Summary
analyzing "the Second Circuit's treatment of statistical evidence in disparate-treatment claims in employment discrimination cases" and noting that "[s]tatistics alone are insufficient [in this type of claim] because an individual plaintiff must prove that he or she in particular has been discriminated against."
Summary of this case from Jarvis v. Cambpell & Dawes, Ltd.Opinion
Case No. 94-CV-5944 (FB) (RML).
July 21, 2005
SAM MADUEGBUNA, ESQ., Madu, Edozie Madu, P.C., New York, New York, Attorney for the Plaintiff.
IRA ROSENSTEIN, ESQ., RENA B. PHILLIPS, ESQ., Orrick, Herrington Sutcliffe LLP, New York, New York, Attorney for the Defendant.
MEMORANDUM ORDER
This litigation arose out of three drug tests between April and October 1993, the last of which led to the termination of plaintiff Richard Drake ("Drake") as a flight attendant for defendant Delta Air Lines ("Delta"). Drake contends that he was not randomly selected for these drug tests but was intentionally targeted because he was a union activist. On October 5, 2004, the Court commenced trial to determine whether the drug tests conducted by Delta constituted a non-random search in violation of the Fourth Amendment. At the conclusion of the liability phase of a bifurcated trial, Delta moved for a judgment as a matter of law ("JMOL") pursuant to Fed.R.Civ.Pro. 50(a) ("Rule 50(a)"), arguing that Drake had not established that he was not randomly selected for the drug tests; the Court reserved judgment. See Tr. at 408, 421. Thereafter, a jury returned a verdict in favor of Drake in the sum of $800,000 in compensatory damages and $1.7 million in punitive damages. Delta timely renewed its motion for a JMOL pursuant to Fed.R.Civ.Pro. 50(b) ("Rule 50(b)"), and in the alternative, moved for a new trial in regard to both liability and damages pursuant to Fed.R.Civ.Pro. 59 ("Rule 59"). For the reasons set forth below, the Court grants Delta's motions, negating the $2.5 million verdict.
As a pro se plaintiff, Drake initially filed a Complaint on December 28, 1994, alleging that the drug tests conducted by Delta violated the Fourth Amendment and Federal Aviation Administration regulations, and that Drake's subsequent termination constituted wrongful termination and violated his procedural due process rights; in two separate opinions, this Court dismissed all of the claims. See Drake v. Delta Airlines, Inc., 923 F. Supp. 387 (E.D.N.Y. 1996) (" Drake I"); Drake v. Delta Air Lines, Inc., 1997 WL 397498 (E.D.N.Y. July 10, 1997) (" Drake II"). The Second Circuit affirmed the dismissal in all respects except one — Drake's Fourth Amendment claim; it explained:
Warrantless drug urinalysis testing of employees in safety-sensitive jobs may be consonant with the Fourth Amendment where part of a systematic, uniformly applied testing program (such as random testing). . . . However, Drake's complaint can be construed to allege that Delta's collection and first test of his urine sample were not administered within the parameters of its systematic testing program. . . .See Drake v. Delta Air Lines, Inc., 147 F.3d 169, 171-72 (2d Cir. 1998). Thereafter, this Court assigned counsel to represent Drake in regard to his Fourth Amendment claim.
Tr. refers to the trial transcript.
DELTA'S RULE 50(B) MOTION
A. Rule 50(b) StandardPursuant to both Rules 50(a) and 50(b), a court may set aside a verdict where "there is no legally sufficient evidentiary basis for a reasonable jury to find for that party on the issue[.]" Fed.R.Civ.Pro. 50(a); see also Raspente v. National R.R. Passenger Corp., 111 F.3d 239, 241 n. 3 (2d Cir. 1997) ("[T]he same standard for granting a Rule 50 motion applies under either subsection [50(a) or 50(b)].").
A district court may set aside a verdict only where (1) there is such a complete absence of evidence supporting the verdict that the jury's findings could only have been the result of sheer surmise and conjecture, or (2) there is such an overwhelming amount of evidence in favor of the movant that reasonable and fair minded [jurors] could not arrive at a verdict against him.DiBella v. Hopkins, 403 F.3d 102, 116 (2d Cir. 2005) (internal quotations and citations omitted; alterations in original). "When considering a post-verdict motion for judgment as a matter of law, the trial court should view all the evidence in the record as a whole and draw all reasonable inferences in favor of the nonmoving party, [but] it may not make credibility determinations or weigh the evidence." Id. (internal quotations and citations omitted; alterations in original).
A Rule 50(b) motion "is limited to those grounds that were specifically raised in the prior [Rule 50(a) motion]." Tolbert v. Queens College, 242 F.3d 58, 70 (2d Cir. 2001). Pursuant to this specificity requirement, the Rule 50(a) motion "must at least identify the specific element that the defendant contends is insufficiently supported." Galdieri-Ambrosini v. National Realty Dev. Corp., 136 F.3d 276, 286 (2d Cir. 1998). The purpose of the specificity requirement is "so that the responding party may seek to correct any overlooked deficiencies in the proof." Id. (quoting Fed.R.Civ.P. 50 Advisory Committee Note (1991)).
B. Facts Adduced at Trial
Viewing the evidence in the light most favorable to Drake, as the Court must in evaluating a Rule 50(b) motion, the following was established during the liability portion of the trial.
1. Drake's Employment at Pan Am and Delta
For thirty years, until 1991, Drake served as a flight attendant for Pan American World Airways ("Pan Am"). See Tr. at 253. In October 1977, while Drake was employed as a flight purser at Pan Am, Drake spearheaded a movement to create a separate union that would better represent the interests of flight attendants than the previous union that had represented several other groups in addition to the flight attendants. See Tr. at 64, 97-98, 259-60. Ultimately, while employed by Pan Am, Drake became the chairperson for the New York base of the union and a strike marshal. See Tr. at 65-66. As chairperson, he represented other flight attendants in grievance proceedings and met with management on a regular basis, see Tr. at 65; as strike marshal, he coordinated two strikes, one in October 1978 and the other in March 1985, which essentially shut down Pan Am's operations for a few days. See Tr. at 66-67.
A flight purser is the lead flight attendant, who oversees the work of the other attendants aboard the aircraft, while performing most of the same duties. See Tr. at 96-97.
In 1991, Delta acquired a subset of Pan Am's European operations, and as part of the acquisition, Delta hired some of Pan Am's employees, including Drake and two other flight attendants, Nancy Jean Hite Speck ("Speck") and Ann Blumensaadt ("Blumensaadt"). While at Pan Am, Blumensaadt and Speck each served at different times as the vice chairperson for the New York base of the union. During the interview process, applicants had to complete an employment application, which included a place to disclose any union membership. See Tr. at 72, 104.
Unlike Pan Am, there was no union that represented the interests of flight attendants at Delta. See Tr. at 261. Drake, therefore, discussed with other flight attendants at Delta the possibility of organizing a union but never directly organized a union or spoke with Delta management about creating one. See Tr. at 263-64; 266-67.
In April 1993, Delta's management hosted a meeting for flight attendants to discuss their grievances. See Tr. at 270. At the meeting, Drake vocalized his concern about a rumor circulating involving a junior flight attendant who was planning to intentionally fail other flight attendants in their language-qualification exams. See Tr. at 270. Vicki Escarra ("Escarra"), who was Director of Flight Services and the leader of the April meeting, summarily dismissed Drake's concern, noting that it was merely a rumor. See Tr. at 270. Shortly thereafter, Drake clarified that it was not just a rumor, but Escarra responded that she was "not going to be bullied into doing anything." Tr. at 271. Escarra then whispered something to another member of Delta's management, Judy Panzarella ("Panzarella"), who had been a flight attendant with Drake at Pan Am, which prompted Panzarella to directly look at Drake. See Tr. at 271.
Federal Aviation Administration ("FAA") regulations require Delta to conduct random drug testing of its approximately 39,000 safety-sensitive employees, which include flight attendants; accordingly, each quarter Delta selects approximately 2,500 of its safety-sensitive employees for random drug testing. See Tr. at 112, 117, 176, 189. In the second quarter of 1993, a week after the April 1993 meeting, Drake was notified that he had been selected for random drug testing. See Tr. at 272-73. Thereafter, he was notified that he had been selected for drug testing in the subsequent two quarters (in July 1993 for the third quarter, and in October 1993 for the fourth quarter). See Tr. at 277. Drake submitted a urine sample in each quarter as required. See Tr. at 278.
2. Delta's Attitude Towards Unionization
During Drake's employment at Delta, Delta instituted a 7.5 per cent pay cut of flight attendants' salaries but could not effect a similar pay cut of the pilots' salaries because the pilots were part of a union, see Tr. at 90; in a speech in October 1993 during which the Senior Vice President ("SVP") thanked the flight attendants for this "give back," the SVP made "derogatory comments" about the pilots because they were able to refuse the "give back" by virtue of their unionization. See Tr. at 280-81. Also in October 1993, Drake had a conversation with David Henson ("Henson"), the head of a program for which Drake was training, during which Henson said, "I understand you were the chairman of the union at Pan Am"; when Drake shared his viewpoint that he believed that unions were mutually beneficial, Henson claimed that they were "antiproductive." See Tr. at 279-80. In 1995, after Drake's termination, Delta's management distributed numerous written publications elaborating on its opposition to unionization in response to an effort by flight attendants to organize a union. See Tr. at 74-75.
The record states only that the SVP made "derogatory comments about the pilot group"; it does not specify whether the SVP expressly mentioned the pilots' ability to refuse the "give back" because of their unionization. See Tr. at 281.
3. Delta's Drug-Testing Program
Diane Tennihan ("Tennihan"), who served as the manager of the drug-testing program at the time of Drake's drug tests, testified in regard to the implementation of the program. See Tr. at 112. At the end of each quarter, Delta used a computer program to select the employees who would be subject to drug testing for the upcoming quarter (i.e., the list would usually be compiled for the second quarter at the end of March), see Tr. at 176, 190; however, because the lists were not dated, the precise date that the lists were created could not be determined. See Tr. at 202. To compile these lists, Delta employed a software program developed by MIT to randomly generate numbers; Delta then created software that would identify the employee numbers that corresponded to the numbers generated. See Tr. at 117-18.
Tennihan admitted that two different individuals asked her to add an employee to the quarterly "randomly" generated lists of employees who would be subject to drug testing; in both instances, she explained to them that it was neither technically feasible nor permissible under the FAA regulations. See Tr. at 142-43. She believed that "their concern was possibly [based on] some nature of suspicion [of drug use]" and that they did not understand " random testing [wa]s not the venue for that[.]" Tr. at 148 (emphasis added).
Tennihan testified that she did not recall when these requests were made. See Tr. at 143.
In regard to the feasibility of intentionally adding someone to the list of employees subject to "random" drug testing, Tennihan testified that "there [wa]s no physical way that this could happen" because "that list [wa]s locked[,]" Tr. at 144; see also Tr. at 201; however, although she "d[id]n't think that [it] would be possible[,]" she acknowledged that she "d[id]n't know every internal detail. . . ." Tr. at 214. Dr. Charles Mann ("Dr. Mann"), Drake's expert, partially contradicted Tennihan's testimony that the selection system could not be manipulated when he testified that theoretically a random number generator could be manipulated in order to continuously generate numbers until one matched Drake's employee number. See Tr. at 393.
Tennihan also testified that she was never asked to add Drake's name to the lists, nor had she ever spoken about Drake with either Panzarella or Escarra, see Tr. at 200-01; however, she admitted that she had no means of ensuring that no such request was ever made to one of the employees who worked for her. See Tr. at 147.
4. Statistical Evidence
Dr. Mann, a statistician, testified on the probability that Drake would have been selected in three consecutive quarters if the selections were truly random. He based his testimony on two assumptions: 1) Drake was a union activist, and 2) Drake was the only union activist known by Delta. See Tr. at 333.
Dr. Mann testified that two other Delta employees were selected for drug testing in each of the three quarters that Drake was selected. See Tr. at 333. The following colloquy between Dr. Mann and the Court then occurred:
Dr. Mann: [I]f you just looked at the sampling to be from among 38,000 individuals, and I used 38,000 as a conservative but round number, what's the chance that an individual would be selected three times?
The Court notes that Dr. Mann's estimate of the number of Delta employees eligible for drug testing (i.e., 38,000 employees) slightly differed from that in Tennihan's testimony (i.e., 39,000 employees). The difference is attributable to simple rounding differences; Tennihan rounded up, whereas Dr. Mann rounded down. See Dr. Mann's Expert Report at 1 ("I consider the actual number of eligible employees to be 38,000. This is conservative for the plaintiff compared with the defendant's maximum of 38,510.").
The Court: Any individual selected three times?
Dr. Mann: That's correct. And that number is .000789, which is less than one chance in 12,600.
The Court: That actually has happened here?
Dr. Mann: By definition, will happen. That applies to anybody.
The Court: That actually did happen here. There were two other people selected three times. So, is that consistent with your statistical analysis?
* * *
Dr. Mann: [I]t's easier if I say it the way it says [in my expert report]. The probability that the union activist, the one single person, would be among the three that were selected . . ., if it was made at random from among 38,000 people, was .000789, less than one chance in 12,600.
The Court: It would be less than one chance in 12,600 —
Dr. Mann: That's correct.
The Court: — to pick the union activist under that assumption that you made three times?
Dr. Mann: That's correct.
The Court: The chance of having people randomly selected three times, just from the general population, not union activists, what would that probability be?
Dr. Mann: That was the probability of, among three selected, one would be the activist.
Tr. at 345-46. Although Dr. Mann's testimony was evasive of the Court's questioning, neither the Court nor counsel further pursued this matter; therefore, Dr. Mann never clearly answered whether the one-in-12,600 statistic applied to any individual or whether it was limited to Drake, as the sole union activist known by Delta.
On cross-examination, Dr. Mann conceded that his assessment would change if Delta knew of more than one union activist. See Tr. at 365. He, however, subsequently explained that for Drake's selection to be sufficiently unusual (or less than one chance in twenty that the selection was random), there only needed to be less than 644 known union activists; thus, his analysis would not materially change unless Delta demonstrated that it knew of more than 644 union activists. See Tr. at 389-90.
In regard to the assumption that Drake was the only known union activist, Drake introduced into evidence a letter submitted by Delta in discovery that stated that it did not know who among the Pan Am flight attendants that it hired had been union activists. See Tr. at 384. Dr. Mann, however, admitted that he was not aware that, of the employees subject to the drug-testing program, nine thousand were pilots who were members of a union, many of whom were active and held union positions. See Tr. at 399.
C. Dr. Mann's Expert Report
Although not admitted at trial, see Tr. at 343, pursuant to Fed.R.Civ.Pro. 26(a)(2)(B), Drake submitted to Delta and to the Court Dr. Mann's expert report, where he outlined the statistical analysis that he employed.
Relying on the same assumptions that he relied upon in his trial testimony, he stated that "[t]he probability that the sole known `union activist' ( or any other individual) would be selected in any specific quarter of 1993 [wa]s .0625." and that "[t]he probability that the sole known `union activist' (or any other individual) would be selected in any specified three quarters of 1993 using simple random sampling [was].0625 × .0625 × .0625 = .000244 (less than one chance in 4,000)." Dr. Mann's Expert Report at 3 (emphasis added). Therefore, unlike Dr. Mann's testimony at trial, his expert report makes clear that the probability that the sole union activist would be selected in any three specified quarters was the same as that probability for any other individual.
Dr. Mann's expert report, however, differs from his trial testimony, in which he stated that the predicted probability that an individual would be selected in three specified quarters was less than one in 12,600 — not less than one in 4,000. See Tr. at 345 ("[I]f you just looked at the sampling to be from among 38,000 individuals, . . . what's the chance that an individual would be selected three times? . . . And that number is .000789, which is less than one chance in 12,600."). The one-in-12,600 statistic was the probability that the union activist would be among the three individuals who were actually selected in the three specified quarters. See Dr. Mann's Expert Report at 2.
Furthermore, based on Dr. Mann's calculations, it was expected that between nine and ten of the 38,000 employees eligible for drug-test selection would be selected in each of the three quarters that Drake was selected (i.e., 38,000 × .000244 = 9.3). Indeed, as Dr. Mann commented, "[t]his corresponds to the result . . . obtained by the FAA that . . . says that 2.4 employees out of 10,000 may be expected to be selected in any specific three selections and that 9 or 10 such selections may be expected out of 40,000 employees." Id. Thus, Delta's actual selection of three individuals in the specified three quarters (i.e., April, July and October 1993) was less than the prediction that nine or ten employees would be selected in those three quarters.
In his report, Dr. Mann also "calculated the probability that the `union activist' would be among the three selected . . ., under the assumption that selection of exactly three individuals for drug testing exactly three times each was made at random from among 38,000 individuals of whom exactly one was a `union activist', to be . . . 3/38,000 [or] .0000789." Id. at 2. The statistic, .0000789, was equivalent to "3 (the number of employees tested three times) divided by 38,000 (the number of employees in the pool)." Delta Mem. at 18. Again, Drake's chance of being among the three selected was the same as any of the 38,000 employees' chances of being among the three selected.
D. Analysis
1. Burden of Proof
The Court charged that the jury that Drake must prove by a "preponderance of evidence" that "Delta's selection of Drake for drug testing was not random[.]" Tr. at 509, 511. Neither of the parties objected to the charge. The Court notes, however, that at least one court, the Tenth Circuit, has instead employed a burden-shifting scheme in a claim similar to Drake's; the Tenth Circuit explained that once the plaintiff has "offer[ed] some evidence raising a genuine factual dispute as to randomness of the tests, the burden shifts to the government to establish the randomness." See Northington v. Warden, 113 F.3d 1246, 1997 WL 242255, at *2 (10th Cir. 1997) (unpublished). In any event, because the Court concludes that the evidence adduced at trial was not sufficient to create even a genuine factual dispute regarding the randomness of the drug-test selection process, the Court need not decide whether it was plain error not to give the jury a burden-shifting charge.
Although "[c]itation of an unpublished decision is disfavored[,]" the Tenth Circuit permits citations to its unpublished decisions where, as here, "(1) it has persuasive value with respect to a material issue that has not been addressed in a published opinion; and (2) it would assist the court in its disposition." 10th Cir. R. 36.3. See also Melissa M. Serfass Jessie W. Cranford, Federal and State Court Rules Governing Publication and Citation of Opinions: An Update, 6J.App.Prac. Process 349, 351-58 (2004) (surveying circuit courts' rules for citing to their unpublished opinions). Moreover, the Court notes that the Federal Judiciary's Committee on Rules of Practice and Procedure recently approved the proposed new Appellate Rule 32.1, which would require courts to permit citation to unpublished opinions; the proposed new rules will now be transmitted "to the Judicial Conference at its September 2005 session with recommendations that they be approved and transmitted to the Supreme Court." See The Federal Judiciary, Standing Rules Committee Approves Proposed Amendments and New Rules, http://www.uscourts.gov/rules/index.html (last visited Jul. 8, 2005).
A plaintiff's burden of demonstrating a genuine factual dispute under Northington resembles a plaintiff's burden of demonstrating a prima facie case in employment-discrimination cases. See McDonnell Douglas Corp. v. Green, 411 U.S. 792 (1973).
Drake contends that the statistical evidence combined with other circumstantial evidence adduced at trial is sufficient to sustain this burden. The Court rejects this contention because neither the statistical evidence nor any other circumstantial evidence is probative of his claim.
2. The Statistical Evidence
The Court's research has not revealed any case addressing the use of statistical evidence in regard to a Fourth Amendment claim. Statistical evidence, however, is commonly used as evidence of discriminatory intent in employment-discrimination cases. Using employment-discrimination cases to guide the Court in its analysis, the Court will address 1) the appropriate weight to place on statistical evidence; 2) the type of statistical evidence that is probative; and 3) the probative value of the statistical evidence in this case.
a. Weight Afforded to Statistical Evidence
Drake correctly concedes that statistical evidence alone is not sufficient to establish his claim of a non-random search. See Drake Mem. of Law in Opp. to Delta's 50(b) Mot. ("Drake Mem.") at 24 ("As counsel for Delta rightly points out in Delta's memorandum of law, it is well settled that statistical data alone would not establish a prima facie case."). Drake's concession corresponds with the Second Circuit's treatment of statistical evidence in disparate-treatment claims in employment discrimination cases. Compare Martin v. Citibank, N.A., 762 F.2d 212, 218 (2d Cir. 1985) (citing Hudson v. International Bus. Machs. Corp., 620 F2d 351, 355 (2d Cir. 1980)) ("[S]tatistical proof alone cannot ordinarily establish a prima facie case of disparate treatment."), with Hudson, 620 F2d at 355 ("In disparate impact cases, where plaintiff alleges that an entire class or category of minority workers are adversely affected by an employer's facially neutral policy, and in class actions alleging racially discriminatory practices, statistical evidence may establish a prima face case of discriminatory effect."); see also Bailey v. City of New York, 2003 WL 21031972, at *9 (S.D.N.Y. May 2, 2003) (statistics alone are not sufficient to establish a discriminatory-treatment claim); Miles v. City of New York, 2002 WL 31410346, at *3 (E.D.N.Y. Oct 24, 2002) (same). Statistics alone are insufficient in a disparate-treatment claim because an individual plaintiff must prove that he or she in particular has been discriminated against. See Hudson, 620 F.2d at 355. In a similar vein, statistics alone are not sufficient to establish a genuine factual dispute regarding the randomness of a drug-test selection process with respect to a single individual.
Although the circuit court's use of the term, "ordinarily," evidences that there may be a case where statistics alone could be sufficient in a disparate-treatment case, the Court's research has not disclosed any such case.
Nonetheless, although it alone is not sufficient, statistical evidence, if probative, is admissible as circumstantial evidence of discrimination, and together with other evidence may be sufficient to establish a genuine factual dispute. See Smith v. Xerox Corp., 196 F.3d 358, 370 (2d Cir. 1999) (in context of a disparate-treatment claim, statistical evidence is "circumstantial evidence of intentional discrimination."); Stratton v. Department for the Aging for New York, 132 F.3d 869, 877 (2d Cir. 1997) (statistics "are admissible to support a claim of discrimination even in a disparate treatment case involving a single plaintiff."). The amount of additional evidence required logically inversely correlates with the probative value of the statistical evidence (i.e., the more probative the statistical evidence, the less additional evidence is required).
The Court's research has revealed only one Second Circuit case involving a disparate-treatment claim that addressed whether the plaintiff had established a prima facie case based on a combination of probative statistical and other evidence. See Stratton, 132 F.3d at 877-80. There, the Second Circuit countenanced a jury verdict in favor of a plaintiff, but summarily noted that the statistical evidence "w[as] accompanied by substantial other evidence showing discrimination[,]" and that "the district court [had] cautioned the jury not to place undue reliance on the [statistical evidence][,]" id. at 877; the substantial other evidence included "[a]ctions taken by an employer that disadvantaged the employee for no logical reason[,]" and a stray comment "evidencing a preference for young mothers over older workers. . . ." Id. at 879-80. Because the circuit court did not elaborate, it is unclear whether the jury verdict could have survived had there been less substantial evidence.
b. When Statistics are Probative
Statistics are probative when they measure the extent that an actual result varies from an expected result. See Ottaviani v. State Univ. of New York at New Paltz, 875 F.2d 365, 367 (2d Cir. 1989) (the proper methodology for statistical determination of disparate treatment is to "measure the difference between the predicted treatment and the actual treatment of those employees"). For example, absent discrimination, it is expected that in selecting 50 employees for random drug testing from a group of 100 employees, where half are known union activists and half are not, about 25 workers in each group would be selected for drug testing. Although the actual number would "vary from that expected number by some small amount due to chance[,] [i]f the obtained result varies too greatly from the expected result [courts are] willing to infer that one group has been discriminated against." Smith, 196 F.3d at 366. "[A] study is found significant — and the hypothesis of chance is rejected — when there exists at most a one in 20 possibility that the observed result could have occurred by chance." Segar v. Smith, 738 F.2d 1249, 1283 (D.C. Cir. 1984).
Even then, however, not all statistical evidence is admissible. Statistical evidence must adequately address the relevant statistical comparison and must account for other possible causes of the disparity. For example, in the context of employment discrimination, because "a disparate treatment claim looks at how an individual was treated compared to her similarly situated coworker[,]" statistics are admissible when they "compare coworkers who competed directly against each other to receive a benefit" and employ some form of regression analysis to eliminate other possible bases for the disparity. See Smith, 196 F.3d at 370-71 (emphasis added); see also Bickerstaff v. Vassar College, 196 F.3d 435, 450 (2d Cir. 1999) (affirming district court's decision to exclude expert statistical analysis that "failed to control for nondiscriminatory causes" and thus "rendered it of no probative value").
c. Whether Dr. Mann's testimony Was Probative of Drake's Complaint
As an initial matter, the Court notes that Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), "require[s] the trial court to `ensure that any and all scientific testimony or evidence admitted is not only relevant, but reliable.'" Brooks v. Outboard Marine Corp., 234 F.3d 89, 91 (2d Cir. 2000) (quoting Daubert, 509 U.S. at 589). However, as the Eighth Circuit has appropriately advised:
It is far better where, in the mind of the district court, there exists a close case on relevancy of the expert testimony in light of the plaintiff's testimony to allow the expert opinion and if the court remains unconvinced, allow the jury to pass on the evidence. Depending on the verdict, the trial court can always refer to Federal Rule of Civil Procedure 50(b) and grant a judgment as a matter of law or a new trial.Lauzon v. Senco Products, Inc., 270 F.3d 681, 695-96 (8th Cir. 2001). Accordingly, although prior to trial the Court ruled that Drake could introduce Dr. Mann's expert testimony, see September 3, 2002, Tr. at 22-23 ("It may be a thin case . . . but I do think that if I were to preclude the plaintiff from going forward and striking the expert's testimony that there would be a significant risk of having to try this case in the future. . . ."), the Court now reevaluates whether Dr. Mann's testimony should have been excluded under Daubert. See Goebel v. Denver Rio Grande Western R.R. Co., 215 F.3d 1083, 1087 (10th Cir. 2000) ("The district court may also satisfy its gatekeeper role [under Daubert] when asked to rule . . . on a post-trial motion. . . ." (citing Daubert, 509 U.S. at 597)). In so doing, the Court reviews not only the evidence adduced at trial but also the underlying expert report, which was not admitted at trial.
"September 3, 2002, Tr." refers to the transcript of oral argument held on September 3, 2004.
As previously noted, Dr. Mann testified that two other employees in addition to Drake were tested during the same three quarters, see Tr. at 333, and that "[t]he probability that the union activist, the one single person, would be among the three that were selected . . ., if it was made at random from among 38,000 people, was. 000789, less than one chance in 12,600." See Tr. at 345. There is no doubt that this statistic — less than one chance in 12,600 — is at least superficially compelling, and the jury presumably heavily relied on Dr. Mann's testimony in finding that Drake's drug-test selection was not random.
As noted, Dr. Mann also testified that the one-in-12,600 statistic applied to the predicted probability that an individual would be selected in three specified quarters, see Tr. at 345; his expert report clarified that this predicted probability was actually one in 4,000. See Dr. Mann's Expert Report at 2.
Nonetheless, Dr. Mann's testimony simply does not have any probative value because, as he testified at trial and repeated in his report, his statistical assessment was identical to the expected probability that any one person — not just a known union activist — would be among the three selected in each April, July and October 1993; instead of bolstering an inference of discrimination, it merely demonstrates that Drake's chance of being among the three selected during each of those three quarters equaled the probability that any one person would be among the three selected. Furthermore, the actual result varied only slightly from the predicted one; in fact, the actual result (i.e., three employees were selected in each of the three specified quarters) was less than Dr. Mann's predicted result (i.e., between nine and ten employees would be selected). The Court thus concludes as a matter of law that Dr. Mann's statistical testimony was not probative of Drake's claim.
Because the Court concludes that Dr. Mann's statistical testimony does not bolster Drake's claim, it need not address Delta's secondary argument, whether there was evidence to support Dr. Mann's assumption that Drake was the only union activist known by Delta), upon which assumption he relied. See Delta Mem. at 19.
3. Whether Other Circumstantial Evidence is Sufficient to Justify the Verdict.
Even assuming arguendo that Dr. Mann's statistical testimony had some probative value, Drake has not presented additional circumstantial evidence to create a genuine factual dispute regarding the randomness of Delta's drug-test selection process. In reaching this conclusion, the Court rejects Drake's contentions that there was sufficient evidence to justify the jury's verdict, namely:
1) Delta refused to provide a description of its program to enable an independent expert [to] determine whether or not the program was random as claimed by Delta; 2) susceptibility to manipulation, of a commercial random number generator like Delta's; 3) Delta's intent and propensity to manipulate its random selection process to include names not randomly selected; 4) Delta's motive for wanting to select Drake in a non-random manner for drug testing; and 5) that no identifiable person, other than Drake, was selected three consecutive times for drug testing by Delta. Drake Mem. at 13. Each of the above will be addressed in turn.
a. Delta's Refusal to Provide a Program Description
Drake first contends that because "Delta refused to provide a description of its program to enable an independent expert [to] determine whether or not the program was random as claimed by Delta[,]" "[t]he jury had every right to consider Delta's failure to present clear and convincing evidence that its system was in fact random." Drake Mem. at 12, 15. Had the Court found that Delta had committed a discovery violation, the Court could have considered, if indicated, whether it should have instructed the jury that it may draw a negative inference from the alleged violation. See Fed.R.Civ.Pro. 37 (for discovery violations, court "may impose . . . appropriate sanctions[, which] . . . may include informing the jury of the failure to make the disclosure."); see also Kronisch v. United States, 150 F.3d 112, 126 (2d Cir. 1998) ("It is a well-established and long-standing principle of law that a party's intentional destruction of evidence relevant to proof of an issue at trial can support an inference that the evidence would have been unfavorable to the party responsible for its destruction."). No such discovery violation was reported to the Court. In any event, Tennihan testified extensively, without contradiction, regarding the methods and procedures that Delta used to ensure that the selection process was randomly administered.
b. Delta's Intent to Manipulate and the Susceptibility of the Random Number Generator to Manipulation
Drake next argues that the program used by Delta to select the employees for drug testing was "susceptib[le] to manipulation" and that Delta had the "intent and propensity to manipulate its random selection process to include names not randomly selected[.]" Tennihan, however, denied both of these contentions.
In regard to Drake's claim that the program was susceptible to manipulation, Tennihan, who was directly responsible for administering the program, testified that it could not be manipulated. Dr. Mann's mere speculation that a software program could be developed to manipulate a random number generator does not address whether the program itself — not the random number generator — was susceptible to manipulation.
In regard to Drake's claim that Delta had the intent and propensity to manipulate the program, Tennihan merely testified that two different people asked her to add an individual on the list. At no time did she testify that these individuals asked her to manipulate the system. To the contrary, she explained that the two instances were based on the inquirers' lack of understanding that the program was only used to randomly select employees and that there was a separate mechanism to drug test individuals when there was a reasonable suspicion of drug use.
Moreover, Drake may not rely solely on the supposition that the jury may have disbelieved Tennihan simply because she denied that Delta's program was susceptible to manipulation and that Delta had an intent and propensity to manipulate the program. See Martin, 762 F.2d at 217 ("[W]hile in strict theory a party having the affirmative might succeed in convincing a jury of the truth of his allegations in spite of the fact that all the witnesses denied them, we think it plain that a verdict would nevertheless have to be directed against him."); 9A Charles Alan Wright Arthur R. Miller, Federal Practice and Procedure § 2527 (2d ed. 1995), cited in Martin, 762 F.2d at 218 ("If all of the witnesses deny that an event essential to the plaintiff's case occurred, the plaintiff cannot get to the jury simply because the jury might disbelieve these denials. There must be some affirmative evidence that the event occurred.").
c. Delta's Motive
Drake also claims that Delta had a "motive for wanting to select Drake in a non-random manner for drug testing," namely Drake's union activism, which was brought to Delta's attention at the April 1993 grievance meeting — just one week prior to his notification that he was selected for drug testing. Drake is certainly correct that Delta disfavored the formation of unions and outwardly campaigned against them in 1995; furthermore, Drake's selection occurred in temporal proximity to the April meeting, which Drake contends provided Delta with a motive to target Drake. However, this motive alone is not sufficient to demonstrate that Delta manipulated its drug-testing program to intentionally include Drake. See Seebald v. Praxair, Inc., 2004 WL 350912, at *15 (E.D. Pa. Jan. 21, 2004) (no causal connection between plaintiff's filing an age discrimination claim and subsequent selection of plaintiff for a "random" drug test because "all that has been shown is that [the plaintiff's] selection for substance screening having occurred nearly contemporaneous with his employer receiving notice of his [Equal Employment Opportunity Commission] charge was an inauspicious coincidence").
d. Drake Was the Only Identifiable Person Selected Three Times
Drake lastly contends that "no identifiable person, other than Drake, was selected three consecutive times for drug testing by Delta." Drake Mem. at 12. Again, the evidence adduced at trial simply does not support Drake's contention; Drake's own expert, Dr. Mann, testified that two others in addition to Drake were selected in the same three quarters.
In sum, Drake has not presented sufficient evidence — statistical or otherwise — to create a genuine factual dispute regarding the randomness of Delta's drug-test selection process. Given "such a complete absence of evidence supporting the verdict[,] . . . the jury's findings could only have been the result of sheer surmise and conjecture." DiBella, 403 F.3d at 116. The verdict must be overturned.
DELTA'S RULE 59 MOTION
Alternatively, Delta argues that it is entitled to a new trial under Rule 59 in regard to both liability and damages. Rule 50 expressly provides:
If the renewed motion for judgment as a matter of law is granted, the court shall also rule on the motion for a new trial, if any, by determining whether it should be granted if the judgment is thereafter vacated or reversed, and shall specify the grounds for granting or denying the motion for the new trial.
Fed.R.Civ.Pro. 50(c) (emphasis added); see also New England Ins. Co. v. Healthcare Underwriters Mut. Ins. Co., 295 F.3d 232, 248 n. 29 (2d Cir. 2002).
"[A] motion for a new trial ordinarily should not be granted unless the trial court is convinced that the jury has reached a seriously erroneous result or that the verdict is a miscarriage of justice." Munafo v. Metropolitan Transp. Auth., 381 F.3d 99, 105 (2d Cir. 2004). Unlike a judgment as a matter of law, there is no preservation requirement and the Court "may grant a new trial even where there is substantial evidence to support the jury's verdict if the court is convinced that the verdict was manifestly erroneous." Manley v. Ambase Corp., 337 F.3d 237, 246 (2d Cir. 2003) (internal citations and quotations omitted). In considering a motion for a new trial, the Court "is free to weigh the evidence [it]self, and need not view it in the light most favorable to the verdict winner." Id. at 244-45.
For the same reasons stated above, the weight of the evidence clearly supports a verdict in favor of Delta. The Court, thus, need not address Delta's arguments in regard to damages.
CONCLUSION
Delta's motion for a judgment as a matter of law ("JMOL") under Rule 50(b) is granted. In the alternative, Delta's motion for a new trial under Rule 59 is granted. Accordingly, the Complaint is dismissed.
SO ORDERED.