MURPHY, Circuit Judge.
Matti Schultz was discharged by his employer, the McDonnell Douglas Corporation, which he then sued for age discrimination under 29 U.S.C. § 621 et seq., and the Missouri Human Rights Act, Mo.Rev.Stat. § 213.010 et seq.
Schultz was a principal technical specialist when he was discharged as part of a reduction in force. In the two performance evaluations before the reduction in force, Schultz had been given the lowest overall ratings of any of the sixteen specialists in his department. Schultz, who was fifty-seven and the fourth oldest specialist in his department when discharged, was the only specialist in that department to be terminated. At trial he attempted to introduce statistical evidence demonstrating a pattern of age discrimination by McDonnell Douglas. The statistics were based on data taken from the company's Production Operations Division, which consisted of 3,731 employees.
The district court excluded the statistical evidence from the divisional level because it was not based on comparable employees, citing Nitschke v. McDonnell Douglas Corp., 68 F.3d 249, 252 (8th Cir.1995), and Hutson v. McDonnell Douglas Corp., 63 F.3d 771, 777 (8th Cir.1995). The statistics dealt with a much larger group of employees than the district court found relevant; they were not comparable in its view because they did not include performance evaluations.
Schultz contends that he had offered additional evidence of age animus at the departmental level to show the reasons offered for discharge were pretextual. He also claims that there was evidence demonstrating the involvement of managers at the higher divisional level in his termination.
We review the denial of a motion for a new trial under an abuse of discretion standard. McKnight v. Johnson Controls, Inc., 36 F.3d 1396, 1400 (8th Cir.1994). The admissibility of evidence is also within the trial court's discretion. The question presented on the post-trial motion was whether exclusion of the evidence was "so prejudicial as to require a new trial which would be likely to produce a different result." O'Dell v. Hercules, Inc., 904 F.2d 1194, 1200 (8th Cir.1990).
The district court did not abuse its discretion in excluding the statistical evidence. While it is true that MacDissi v. Valmont Indus. Inc., 856 F.2d 1054 (8th Cir.1988), indicated that other evidence of discrimination may add to the relevance of statistical evidence, it did not contradict the rule that the statistics must evaluate comparable employees to be meaningful indicators of pretext. 856 F.2d at 1058-59; Hutson, 63 F.3d at 778 (discussing MacDissi). Even if the decision to discharge Schultz had been made with the involvement of division level managers, he has not shown that the 3,731 division employees reflected in the statistics were comparable to him. McDonnell Douglas claimed Schultz was selected for discharge because of his performance ratings, and the statistical evidence did not incorporate the performance ratings of other employees. The probative value of statistical evidence
For these reasons the district court did not err in denying the motion for a new trial. Accordingly, its order is affirmed.