OPINION AND ORDER
FEIKENS, District Judge.
Plaintiff Voncile Mims (Mims) worked for defendant EDS for eight years before she was terminated in June 1993, as part of a reduction in force. She is African American and was 42 years old at the time of her discharge. Mims sues on the basis of diversity jurisdiction, alleging age and race discrimination in violation of the Elliott-Larsen Civil Rights Act, M.C.L. 37.2101
et
seq. She puts forward disparate treatment and disparate impact claims as to both age and race. Defendant has moved for summary judgment on all claims. For the reasons stated below, defendant’s motion is stayed pending argument on the admissibility and probative value of plaintiffs statistical evidence.
I.
Standard of Review
Summary judgment may be granted “if the pleadings, depositions, answers to interrogatories, and admissions on file, together with affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to judgment as a matter of law.” Fed.R.Civ. P. 56(c). The moving party carries the initial burden of making this showing. Once that burden is met the non-moving party must present evidence to show that a genuine issue of material fact exists.
Celotex Corp. v. Catrett,
477 U.S. 317, 324-25, 106 S.Ct. 2548, 2553-54, 91 L.Ed.2d 265 (1986);
Betkerur v. Aultman Hospital Association, 78
F.3d 1079, 1087 (6th Cir.1996). In deciding a summary judgment motion a court must determine whether there is sufficient evidence for a reasonable jury to return a verdict for the non-movant, viewing the facts in the light most favorable to the non-moving party.
Monette v. Electronic Data Systems,
90 F.3d 1173 (6th Cir.1996), citing
Adickes v. S.H. Kress & Co.,
398 U.S. 144, 157, 90 S.Ct. 1598, 1608, 26 L.Ed.2d 142 (1970);
Street v. J.C. Bradford & Co.,
886 F.2d 1472, 1478 (6th Cir.1989).
II.
Analysis
A. Prima Facie Case of Disparate Treatment
Plaintiff claims both age and race discrimination. Under Michigan law, a plaintiff may establish a claim of disparate treatment by showing direct evidence that age or race was a “determining factor” in the decision to discharge plaintiff. Alternatively, plaintiff may employ the disparate treatment framework established under federal law in
McDonnell Douglas Corp. v. Green,
411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973). Thus she may establish a prima facie case of age or race discrimination by showing (1) she was a member of the protected class; (2) she was discharged; (3) she was qualified for the position; and (4) she was replaced by a younger or non-minority person.
Matras v. Amoco Oil Company,
424 Mich. 675, 683, 385 N.W.2d 586 (1986). By proving a prima facie case, a plaintiff creates an inference of dis
crimination, which defendant must rebut with a legitimate nondiscriminatory explanation for its decision.
Mims was clearly a member of protected classes for age and race discrimination. She was discharged. She was qualified for the job she held, as defendant does not contest she was fired for incompetence (though EDS does question her performance relative to other employees). But the final prong of the
McDonnell Douglas
test, replacement by a younger person, does not apply, since plaintiff was discharged as part of an economically-motivated reduction in force.
Michigan case law has recognized the problem posed by discrimination claims in the context of a reduction in force, where by definition the plaintiff was not replaced.
Evidence that a competent older employee was terminated, and a younger employee was retained, is insufficient standing alone to establish a prima facie case when the employer reduces his work force because of economic necessity. The rationale behind the
McDonnell Douglas
formula is that its four-part test alone “eliminates the most likely legitimate causes for the employer’s adverse action.” This formulation is incomplete in the work-foree-reduction situation.
Matras
at 684, 385 N.W.2d 586 (internal citations omitted). In such a situation, the plaintiff must “present sufficient evidence on the ultimate question — whether age was a determining factor in the decision to discharge the older protected employee.”
Id.
To meet this “ultimate burden,” a plaintiff in a work reduction case “must come forward with additional direct, circumstantial, or statistical evidence that age was a determining factor in his job displacement.”
Ridenour v. Lawson Co.,
791 F.2d 52, 57 (6th Cir.1986);
see also Wilson v. Firestone Tire & Rubber Co.,
932 F.2d 510, 517 (6th Cir.1991)(citing
Ridenour
at 57 and
Barnes v. GenCorp Inc.,
896 F.2d 1457, 1469 (6th Cir.1990),
cert den.
498 U.S. 878, 111 S.Ct. 211, 112 L.Ed.2d 171 (1990)).
As analyzed below, I find that Mims has presented evidence from which a jury could reasonably conclude that similarly situated younger and non-minority employees were retained. However, under
Matras, Riden-our,
and
Wilson
this is not sufficient to maintain her prima facie case in the reduction in force context; she must come forward with “additional direct, circumstantial, or statistical evidence” that age or race was a determining factor in her discharge.
Ridenour
at 57.
Plaintiff has submitted statistical evidence showing age- and race-based disparities in defendant’s discharges of its employees which might satisfy this additional burden. I have not yet heard argument on defendant’s motion to strike this evidence because plaintiff allegedly failed to properly authenticate or disclose it. Neither have I heard argument on the methodology or probative value of these statistics. Thus for purposes of summary judgment, I cannot find at this time that plaintiff has failed to present sufficient evidence to establish a prima facie ease of discrimination.
Mims was one of two members of a thirteen-person Asset Management team who were dismissed. This team was part of the division managed by Stephan Lutz, who testified that he had assembled all group supervisors within his division to discuss staff reductions. The other employee to be discharged from Lutz’s division was Sandra Lawler, a
white woman who was over 40 at the time.
Lawler, like Mims, worked on the Asset Management team.
Mims does not argue that she performed better than others in the Asset Management team who were not discharged. Mims’ October 1992 evaluation had been favorable, “Solid Meets Expectations” (SM), but no member of the group had been rated lower. Rather, Mims compares her performance to two younger, white employees who worked in Lutz’s division, though not in the Asset Management team, Glenn Southard and Christopher Clark. Mims alleges that because Southard and Clark had lower performance ratings they should have been discharged before she was.
The central dispute is thus whether Clark and Southard were comparably situated to Mims, such that disparate treatment of them evidences discrimination. EDS urges that employees must be identically situated in order to be compared for purposes of showing discrimination. In support, EDS points to federal court decisions on federal law that have held that the plaintiff “must show that the ‘comparables are similarly-situated in all respects.’ ”
Mitchell v. Toledo Hospital,
964 F.2d 577, 583 (6th Cir., 1992), citing
Stotts v. Memphis Fire Department,
858 F.2d 289 (6th Cir.1988). Thus the comparable employees “must have dealt with the same supervisor, have been subject to the same standards, and have engaged in the same conduct without such differentiating or mitigating circumstances that would distinguish their conduct or the employer’s treatment of them for it.”
Mitchell
at 583. In other words, the plaintiff must show that “all relevant aspects of his employment situation are ‘nearly identical’ ” to the comparable employees.
Pierce v. Commonwealth Life Ins.,
40 F.3d 796, 803 (6th Cir.1994)(internal cite omitted).
Plaintiffs claim is brought solely under Michigan’s Elliott Larsen Civil Rights Act (ELCRA), M.C.L. 37.2101
et seq.
Defendant argues that the ELCRA standard for comparable employees is the same as the “identically situated” federal standard, citing
Larson v. First of America Bank Corp.,
Slip Op.No. 189717 (Mich.Ct.App. February 7, 1997). The
Larson
decision adopts the language in
Mitchell
demanding that comparables be situated “in all respects,” including supervisor, standards, and conduct.
Larson
at pp. 2-3.
Plaintiff maintains that courts interpreting the ELCRA have applied a less stringent “substantially similar” test to comparable employees. In
Sisson v. Bd. of Regents of the University of Michigan,
174 Mich.App. 742, 747, 436 N.W.2d 747 (1989), the court stated that the fact that plaintiff worked in a different department from the employees with whom she sought comparison, and under different supervisors, “will not preclude a finding of dissimilar treatment. The law requires only that the co-workers have similarity, not identity, in place and time.”
Id.
at 747, 436 N.W.2d 747. In another case, the Michigan Court of Appeals permitted an age discrimination plaintiff to compare his treatment to that of employees in a different job classification.
Wolff v. Automobile Club,
194 Mich.App. 6, 12, 486 N.W.2d 75 (1992). The two classifications were comparable because of the “similarity of the job responsibilities,” even though employees in one classification did not recruit new customers as did the others, and one group was paid a salary while the other worked on commission.
Id
The
Sisson
and
Wolff
decisions were issued before the
Mitchell
decision, though Mitchell’s interpretation of federal law could not overrule the earlier state court decisions. However, a Michigan ease decided with the advantage of
Mitchell’s
analysis also supports the application of a less stringent test.
Coleman-Nichols v. Tixon Corp.,
203 Mich. App. 645, 653-4, 513 N.W.2d 441 (1994). There, a plaintiff alleging sex discrimination
was permitted to compare her treatment to that of another assistant manager whose duties were different and whose department had a different function; additionally, the male had a college degree, while the female plaintiff did not.
Id.
at 652, 513 N.W.2d 441.
Defendant argues that the Michigan cases do not contradict the federal test, because in the state cases the differences between plaintiffs and the “comparables” were found to be insignificant, thus preserving an “identically situated” standard. Defendant also cites cases where Michigan courts have found employees not to be similarly situated because differences were found to be significant.
Schultes v. Naylor,
195 Mich.App. 640, 491 N.W.2d 240 (1992)(conduet was not the same);
Michigan Civil Rights Comm. ex rel. Boyd v. Chrysler Corp., Mack Ave. Stamping Plant,
80 Mich.App. 368, 377-78, 263 N.W.2d 376 (1977)(duties were different);
Bouwman v. Chrysler Corp.,
114 Mich.App. 670, 681, 319 N.W.2d 621 (1982)(different skills and experience);
C.Thorrez Industries, Inc. v. Michigan Dep’t. of Civil Rights ex rel. Fullerton,
88 Mich.App. 704, 278 N.W.2d 725 (1979)(different skills and experience).
This argument only illustrates that the difference between the “identically situated” and “similarly situated” test is largely semantic. Even under the supposedly stricter federal test, courts consistently examine whether the situation of plaintiffs and comparable employees is the same in all
relevant
aspects.
Pierce, supra.
Those aspects of employment that are considered relevant must depend on the factual context of each case. This is consistent with cases treating the similarity of the comparables to the plaintiff as a jury question.
Coleman-Nichols,
supra;
Douglass v. Eaton Corp.,
956 F.2d 1339, 1342 & 1343-6 (6th Cir.1992). For purposes of summary judgment, even applying the stricter federal test, the question thus becomes whether a reasonable jury could find that Mims was identically situated to Southard and Clark in all
relevant
aspects of employment.
As a threshold matter, there is evidence that the cuts could have come from anywhere in Lutz’s division, and not just the Asset Management team. It appears that Lutz was told of the need to terminate people, and called together the team managers under his supervision to inform them of this need, including Mims’ supervisor, Yvonne Bonnici, who was the head of the Asset Management team at the time the cuts were made. Bon-nici testified that she had not wanted to make the cuts from her group but did so under protest. Lutz testified to the contrary, that the decision had been a joint one, and that Bonnici had eventually volunteered to do with fewer staff in her group. Drawing inferences in the favor of the non-moving party for purposes of summary judgment, I must assume that Lutz intended the cuts to come from anywhere within his division.
The fact that Mims had a different supervisor from Clark and Southard may well be damaging to her claim, but does not preclude a finding that the employees were similarly situated. While Lutz asked each of the supervisors under his authority to identify employees who could be eliminated, we may assume for purposes of summary judgment that he had ultimate responsibility for deciding staff cuts within his division. In addition to relying on the supervisors’ rankings of the members of their teams, Lutz also directly reviewed plaintiffs performance ratings, as he acknowledges in his affidavit. Thus it appears that Lutz, and not the team supervisors, was arguably the relevant decision maker.
Although EDS claims that Mims had performance problems and customer complaints, Mims points to evidence that Southard and Clark had records that were, if anything, beneath hers. Both men had received “Solid Generally Meets Expectations” (SG) ratings in their most recent review, lower than Mims’ SM rating, though Mims in the past had also been rated SG. EDS argues that this “could” indicate that the two men were “new to the position and [had] not yet reached a Solid Meets level.” Supplemental Brief in Support of Defendant’s Motion for Summary Judgment at 8. An equally plausible inference may be drawn that the two were performing below plaintiffs level.
EDS argues that the differing job duties and skills of Mims and Southard and Clark prevent them from being viewed as compara-
bles. They characterize Mims’ duties as clerical, while Southard was a “highly skilled technician whose job duties were essential,” and Clark was involved with the maintenance of computer hardware and software. However, these statements alone do not show relevant differences in job duties. EDS provides no specifics as to Southard’s actual duties beyond a statement in the deposition testimony of an EDS employee that Southard was “highly skilled” and “a real technical systems engineer.” Deposition of Ron Sipes at 52. Neither does EDS present a comparison between Clark’s and Mims’ computer responsibilities. Defendant’s statements that Clark and Southard were “essential” and “highly skilled” are not sufficient to force summary judgment for defendant on this issue.
As noted above, even though Mims may establish the existence of comparably situated employees who were not discharged, this is insufficient to establish a prima facie case in the reduction in force context. Plaintiff must come forward with “additional direct, circumstantial, or statistical evidence” that age or race was a determining factor in her case. Ridenour at 57. The viability of her prima facie case will depend on the additional statistical evidence she has attempted to present.
B. Legitimate Business Reason and Pretext
Assuming that plaintiff can establish her prima facie case, she must also demonstrate that EDS’ stated legitimate business reason for her discharge is pretextual. I find that she can make such a showing, based on non-statistical evidence she has offered. Of course, if her statistical evidence is inadmissible or inadequate to support her prima facie case, defendant will be entitled to summary judgment irrespective of evidence of legitimate business reason or pretext.
Michigan courts interpreting the ELCRA rely on precedent under analogous federal antidiscrimination law.
Radtke v. Everett,
442 Mich. 368, 361-382, 501 N.W.2d 155 (1993). Under this framework, once a plaintiff has established a prima facie case a presumption of discrimination is created, which the defendant may rebut by producing evidence of a legitimate, nondiscriminatory reason for its actions. The plaintiff, who carries the burden of persuasion throughout, must demonstrate by a preponderance of the evidence that the defendant’s explanation is a pretext for discrimination, either directly, by persuading the fact finder that discrimination more likely motivated the employer, or indirectly, by showing the purported nondiscriminatory reason is not credible.
St. Mary’s Honor Center v. Hicks,
509 U.S. 502, 113 S.Ct. 2742, 125 L.Ed.2d 407 (1993);
Clark v. Uniroyal,
119 Mich.App. 820, 825, 327 N.W.2d 372 (1982);
Lytle v. Malady,
209 Mich.App. 179, 187, 530 N.W.2d 135 (1995),
appeal granted
451 Mich. 920, 550 N.W.2d 535 (1996).
Defendant maintains that Mims was discharged for unsatisfactory performance. Mims was one of thirteen Business Relations Analysts on the Asset Management team, which was charged with maintaining the inventory of corporate assets of EDS’ clients on a computer system. Thus she was responsible for entering the purchase or transfer of equipment on the computer system, and for preparing monthly inventory reports. EDS maintains that Mims failed to perform necessary tasks such as transferring assets on customer accounts, generating numerous customer complaints. In addition, EDS says that the fact that she did not volunteer for special projects or work overtime was a factor in her discharge.
In challenging defendant’s asserted reason, Mims points out that she had received a “SM” (satisfactory meets expectations) rating in her last evaluation, similar to her teammates, and had received a salary increase in March 1993, which would not have happened if her work was unsatisfactory. Defendant maintains that the SM is a “very broad category,” and Mims fell on the lower end, and also that her performance deteriorated between the October 1992 review and her June 1993 discharge. Mims points out that her supervisor, Bonnici, could only recall one incident where there was a problem with a customer account, and she asserts that this problem was due to another person’s failure to submit paperwork rather than Mims’ ne-
gleet. Mims also challenges the credibility of Bonnici’s assertions regarding complaints from another EDS area, maintaining that she had stopped servicing this area before Bonni-ci arrived.
This is sufficient to meet Mims’ burden to show pretext. As an initial matter, there is a material dispute of fact regarding the credibility of defendant’s asserted nondiscriminatory reason for Mims’ discharge. EDS has provided no contemporaneous documentation of performance problems, though Mims asserts that the company’s policy requires it to retain such documentation. The only documentation provided by EDS is a critique of the performance of Mims and Lawler written at the time of the decision to fire them. Thus EDS’ evidence of inadequate performance is primarily the deposition testimony of its own employees, and Mims’ evidence of good performance reviews and a raise during her last year is sufficient for a reasonable trier of fact to find pretext.
In addition, defendant has insisted in court and in its papers that Mims was not fired because she was unfit for her job in absolute terms, but rather because of her performance relative to others in the Asset Management team who performed better. This might suffice as a legitimate reason if the universe for reduction in force was only the thirteen-person team. But, as discussed in relation to the prima facie case, Mims has presented evidence that the staff cuts could have come from anywhere in Lutz’s division. Thus defendant’s admission that Mims was not unfit in absolute terms may cut against its proffered nondiscriminatory reason.
C. Disparate Impact
Plaintiff claims that defendant’s subjective procedures for determining reductions in force had a disproportionate impact on older and African American employees. In order to sustain a disparate impact claim, a plaintiff must show that a facially neutral employment practice caused a negative impact on the protected class of which the plaintiff is a member. Because this showing is entirely-dependant on the admissibility and probative value of Mims’ statistical evidence, I defer ruling on summary judgment at this time.
III.
Conclusion
For the reasons stated above, defendant’s motion for summary judgment shall be stayed pending argument on the admissibility and probative value of plaintiffs statistical evidence.
IT IS SO ORDERED.
Defendant Electronic Data Systems Corporation (EDS) seeks summary judgment on complaints of employment discrimination filed by plaintiff Voncile Mims. Mims, an African American employee, was discharged in June 1993 as part of a reduction in force. Her lawsuit is based on diversity jurisdiction, and she alleges both disparate treatment and disparate impact claims of age and race discrimination under the Elliott-Larsen Civil Rights Act, M.C.L. 37.2101
et seq.
In an earlier opinion, I stayed defendant’s motion pending hearing on the admissibility and probative value of statistical evidence offered in support of Mims’ claims. I held that Mims had presented sufficient evidence to show that comparably situated white and younger employees were retained while she was fired. But because Mims’ discharge took place during a reduction in force, I held that the existence of these “comparables” could not, standing alone, sustain her disparate treatment claim. Thus her disparate treatment claim is dependent on the statistical evidence she seeks to offer in support. Her disparate impact claim is entirely dependent on a showing of disparities affecting older and non-white employees, and therefore I held that this claim also stands or falls on her statistical evidence.
I now hold that plaintiffs proffer of statistical evidence is not probative of either age or race discrimination. Summary judgment on the disparate treatment and disparate impact theories is appropriate. I need not decide defendant’s contention that the evidence was not properly authenticated or noticed under Fed.R.Civ.Pr.56. and Local Rule 7 of this court.
A.Disparate Treatment
My previous opinion held that when an allegedly discriminatory firing takes place as part of a reduction in force, the usual prima facie case of disparate treatment, as set forward in
McDonnell Douglas Corp. v. Green,
411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973), is insufficient to establish an inference of discrimination.
Matras v. Amoco Oil Company,
424 Mich. 675, 684, 385 N.W.2d 586 (1986). Plaintiff must “present sufficient evidence on the ultimate question'— whether age was a determining factor in the decision to discharge the older protected employee.”
Id.
The statistics offered here do not meet that requirement. I found that the relevant employment unit was the EDS division managed by Stephan Lutz, who was the effective decisionmaker in Mims’ discharge. Lutz had some 40 to 50 employees under his supervision. The statistics that Mims offers include some 14,000 members of EDS’ GM North American Operations (NAO) group, headed by Frank Stoneking, who is three levels of management above Lutz. The statistics do not show discriminatory treatment in Mims’ division.
Evidence of disparities in an employment unit larger than the plaintiffs may be probative in cases where there is a single policy governing the larger unit.
McAlester v. United Air Lines,
851 F.2d 1249, 1258 (10th Cir.1988). But in this case there is no evidence of any company-wide policy. There is no evidence that Lutz received direction as to discharges in his department from higher levels of management. The NAO-wide statistical disparities thus are not probative of discriminatory treatment by Lutz or the managers under his supervision.
Several cases have denied discovery of statistical information beyond the relevant employment unit.
Scales v. J.C. Bradford and Co.,
925 F.2d 901, 906-07 (6th Cir.1991);
Earley v. Champion International Corp.,
907 F.2d 1077, 1084 (11th Cir.1990);
In re Western Dist. Xerox Litigation,
140 F.R.D. 264, 269-70 (W.D.N.Y.1991). While inconvenience and prejudice to defendants from the discovery is an important factor, these cases also reflect the diminished probative value of such statistics.
Because Mims’ proffered statistical evidence is based on a population far larger than her own employment unit, and there is no evidence that decisionmakers in her unit were governed by any broader policy, the evidence is not probative of disparate treatment on the basis of race or age.
B.Disparate Impact
Neither do the NAO-wide statistics establish Mims’ disparate impact case on the basis of race or age. Statistics alone can make out a prima facie case of disparate impact.
See Wards Cove Packing Co., Inc. v. Atonio,
490 U.S. 642, 650, 109 S.Ct. 2115, 2121, 104 L.Ed.2d 733 (1989) (citing
International Brotherhood of Teamsters v. United States,
431 U.S. 324, 339, 97 S.Ct. 1843, 1856, 52 L.Ed.2d 396 (1977);
Hazelwood School Dist. v. United States,
433 U.S. 299, 307-308, 97 S.Ct. 2736, 2741-2742, 53 L.Ed.2d 768 (1977)). But the statistics must be probative of an imbalance caused by a specific employment practice in the particular pool of workers to which the plaintiff belongs.
Wards Cove
at 650-651, 109 S.Ct. at 2121-2122. Mims’ statistics do not support any such practice.
The use of subjective criteria can constitute an identifiable practice in disparate impact cases.
Watson v. Fort Worth Bank and Trust,
487 U.S. 977, 989-90, 108 S.Ct. 2777, 2786-87, 101 L.Ed.2d 827 (1988);
Scales v. J.C. Bradford & Co.,
925 F.2d 901, 908 (6th Cir.1991);
Squire v. General Motors Corporation,
434 Mich. 884, 452 N.W.2d 210 (1990). But the plaintiff must link the statistics to a particular practice that caused the alleged disparity. This Mims has not done.
C.Conclusion
My scrutiny of the proffered statistical evidence is in keeping with the Supreme Court’s holding in
Daubert v. Merrell Dow Pharma
ceuticals, Inc.,
509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), that under the Federal Rules of Evidence (FRE), “the trial judge must ensure that any and all scientific testimony or evidence admitted is not only relevant, but reliable.” 509 U.S. at 589, 113 S.Ct. at 2795. This screening obligation of the judge is based on the requirement of FRE 702 that scientific evidence or testimony “assist the trier of fact to understand the evidence or to determine a fact in issue.”
Id.
at 591, 113 S.Ct. at 2795. “Rule 702’s ‘helpfulness’ standard requires a valid scientific connection to the pertinent inquiry as a precondition to admissibility.”
Id.
at 591-92, 113 S.Ct. at 2796.
Because statistical disparities within the NAO unit do not prove disparate treatment within Mims’ employment unit, the proffered statistical evidence is not relevant to her disparate treatment claim. Because these disparities do not tend to prove the existence of an identifiable practice which caused harm to Mims and others in the same protected class, the proffered evidence is not relevant to her disparate impact claim.
For the above reasons, and for the reasons set forth in my opinion of July 3, 1997, defendant’s motion for summary judgment is granted.