t]he total number of employees [wa]s radically increased because the actual comparators for the December RIF are not known, so all potential comparators [we]re included." Id. at 9. Analysis 3 "adds the December RIFs like Analysis 2, but also adds retirements (n = 16) and other apparent terminations due to the RIF (i.e., terminations with severance n = 7)." Id. Analysis 4 "adds retirements and other apparent terminations due to the RIF to Analysis 1 for the period one month pre and one month post the March 2009 RIF, [to] include[ ] February to April (n = 14). The larger total is again used because the potential comparators are unknown (n = 1,594)." Id. Analysis 5 "adds retirements and other apparent terminations due to the RIF to Analysis 1 from January to September (n = 18) because other terminations in the base dataset counted terminations in that time period as RIF related." Id. at 10. And "Analysis 6 adds all retirements up to September in 2009 (n = 16) to Analysis 1." Id. As with Analysis 1, Dr. Campion concludes that the results of Analyses 2-6 "suggest that there is evidence of adverse impact based on employee age in the RIF conducted at PGW." Id.
In its motion, PGW asserts that Dr. Campion's analyses contain several errors. To summarize, PGW contends that Dr. Campion, as a non-statistician, is not qualified to opine in the field of statistics, that Analysis 1 uses incorrect data, the wrong methodology and fails to specifically examine Plaintiffs, that Analyses 2-6 strays outside the March 2009 RIF to examine an ad hoc collecting of numbers predating and post-dating the RIF, that Dr. Campion engaged in data manipulation, and that he improperly draws his own inferences from the record and offers legal conclusions, usurping the role of the factfinder.
Plaintiffs disagree with PGW's criticisms and suggest that the Court should "avoid intervening in the battle of the experts" between Dr. Campion and Dr. James L. Rosenberger who has been retained by the defendant. In support, Plaintiffs argue that "Dr. Campion's factual assumptions regarding the composition of the data set" and "[t]he proper statistical means to conduct [a] disparate impact analysis" are matters for the jury to decide. Pls.' Br. in Opp. at 3-4, ECF No. 393. Plaintiffs also cite to Brand v. Comcast Corp., Inc., 302 F.R.D. 201 (N.D. Ill. 2014), a decision in which a district court rejected a challenge to Dr. Campion's qualifications. See id. at 209 ("Although Campion is not a statistician—he is a professor of management—he has a sufficient background in statistics to testify on the subjects covered in his report.").
Dr. Rosenberger holds a doctoral degree in Biometry (i.e., "statistics with applications in biological fields"), which he earned from Cornell University in 1977. See Rosenberger Report at 2, ECF No. 381-6. Dr. Rosenberger is a Professor of Statistics at The Pennsylvania State University, holds the departmental positions of Director of the Statistical Consulting Center and Director of Outreach and Online Programs, served as the Department Head from 1991 to 2006 and Director of the Bioinformatics Consulting Center from 2002 to 2007, worked at the National Science Foundation as Statistics Program Director, and has held visiting academic positions at University of Leeds (UK), Stanford University, Swiss Federal Institute (Zurich), and Harvard University. Id. Among his many academic achievements, Dr. Rosenberger has been elected Fellow of the American Statistical Association and Fellow of the American Association for the Advancement of Science, received the Distinguished Service Award from the National Institute of Statistical Science and currently serves as vice president of the American Statistical Association. Id. In addition, Dr. Rosenberger has authored more than fifty refereed scientific publications in statistics related topics. Id. As part of this case, Dr. Rosenberger has authored an expert report in which he concludes that "based on the statistical evidence, it is [his] opinion as a statistician applying generally accepted principles in [his] field that there is no evidence of an adverse impact of the RIF." Id. at 13.
The Court cannot agree that Dr. Campion meets the standards of Rule 702. Even assuming that Dr. Campion is qualified to opine on statistical matters of this sort—of which the Court is not convinced—his report is not based on sufficient data. That is, Dr. Campion includes the Evart Terminees as remaining with PGW after March 31, 2009 and relies on the erroneous DUM rather than the Agreed Data Set. For their part, Plaintiffs attempt to explain away this error: "Dr. Campion did a further analysis and determined that removing the Evart employees from the data set (and thus treating them as having been neither fired or retained in the RIF) had no effect on the outcome of [Dr. Campion's] statistical analysis." Pls.' Br. in Opp. at 4 n.4, ECF No. 393.
Even so, Dr. Campion's methodology is not reliable. To support his conclusion, Dr. Campion relies on the "four-fifths rule," the reliability of which has been criticized. See Delgado v. Ashcroft, No. CIV.A. 99-2311(JR), 2003 WL 24051558, at *8 (D.D.C. May 29, 2003) (collecting cases); see also Stagi, 391 F. App'x at 138 ("The '80 percent rule' or the 'four-fifths rule' has come under substantial criticism, and has not been particularly persuasive, at least as a prerequisite for making out a prima facie disparate impact case.") At the same time, Dr. Campion also offers a z-score approach. See generally, Ogletree v. City of Auburn, 619 F. Supp. 2d 1152, 1177 (M.D. Ala. 2009) ("[O]ne district court concluded that, at least where the sample size is relatively small (189 in that case), 'the 4/5ths Rule is recognized as a rule of thumb to be used in conjunction with standard deviation or other statistical evidence.'") (quoting Jones v. Pepsi-Cola Metro. Bottling Co., 871 F. Supp. 305, 311 (E.D. Mich. 1994)). But that aspect of his analysis is also improper. Using that method, Dr. Campion claims to find evidence of disparate impact in older subgroups despite not finding any statistical significance in the base forty-plus age-group analysis. And in doing so, Dr. Campion does not apply any of the generally accepted statistical procedures (i.e., the Bonferroni procedure) to correct his results for the likelihood of a false indication of significance. This sort of subgrouping "analysis" is data-snooping, plain and simple.
Of course, the subgrouping analysis would only be helpful to the factfinder if this Court held that Plaintiffs could maintain an over-fifty disparate impact claim. It has not done so. The Court instead notes that "[w]hile the Third Circuit does not appear to have considered the issue of subgroup disparate impact claims, [(that is[,] claims based upon evidence suggesting that a particular employment practice affected a subset of individuals within the protected class of those aged 40 or older)], the majority view amongst the circuits that have considered this issue is that a disparate impact analysis must compare employees aged 40 and over with those 39 and younger, and therefore it is improper to distinguish between subgroups within the protected class."). Petruska v. Reckitt Benckiser, LLC, No. CIV.A. 14-03663 CCC, 2015 WL 1421908, at *6 (D.N.J. Mar. 26, 2015) (citations omitted).
Dr. Campion attempts to negate the "data-snooping" label in his rebuttal report. See ECF No. 381-8. That report is rife with inadmissible legal conclusions and relies upon an unreliable source: Wikipedia. See id. To be sure, "district courts prohibit experts from offering legal opinions because such testimony is not helpful to the trier of fact." FedEx Ground Package Sys., Inc. v. Applications Int'l Corp., 695 F. Supp. 2d 216, 221 (W.D. Pa. 2010).
Analyses 2-6 fare no better; they are rooted in rank speculation. In Analysis 2, Dr. Campion includes (inaccurate) data from a separate 2008 RIF on the assumption that the December and March terminations were a single employment decision. See Dep. of Campion at 190-194, ECF No. 190. There is no support in the record for this factual (and fanciful) theory. In Analyses 3-6, Dr. Campion proceeds to add retirees from 2008 and 2009 to Analysis 2 and considers them all as if they were involuntarily terminated. The parties took no discovery on this issue; this opinion evidence is yet again connected to the existing data only by the ipse dixit of Dr. Campion. To be sure, Dr. Campion simply cites his own research and opines that "it is possible that 'retirement' was recorded, rather than 'terminated.'" That sort of conjecture would hardly assist the factfinder in this case. Accordingly, the Court finds that Dr. Campion's statistical report does not meet the requirements of Rule 702, and therefore, it will bar his testimony regarding same.
b. Reasonable Human Resource Practices
For this report, Dr. Campion conducted a review of the "RIF process followed by PGW based on depositions and documents to evaluate whether the procedures conformed to reasonable HR principles and procedures for conducting RIFs based on research and practice literature in HR." Campion HR Practices Report at 6, ECF No. 382-3. In his opinion, "PGW failed to follow reasonable HR practices in many different ways [(no less than nineteen)] when it conducted the RIF terminations." Id. To reach that conclusion, Dr. Campion evaluated "the actual practices at the company" that he gleaned from select deposition transcripts and compared them to a list of "reasonable HR practices" that he previously delineated.
"Reasonable HR practices," as Dr. Campion explains, "are the practices that are recommended by the practice literature, supported by the research literature, or clearly implied by one of these literatures. Also, these are the practices that are taught in the MBA courses by [him]." Id. at 9. According to Dr. Campion, "[t]hese are not 'best practices' because it is a reasonable expectation that they should have been used as opposed to being ideal standards." The "reasonable HR practices" include the following:
(1) identify future work; (1.1) identify jobs and tasks to be performed; (1.2) identify skill, knowledge and experience requirements; (2) evaluate employees against future work; (2.1) consider past job performance heavily; (2.2); consider seniority; (2.3) emphasize job-related criteria (e.g., skills, knowledge, experience, etc.); (2.4) avoid inherently age-related criteria; (2.5) emphasize objective (vs. subjective) criteria; (2.6) base selections on accurate and complete information; (2.7) prevent exposure of demographic information to evaluators; (2.8) allow employees to compete for the remaining or new positions, or positions elsewhere in the organization; (2.9) use multiple independent evaluators to ensure reliability; (2.10) provide clear instruction and train evaluators on use of system, including equal employment considerations; (2.11) ensure adequate documentation; (2.12) allow appeal mechanism; (2.13) apply methodology in consistent manner; (2.14) communicate to employees and provide considerate interpersonal treatment; (3) analyze adverse impact; (3.1) conduct analyses in timely fashion; (3.2) take action if adverse impact is shown; (4) evaluate the process and outcomes; and (5) ensure an informed and independent human resources staff.Id. at 12-52. To gather this list, Dr. Campion explains that he reviewed volumes of publications, yielded eighty-four relevant articles and books on reasonable HR practices during RIFs, and identified what he judged to be a representation of "the most common advice in the professional and scientific HR literature." Id. at 9-11. Dr. Campion later published an article in a peer-reviewed journal summarizing this research, some of which draws from "procedural justice"—i.e., a reference to the "fairness of the procedures used to make decisions." Id. at 11.
Among its many challenges to Dr. Campion's report, PGW argues that his statements are merely ipse dixit conclusions that offer no information regarding the remaining Plaintiffs. PGW also challenges Dr. Campion's qualifications as well as the reliability of the data underlying his opinion(s) and his methodology. In addition, PGW submits that Federal Rule of Evidence 404(b) would also bar Dr. Campion from offering his opinion at trial.
For their part, Plaintiffs argue that "[Dr.] Campion's testimony is relevant to [their] disparate impact claims as it will be helpful to the jury in understanding how PGW's RIF—even if facially neutral—could have a disparate impact on employees fifty and older and how the RIF guidelines that had been the policy of the organization until a few months earlier would have prevented a disparate impact from occurring had they been applied." Pls.' Br. in Opp. at 1, ECF No. 392. Elsewhere, Plaintiffs assert that "Dr. Campion's opinions on reasonable [HR] practices will be relevant to rebut PGW's reasonable factors other than age [("RFOA")] defense." Pls.' Sur-Reply at 2, ECF No. 429.
Plaintiffs submit that "PGW will argue at trial that the decisions it made in selecting who to terminate in the RIF and who to retain were reasonable." Pls.' Sur-Reply at 2, ECF No. 429. From that premise, PGW contends that "[s]ince ordinary jurors are not knowledgeable about reasonable [HR] practices—especially regarding industry-accepted best practices for large-scale reduction in force, they are unlikely to be able to effectively determine whether Plaintiffs' selections were in fact the result of reasonable factors other than age without expert guidance." Id. In its summary judgment brief(s), PGW asserts that its reasonable factor other than age defense relates to its financial situation coupled with the decline in the U.S. economy and its effect on the automotive industry in 2008 and 2009. See Def.'s Br. at 14-18, ECF No. 375.
At least one district court has excluded Dr. Campion from testifying about his reasonable HR practices in an ADEA action related to a RIF and based on a theory of disparate impact. See Powell v. Dallas Morning News L.P., 776 F. Supp. 2d 240, 247 (N.D. Tex. 2011), aff'd, 486 F. App'x 469 (5th Cir. 2012). There, in examining Defendants' RFOA defense, the district court held that Dr. Campion's "20 Reasonable HR Practices that Defendants allegedly should have known about and used during the RIF are irrelevant" because "Plaintiffs c[ould] rebut Defendants' RFOA defense only by demonstrating that the factors offered by Defendants [we]re unreasonable." Id. Moreover, the district court also reasoned that "Defendants have no duty to establish the lack of better alternatives because the RFOA inquiry does not require a determination as to 'whether there are other ways for the employer to achieve its goals that do not result in a disparate impact on a protected class,'" and therefore, "[t]he fact that other RIF practices were available [wa]s not relevant to, and d[id] not rebut, Defendants' RFOA defense." Id. at 266 (quoting Smith v. City of Jackson, Miss., 544 U.S. 228, 243 (2005)).
The United States Court of Appeals for the Seventh Circuit has also stated that a district court "adequately considered and rejected the testimony of Dr. Michael Campion, who stated that the subjective decision-making process used provided the 'perfect atmosphere' for age stereotyping." Richter v. Hook-SupeRx, Inc., 142 F.3d 1024, 1032 n.4 (7th Cir. 1998).
The Court finds this decision instructive and persuasive. In addition, the Court does not consider Dr. Campion's opinion regarding his ideal (and invented) "reasonable HR practices" relevant in assisting the factfinder in understanding or deciding the disparate impact claim in this case: it sheds no light on the relationship between the alleged failures in the RIF and the impact on the remaining Plaintiffs, despite Plaintiffs' inconsistent statement to the contrary. Compare Pls.' Br. in Opp. at 2, ECF No. 392 ("Dr. Campion's opinions have a firm factual grounding and will assist the jury by providing a framework to evaluate whether Plaintiffs' age caused their selection for the RIF.") with Pls.' Sur-Reply at 2 ("As a threshold matter, Plaintiffs argue not that [Dr.] Campion's opinions are relevant to causation, but rather, that they are relevant in rebutting PGW's RFOA defense."). Nor does the Court consider Dr. Campion's opinion particularly reliable; he reviewed deposition excerpts selected by and provided to him by counsel for Plaintiffs and applied those factual snippets to an untested hypothesis. See Dep. of Campion at 92-93, 146-48 ECF No. 382-5; accord supra at 14. As with Dr. Greenwald, Dr. Campion may be well-respected in this particular field, "but if he cannot explain how his conclusions satisfy Rule 702's requirements, then he is not entitled to give expert testimony." E.E.O.C., 2010 WL 3466370, at *15. Accordingly, the Court finds that Dr. Campion's report on "reasonable HR practices" does not meet the requirements of Rule 702, and therefore, his testimony regarding same will be barred.
C. David Duffus (Purported Rebuttal Expert)
David M. Duffus, CPA/ABV/CFF, CFE is a Partner in the Pittsburgh office of ParenteBeard LLC. Prior to this position, Mr. Duffus served as the President of Duffus & Associates (2003-2004), Senior Mangager at Sisterson & Company (2001-2003), Senior Manager, Financial Advisory Services at PricewaterhouseCoopers LLP (1992-1994, 1995-2011), Senior Associate at Linduist, Avey, McDonald, Baskerville (1994-1995), and Corporate Credit Analyst at Meridian Bank (1989-1991). Since 1992, Mr. Duffus "has specialized in working on complex litigation services, forensic accounting and valuation services assignments for businesses ranging from start-up entities to Fortune 100 companies." Duffus Report at 11, ECF No. 391-5. According to Mr. Duffus, "[h]is current responsibilities entail managing the firm's Forensic, Litigation & Valuation Services practice in Pittsburgh, where he oversees assignment planning, supervision of staff, performing analyses, discovery assistance, and expert testimony." Id.
Mr. Duffus has been retained by Plaintiffs to provide his expert opinion in this case and rebut PGW's certified expert in turnaround and restructuring practices, Thomas S. O'Donoghue, Jr. who presently serves as a Principal of Deloitte Financial Advisory Services LLP in Chicago, IL.
The timing of the competing expert reports is a core issue of the present dispute. Importantly, opening expert reports were due on or before May 10, 2013 and rebuttals thereto were due on or before May 30, 2013. See Revised Sch. Order, ECF No. 258; May 10, 2013 Text Order.
Mr. O'Donoghue submitted his expert report on May 10, 2013 in which he analyzes certain aspects of PGW's restructuring activities. See Report of O'Donoghue, ECF No. 391. PGW also presented fact and opinion testimony of Scott Lyons, its Chief Financial Officer and James Wiggins, the Chairman of its Board and Chief Executive Officer, both of whom testified about the conditions of the automotive industry in 2008 and 2009, the circumstances of PGW during that time and the financial performance of the company.
In Mr. O'Donoghue's opinion, PGW management approached the "evaluation of the business, development of turnaround strategies, and implementation of actions in a proactive, intentional manner using the best available information to improve the underlying business. As part of his analysis, Mr. O'Donoghue discusses the market dynamics of the automotive industry as well as the overall decline in the U.S. economy, often referred to as the "Great Recession." --------
Mr. Duffus also submitted an expert report on May 10, 2013, providing "an analysis of the financial damages that ha[d] been claimed by" the then-Plaintiffs. Report of Duffus at 2, ECF No. 391-3. Moreover, as Mr. Duffus stated in his report, "[t]he objective of [his] analysis [wa]s to determine the extent of economic damages accruing to the Class Members as a result of their alleged wrongful termination." Id.
Mr. Duffus submitted two additional expert reports on June 6, 2013. As detailed by Mr. Duffus, the first "report outlines [his] observations and opinions regarding Mr. O'Donoghue's Report," ECF No. 391-4 at 2, while the second report outlines his observations and opinions after having reviewed the expert disclosures of Mr. Lyons and Mr. Wiggins, ECF No. 391-5 at 2. In sum, Mr. Duffus concluded that "(1) Mr. O'Donoghue's analyses, and ultimately his opinions, have an overly narrow scope and ignore key information that is relevant to this matter" and the "opinions set forth in the O'Donoghue Report are based upon incomplete analyses that render his opinions unreliable," ECF No. 391-4 at 15; and (2) that "any expert opinions provided by either Mr. Lyons or Mr. Wiggins suffer from a lack of independence, integrity, and objectivity, thereby rendering the purported expert opinions unreliable," ECF No. 391-5 at 7. Mr. Duffus also concluded "that the disclosures offered with respect to the proposed testimony of Mr. Lyons and Mr. Wiggins are overly broad and lack specificity with respect to the opinions that each holds, and fail to identify in any level of specificity the documents, other information, principals, methodologies and approaches used by each to develop their purported expert opinions." ECF No. 391-5 at 7.
PGW now seeks to bar the purported rebuttal expert opinion of Mr. Duffus due primarily to the late submission by Plaintiffs' counsel and based on its contents being "new, affirmative matter." In the alternative, PGW submits that "at a minimum, [Mr. Duffus' expert reports] should be limited to a rebuttal case[,] and PGW should be given leave to respond." Def.'s Br. at 6, n.5, ECF No. 383; Def.'s Reply at 3, n.1, ECF No. 416. For the reasons that follow, the Court will adopt this latter approach.
PGW's squabbles regarding the untimeliness and affirmative nature of Mr. Duffus' purported rebuttal reports elevates form over substance and does little to advance this five-year-old case. Although the Court certainly does not countenance late submissions it is reluctant to resolve a critical issue based on a procedural technicality rather than the merits of the claims.
Regarding the substantive aspects of PGW's motion—i.e., its attacks on Mr. Duffus' qualifications, credibility, methodology, etc.—the Court finds that they too are without merit. Mr. Duffus is not unqualified as PGW repeatedly suggests based on snippets of his deposition in which he reiterates that he is not a "restructuring expert" or "turnaround guy." Plaintiffs offer Mr. Duffus to rebut PGW's RFOA defense—the 2008-2009 economic downturn and its financial condition—rather than to testify about manner in which the RIF was conducted. Mr. Duffus has recently been qualified to render an expert opinion on similar matters by a member of the United States Court of Appeals for the Third Circuit, presiding over a trial in the Western District of Pennsylvania. See U.S. Fire Ins. Co. v. Kelman Bottles LLC, No. 11CV0891, 2014 WL 3890355, at *5 (W.D. Pa. Aug. 8, 2014). As Judge Fisher stated:
To the extent that Belack and Duffus will testify about Kelman's financial condition during the time period leading up to March 15, 2011 based upon financial records and other documentation that they have reviewed, that testimony will be permitted. Such testimony may include the opinion that Kelman was in financial distress during that time period. The Court notes that Belack and Duffus' backgrounds as certified public accountants qualifies them to testify about Kelman's financial condition, particularly where their testimony is based upon financial records that they have reviewed and incorporated in their expert reports. Such testimony is therefore sufficiently based in fact to meet the standard set forth in Rule 702.Id. The Court finds no legitimate reason to stray from that ruling. Additionally, the Court is not persuaded by PGW's repeated suggestion that Mr. Duffus' qualifications are to be measured solely against those of Mr. Donoghue. In essence, PGW asks this Court to pronounce that less qualified means unqualified. The Court declines to do so.
The Court also cannot agree with PGW's attacks on the reliability of Mr. Duffus' opinion or his methodology. PGW cherry-picks aspects of Mr. Duffus' rebuttal of O'Donoghue with which it disagrees—i.e., his discussion of the Platinum Equity offer, the strategies of private equity firms, and PGW's post-RIF financials—and claims that Mr. Duffus' entire report is unreliable. Once again, disagreement between experts about the scope of a financial analysis is not a proper basis to grant a Daubert motion. To the extent that Mr. Duffus opines on matters that this Court has already ruled to be irrelevant, PGW may submit a motion in limine to limit his testimony at the appropriate time. See, e.g., Karlo v. Pittsburgh Glass Works, LLC, No. 2:10-CV-1283, 2014 WL 1317595, at *21 (W.D. Pa. Mar. 31, 2014) ("Plaintiffs may not, however, include irrelevant averments regarding KKR or Kohlberg," such as "'19. Kohlberg is a Chicago-based private equity firm with $8.0 billion in assets. It was founded by Jerome Kohlberg, Jr., the investment banker who is the son of one of the founders of KKR (Kohlberg, Kravis and Roberts), the leveraged buy-out firm notorious for its controversial practices involving downsizing of companies acquired by it in the junk-bond era. Those traditions established by KKR continued at Kohlberg.'") (quoting Pls.' Proposed Sec. Am. Compl., ECF No. 299-1 at 4, ¶¶ 19, 20).
Finally, the remainder of the motion—its Rule 404(b) challenges to Mr. Duffus' commentary on corporate culture and his apparent personal attacks on Mr. Wiggins and Mr. Lyons—are also more appropriate for pre-trial motions in limine by PGW. At this time, however, the Court will specifically limit the testimony of Mr. Duffus regarding Mr. O'Donoghue, Mr. Wiggins and Mr. Lyons to rebuttal and grant PGW leave to respond. Plaintiffs untimely submitted their rebuttal reports, which deprived PGW of the opportunity to reply. The Court will issue a scheduling order after it rules on the pending motions for summary judgment, assuming they are not granted. Accordingly, the motion to bar the rebuttal expert opinion of Mr. Duffus will be granted in part and denied in part.
IV. Conclusion
For the reasons hereinabove stated, the Court will grant PGW's motions to bar the testimony of Dr. Greenwald and Dr. Campion; and grant in part and deny in part its motion to bar the rebuttal expert reports of Mr. Duffus. An appropriate Order follows.
McVerry, S.J. ORDER OF COURT
AND NOW, this 13th day of July, 2015, in accordance with the foregoing Memorandum Opinion, it is hereby ORDERED, ADJUDGED and DECREED that as follows:
(1) DEFENDANT'S RENEWED MOTION TO BAR PROPOSED EXPERT OPINION OF ANTHONY G. GREENWALD RELATED TO PURPORTED IMPLICIT SOCIAL BIAS (ECF No. 380) is GRANTED;
(2) DEFENDANT'S RENEWED POST-DECERTIFICATION MOTION TO BAR DR. MICHAEL CAMPION'S STATISTICAL ANALYSIS (ECF No. 381) is GRANTED;
(3) DEFENDANT'S RENEWED MOTION TO BAR DR. CAMPION'S EXPERT OPINION ON REASONABLE HUMAN RESOURCE PRACTICES (ECF No. 382) is GRANTED; and
(4) DEFENDANT'S MOTION TO BAR PURPORTED REBUTTAL EXPERT OPINION OF DAVID DUFFUS (ECF No. 383) is GRANTED IN PART AND DENIED IN PART, and the Court will limit the testimony of Mr. Duffus regarding Mr. O'Donoghue, Mr. Wiggins and Mr. Lyons to rebuttal and permit PGW leave to respond, with a scheduling order to issue after the Court rules on the pending motions for summary judgment.
BY THE COURT:
s/Terrence F. McVerry
Senior United States District Judge
cc: All counsel of record.