HOWARD v. PHILADELPHIA HOUSING AUTHORITY et al

Filing 61

MEMORANDUM AND/OR OPINION. SIGNED BY MAGISTRATE JUDGE DAVID R. STRAWBRIDGE ON 6/16/16. 6/17/16 ENTERED AND COPIES E-MAILED.(mbh, ) .

IN THE UNITED STATES DISTRICT COURT
FOR THE EASTERN DISTRICT OF PENNSYLVANIA
SHENECQUA BUTT, et al
v.
UNITED BROTHERHOOD OF
CARPENTERS & JOINERS OF
AMERICA, et al
THERESA HOWARD
v.
METROPOLITAN REGIONAL
COUNSEL OF CARPENERTS &
JOINERS OF AMERICA, et al

:
:
:
:
:
:
:

CIVIL ACTION

:
:
:
:
:
:
:

CIVIL ACTION

NO. 09-4285

NO. 13-374 (consolidated with 09-4285)

MEMORANDUM OPINION ON MOTION IN LIMINE
CENTER FOR FORENSIC ECONOIC STUDIES
DAVID R. STRAWBRIDGE
UNITED STATES MAGISTARTE JUDGE
I.

June 16, 2016

Introduction
Defendants Metropolitan Regional Council of Carpenters and Joiners of America,

Edward Coryell, and Mark Durkalec (collectively, the “Union Defendants”) bring a Motion in
Limine seeking to exclude trial testimony from the Center for Forensic Economic Studies
(“CFES”). Doc. 143.
This matter comes before the Court pursuant to 28 U.S.C. § 636(b)(1)(A) and Federal
Rule of Civil Procedure 72(a). In conformity with the scheduling orders governing these cases,
Plaintiffs obtained reports from CFES in July and August 2015. The Union Defendants filed
their Motion in Limine on November 9, 2015. Doc. 143. All Plaintiffs filed Responses in
Oppositions on December 21, 2015, Doc. 153, 154, and the Union Defendants filed Replies on
December 30, 2015. Doc. 160, 161. The Court held oral argument on the Motion on June 7,
1

2016,1 and the matter is now ripe for resolution.

For the reasons that follow, the Union

Defendants’ Motion shall be granted. In light of the Court’s conclusions as to the reliability of
CFES’s statistical analyses in this case, Plaintiffs shall additionally be prohibited from
introducing CFES’s purported expert reports.
II.

Factual Background
CFES provided reports to Plaintiffs on July 28, August 7, and August 12, 2015. Doc.

143-4. Plaintiffs contend that the CFES reports provide a measure of the wages they lost as a
consequence of the Union’s alleged sex discrimination.
The July report begins with a comparison of average hours worked by male and female
members of Local 1073 from 2003 through 2012. According to CFES, a “t-test”2 performed on
the data demonstrates that there is a statistically significant difference in the number of hours

1

The Plaintiffs filed unauthorized submissions in advance of this hearing to address several
questions the Court posed for oral argument. Doc. 191, 192. While the Court has no obligation
to consider these filings, it has reviewed them. Nothing in them changes the analysis outlined
herein.
2

None of the CFES reports define what a “t-test” is or precisely how it was applied in this case.
Counsel for Plaintiff Mitchell submitted a supplemental filing, with the Court’s authorization,
that called the Court’s attention to the Reference Manual on Scientific Evidence, Third Edition
from the Federal Judicial Center. Doc. 194. The Court notes that in defining the phrase “t-test,”
the Manual is explicit in cautioning, “A t-test is not appropriate for small samples drawn from a
population that is not normal.” Doc. 194-1 at 7. The sample size in CFES’s analysis was just 17
women and 388 men on average across ten years. There is no evidence of exactly how many
women or men were employed in any given year or whether or not the sample sizes in this case
were statistically adequate. While Plaintiff Mitchell is correct that CFES purportedly considered
the entire population of carpenters of Local 1073 in its analysis, that is not necessarily
dispositive: there would be no statistical significance in a comparison of pay discrepancies
between one man and one woman, even if those two individuals represented the entire
population of a given union. Even if the Court were to put aside its serious reservations and
accept Plaintiffs’ assertion that a “t-test” was appropriate in this case, the Court would still
conclude that the Union Defendants’ Motion should be granted given other flaws in CFES’s
methodology as outlined herein.
2

worked between men and women that is not attributable to chance. CFES appears to assume that
because the discrepancy is not random, it is the result of intentional gender discrimination.
In all three reports, CFES compared the number of hours each Plaintiff worked to the
average number of hours a male Union member worked in a given year and created a lost wages
damage figure based on the difference. For instances in which a Plaintiff worked more hours
than the average male carpenter, CFES listed damages as zero. The reports note that “hours after
2012 were projected at 1,303 hours per year, the average male hours from 2007 to 2012.” Doc.
143-4 at 10, 12.
III.

Analysis
A. Standards for Expert Testimony
Federal Rule of Evidence 702 governs the admissibility of expert testimony. The rule

“has 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 v. Ford Motor
Co., 520 F.3d 237, 244 (3d Cir. 2008). “[T]he testimony must be reliable; it must be based on
the methods and procedures of science rather than on subjective belief or unsupported
speculation; [and] the expert must have good grounds for his on her belief.” Schneider ex rel.
Estate of Schneider v. Fried, 320 F.3d 396, 404 (3d Cir. 2003) (citations omitted). Determining
if testimony will assist the trial of fact “entails a preliminary assessment of whether the reasoning
or methodology underlying the testimony is scientifically valid and of whether that reasoning or
methodology properly can be applied to the facts in issue.” In re TMI Litig., 193 F.3d 613, 663
(3d Cir. 1999) (citations omitted).

“[E]ven if an expert’s proposed testimony constitutes

scientific knowledge, his or her testimony will be excluded if it is not scientific knowledge for

3

purposes of this case.” In re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 741 (3d Cir. 1994)
(emphasis in original).
Plaintiffs, as proponents of the CFES reports, must prove all elements of admissibility by
a preponderance of the evidence. In re TMI Litig., 193 F.3d at 663. The Union Defendants’
decision to not depose the CFES report authors to examine their methodology does not impact
Plaintiffs’ burden. Because Defendants have not challenged CFES’s qualifications, the Court
proceeds to analyze the reliability of their reports and their “fit” to this case.
B. The CFES Reports Are Inadmissible to Prove Disparate Impact Because
There Are No Disparate Impact Claims At Issue in This Case
At the June 7 hearing, Plaintiffs asserted – for the first time in this litigation3 and in spite
of materials they submitted to the Court shortly before the hearing began4 – that they are seeking
to bring a disparate impact claim alongside their disparate treatment claims. The Court rejects
what amounts to Plaintiffs’ untimely attempt to amend their Complaints to assert such a claim
after discovery has closed. Josey v. John R. Hollingsworth Corp., 996 F.2d 632, 641 (3d Cir.

3

In contrast to what Plaintiff Mitchell argued, paragraphs 20 and 21 of the Amended Complaint
in Docket Number 09-cv-4285 (Doc. 6) do not plead disparate impact. These paragraphs state,
“During Plaintiffs’ employment, males and non-black employees were given assignments before
black female employees and were also given preferential assignments. As a result of the
discriminatory assignment practices, Plaintiffs received substantially less work than other
employees and suffered long periods without any work or income from Defendants.” These
paragraphs, as well as all of the Complaints that have been filed in this litigation (see Doc. 1 in
Docket Number 10-cv-2633; Doc. 1 in Docket Number 10-cv-3269; Doc. 28 in Docket Number
13-cv-374), have never identified the Union’s system of job assignments as a facially neutral
practice causing a disparate impact on women and/or minority workers.
4

The Court notes that in both of the unauthorized Plaintiff submissions, Plaintiffs disclaimed any
reliance on a disparate impact theory. Doc. 192 at 7 (“Although this is a disparate treatment case
. . .”); Doc. 191 at 4 (“To the extent Plaintiff makes mention of cases involving disparate impact
claims, she does so by way of example for reference only. She also points out that the heightened
standards for admission in disparate impact claims do not apply in disparate treatment claims.”).
4

1993). Therefore, the CFES reports may only be consider here to support claims for disparate
treatment.
C. The CFES Reports Are Unreliable Because They Reflect Misunderstandings
About the Record in This Case
The CFES reports repeatedly assert that the Plaintiffs “sustained lost income and benefits
. . . as a result of termination and layoffs from the Carpenters union.” Doc. 143-4 at 3, 6, 10; see
also id. at 13 (“After Ms. Mitchell and Ms. Butt were terminated from the Carpenters in
2009…”). This allegation is unsupported by the record. While Plaintiffs attempt to brush that
language aside as mere sloppy drafting, the Court concludes that this mistake calls into question
the reliability of the CFES reports.

CFES appears to have been misinformed about the

allegations at issue in this litigation.
D. The CFES Reports Are Unreliable Because They Engage in Speculation
The CFES reports purport to assign damages to Plaintiffs Howard and Mitchell for lost
income in 2013, and for Plaintiff Bronson in 2013 and 2014, despite the fact that the CFES
comparison of average hours between male and female carpenters ends with data from 2012.
The reports note, “hours after 2012 were projected at 1,303 hours per year, the average male
hours from 2007 to 2012” despite the fact that in 2012 itself, the average number of male hours
was only 1,153. Doc. 143-4 at 10, 12. In fact, carpenter hours were on a downward trend from
2010, suggesting that using a projected average for 2013 and 2014 is improperly speculative.
See Calhoun v. Yamaha Motor Corp., U.S.A., 350 F.3d 316, 322 (3d Cir. 2003) (affirming
exclusion of “speculative and unreliable” expert evidence).

5

E. The CFES Reports Are Unreliable Because They Assume Causation and
Fails to Control for Individual Circumstances or Other Variables in Work
Assignments
We are guided in part by Judge Huyett’s opinion in Stair v. Lehigh Valley Carpenters
Local Union No. 600 of United Brotherhood of Carpenters and Joiners of America, No. CIV. A.
91-1507, 1993 WL 235491, at *7, *18 (E.D. Pa. July 24, 1993), aff’d, 43 F.3d 1463 (3d Cir.
1994), in which he noted, in a case bearing significant factual similarities, that “[m]any variables
determine the number of hours worked by a carpenter: whether the carpenter solicits
employment, whether he or she uses the referral system, whether an employer asks for that
carpenter by name, whether the carpenter will accept out of town employment, and whether the
carpenter has the skills requested by an employer when that employer calls the Union for a
referral.”5
Much as in Stair, the Union Defendants do not have exclusive control of the hours that
each of their members work. Among the highly individualized variables that go into a Union
member’s ability to obtain work assignments are:


how much time and effort each individual union member spends in soliciting work on
their own pursuant to “open solicitation” or “free solicitation” policies under local
collective bargaining agreements (see Doc. 151-4 at 3-4);



the willingness of an employer to accept a particular member that the Union refers; and

5

At the June 7 hearing, the Court asked the parties to discuss how if at all Stair differs from this
case. Plaintiffs noted that Stair’s conclusion that the proffered statistical evidence was
insufficient to support a claim of disparate treatment occurred in the context of the court issuing
a decision on the merits as factfinder, not in a motion to exclude evidence pursuant to Rule 702.
Indeed, it is not clear from Stair if there was an expert proffered as opposed to just the data itself.
We also recognize that Stair post-dated the Supreme Court’s decision in Daubert v. Merrell Dow
Pharmaceuticals, Inc., 509 U.S. 579 (1993), by approximately one month. Nevertheless, the
Court finds the variables outlined in Stair go directly to the reliability of the CFES reports per
Daubert such that Stair is helpful in resolving the Union Defendants’ Motion.
6



contractors’ requests for specific union members to perform work given that the
contractor has worked with that carpenter in the past.6
The CFES reports note a statistical disparity between the average number of hours

worked by men and women. They do not, however, analyze or address any variables other than
assumed gender discrimination, or otherwise put the Union’s data in context, in concluding that
disparate treatment is to blame. Most glaringly, the CFES reports do not control for what effect,
if any, the Union Defendants’ alleged racial discrimination played in the hours disparity. While
it is certainly true that statistical evidence in individual disparate treatment cases “need not be so
finely tuned” as in class-action or pattern and practice cases, see Abrams v. Lightolier Inc., 50
F.3d 1204, 1217 (3d Cir. 1995), the CFES reports simply have not considered enough
circumstances to conclude that gender discrimination is the cause of, rather than merely
correlated with, the discrepancy in hours observed. See Molthan v. Temple Univ. of Commw.
Sys. of Higher Educ., 778 F.2d 955, 963 (3d Cir. 1985) (“Because the considerations affecting
promotion decisions may differ greatly from one department to another, statistical evidence of a
general underrepresentation of women in the position of full professor adds little to a disparate
treatment claim.”); Riding v. Kaufmann’s Dep’t Store, 220 F. Supp. 2d 442, 459 (W.D. Pa. 2002)
(“Plaintiff’s statistical evidence is mildly interesting, but she does not put the data in context
(how old were the women?) [or] tell us what to do with it or what inferences should be gathered
from it . . .”); Brown v. Cost Co., No. Civ.A. 03-224 ERIE, 2006 WL 544296, at *3 (W.D. Pa.
Mar. 3, 2006) (affirming exclusion of statistical evidence in disparate treatment case). Nor have
the reports sufficiently explained why the appropriate damages metric is a comparison of each
6

It may well be that this kind of process, based largely on past personal relationships, had a
disparate impact on African American women carpenters. As explained earlier, however,
Plaintiffs have never asserted a disparate impact claim in this litigation and the Court is
unwilling to let them do so now at this late stage of the proceedings.
7

Plaintiff’s hours to that of an average man in the Union given how each union member’s hours
fluctuate over time and with individual circumstances.
IV.

Conclusion
Given the above analysis, we preclude testimony from CFES, as we find Plaintiffs have

failed to satisfy their burden of proof as to the admissibility of the reports. Accordingly, we
grant the Union Defendants’ Motion in Limine.
An appropriate Order follows.

BY THE COURT:

/s/ David R. Strawbridge
DAVID R. STRAWBRIDGE
UNITED STATES MAGISTRATE JUDGE

8