*309MEMORANDUM **

Don Zottola, a white male, alleges on appeal that the oral interview component of the City of Oakland’s (“Oakland”) entry level examination for the position of a firefighter discriminates against white males in violation of Title VII of the Civil Rights Act, 42 U.S.C. § 2000e. Specifically, Zottola contends that the district judge erred when she: (1) dismissed his disparate treatment claim as a matter of law; (2) denied his motion to grant judgment as a matter of law in his favor on the disparate impact claim; (3) improperly instructed the first jury about the meaning of “business” “necessity” and “job relatedness”; (4) granted a new trial on alternative measures only rather than permitting a new trial on the entire disparate impact claim; and (5) excluded evidence of possible alternatives at the second trial. We affirm on all grounds.

A. Disparate Treatment

Zottola argues that he presented sufficient evidence to establish a prima facie case of intentional discrimination and that, as a result, the district court erred when, pursuant to Federal Rule of Civil Procedure 50 (“FRCP 50”), it dismissed his disparate treatment claim as a matter of law following the presentation of evidence. A prima facie ease of intentional discrimination may be proven either through direct evidence of intentional discrimination or through the presentation of evidence sufficient to satisfy the four elements of a prima facie case set forth in McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973). The prima facie McDonnell Douglas showing may, however, be rebutted if the employer articulates a legitimate, nondiscriminatory reason for the employee’s rejection. The employee must then present evidence showing that the nondiscriminatory reason offered by the employer is a mere pretext for discrimination. See id at 802-04, 93 S.Ct. 1817.

Zottola’s direct evidence of discrimination consisted of statistics showing that the eight black candidates interviewed by Panel L — the panel that interviewed Zottola — scored significantly higher than the 30 white candidates interviewed by the panel, and that all three members of the panel were black. Athough statistical evidence is relevant in disparate treatment cases because “it can be used to establish a general discriminatory pattern in an employer’s hiring or promotion practices [and] is probative of motive,” Diaz v. American Telephone & Telegraph, 752 F.2d 1356, 1363 (9th Cir. 1985); Lowe v. Monrovia, 775 F.2d 998, 1008 (9th Cir. 1985), the sample size in Zottola’s case is too small, standing alone, to establish a prima facie case of intentional discrimination. See Stout v. Potter, 276 F.3d 1118 (9th Cir. 2002) (affirming the district court’s dismissal of a disparate impact claim because “[a] sample involving 6 female applicants in a pool of 38 applicants is likely too small to produce statistically significant results”); see also Watson v. Fort Worth Bank and Trust, 487 U.S. 977, 996, 108 S.Ct. 2777, 101 L.Ed.2d 827 (1988); Contreras v. City of Los Angeles, 656 F.2d 1267, 1272 (9th Cir. 1981) (“[S]mall sample size may, of course, detract from the value of [statistical] evidence.”); see also Morita v. Southern Cal. Permanente Med. Group, 541 F.2d 217, 220 (9th Cir. 1976) (“[Statistical evidence derived from an extremely small universe ... has little predictive value and must be disregarded,”). Moreover, given that roughly 38% of Oakland’s popu*310lation is African-American, it is not surprising that three members of an Oakland interview panel would be black. See Bay-Area Census Report as of March, 2001, available at http://census.abag.ea.gov/eities/Oakland.htm. The racial composition of the panel is certainly not sufficient, in combination with the insubstantial statistical evidence, to show intentional discrimination.

A plaintiff may also assert a prima facie case of intentional discrimination by presenting evidence that meets the four requirements of McDonnell Douglas, 411 U.S. at 792, 93 S.Ct. 1817. See also Texas Dep’t of Community Affairs v. Burdine, 450 U.S. 248, 101 S.Ct. 1089, 67 L.Ed.2d 207 (1981). To establish a McDonnell Douglas prima facie case of discrimination, a plaintiff must prove:

(i) that he belongs to a racial minority;

(ii) that he applied and was qualified for a job for which the employer was seeking applicants; (iii) that, despite his qualifications, he was rejected; and (iv) that, after his rejection, the position remained open and the employer continued to seek applicants from persons of complainants’ qualifications.

Burdine, 450 U.S. at 254 n. 6, 101 S.Ct. 1089; McDonnell Douglas, 411 U.S. at 802, 93 S.Ct. 1817. In McDonald v. Santa Fe Trail Trans., 427 U.S. 273, 280, 96 S.Ct. 2574, 49 L.Ed.2d 493 (1976), the Court held that Title VII prohibits racial discrimination against white applicants as well as minority applicants. Although we have held that reverse discrimination cases “fit[ ] within the analytic framework established in McDonnell Douglas Corp. v. Green,” Higgins v. Vallejo, 823 F.2d 351, 355 (9th Cir. 1987), we have not discussed how the courts should modify the first McDonnell Douglas requirement that the plaintiff “belongs to a racial minority” when analyzing a reverse discrimination claim. There is currently a circuit split on this issue. Compare Duffy v. Wolle, 123 F.3d 1026, 1036 (8th Cir. 1997) (holding that a plaintiff in a reverse discrimination case does not present a prima fade of discrimination unless he shows that “background circumstances support the suspicion that the defendant is that unusual employer who discriminates against the majority”); Murray v. Thistledown Racing Club, Inc., 770 F.2d 63, 67 (6th Cir. 1985) (same);1 Mills v. Health Care Serv. Corp., 171 F.3d 450, 457 (7th Cir. 1999) (same); Notari v. Denver Water Dep’t, 971 F.2d 585, 589 (10th Cir. 1992) (same); Russell v. Principi, 257 F.3d 815, 818 (D.C.Cir. 2001) (same) with Byers v. Dallas Morning News, Inc., 209 F.3d 419, 426 (5th Cir. 2000) (holding that a plaintiff in a reverse discrimination case need show only that he is a member of “a protected group” and whites are a protected group under Title VII); Bass v. Board of County Commissioners, 256 F.3d 1095, 1103-04 (11th Cir. 2001) (same).2

*311We need not take sides in this inter-circuit dispute to resolve Zottola’s case. Even assuming that Zottola was able to satisfy the four requirements of McDonnell Douglas and establish a prima facie case, the district judge’s dismissal of his intentional discrimination claim was still proper because the statistical evidence that Zottola presented was insufficient as a matter of law to show that Oakland’s articulated reason for his rejection — that it did not hire him because his score on the oral examination was not high enough — was pretextual. Accordingly, the district judge’s decision to grant Oakland’s Rule 50 motion to dismiss was not erroneous.

B. Disparate Impact

(1) Business Necessity

Zottola contends that the district court erred in denying his motion for judgment as a matter of law under FRCP 50 at the close of the evidence because Oakland failed to provide sufficient evidence that the oral interview selection process had been properly validated and therefore could not establish business necessity. The question whether a test has been validated properly is “primarily a factual question, which depends on underlying factual determinations regarding the content and reliability of the validation studies that a defendant utilized.” Association of Mexican-American Educators (“AMAE”) v. California, 231 F.3d 572, 585 (9th Cir. 2000) (en banc). In cases that involve a scored test, as this case does, we “require that the test be ‘job related’ — that is, ‘that it actually measures skills, knowledge, or ability required for successful performance on the job.’ ” AMAE, 231 F.3d at 585; Contreras, 656 F.2d at 1271. Demonstrating “job-relatedness” is a three-step process:

The employer must first specify the particular trait or characteristic which the selection device is being used to identify or measure. The employer must then determine that the particular trait or characteristic is an important element of work behavior. Finally, the employer must demonstrate by “professionally acceptable methods” that the selection device is “predictive of or significantly correlated” with the element of work behavior identified in the second step.

Craig v. County of Los Angeles, 626 F.2d 659, 662 (9th Cir. 1980) (quoting Albemarle Paper Co. v. Moody, 422 U.S. 405, 431, 95 S.Ct. 2362, 45 L.Ed.2d 280 (1975)). Here, Zottola concedes that Oakland’s job analysis satisfied the first two steps, and it is only the third requirement — that Oakland demonstrate by professionally accepted methods that its oral interview process is predictive of or significantly correlated with the knowledge, skills, and abilities identified in its job analysis as important characteristics for an entry-level firefighter — that is at issue.

Oakland relied on content validation to demonstrate that its oral interview was predictive of the knowledge, skills, and abilities for an entry-level firefighter. According to the EEOC’s Uniform Guidelines on Employee Selection Procedures (“Guidelines”), “[ejvidence of validity of a test or other selection procedure by a content validity study should consist of data showing that the content of the selection procedure is representative of important aspects of performance on the job for which the candidates are to be evaluated.” 29 C.F.R. § 1607.5(B). The evidence that *312Oakland presented to validate its oral interview included pre-testing results collected as part of the job analysis; statistical inter-rater reliability studies that were conducted to ensure that the panel scores were reflective of the candidate rather than the rater; anecdotal evidence that the Fire Department “raved” about the candidates who had been hired and that they were performing well in the academy; and expert testimony that these “behavioral consistency orals yield the most reliable and valid results” and that the use of open-ended oral interview questions is appropriate, professionally accepted, and prevalent throughout the country. We hold that there was sufficient evidence of validation to send the question to the jury and that the district court did not err in denying Zottola’s FRCP 50 motion.

(2) Jury Instruction

Zottola contends that the district court improperly instructed the jury when it defined “job relatedness” and “business necessity.” Whether a jury instruction properly states the elements to be proved at trial is a question of law that the panel reviews de novo. Gizoni v. Southwest Marine Inc., 56 F.3d 1138, 1142 (9th Cir.), cert. denied, 516 U.S. 944, 116 S.Ct. 381, 133 L.Ed.2d 304 (1995). Because the instruction that the district court gave adequately defined “job relatedness” and “business necessity,” we hold that there was no error.

(3) Alternative Measures

Zottola contends that the district court erred when it granted Oakland a new trial on alternative measures. We disagree. After the first jury returned its verdict, Oakland moved, under FRCP 50, for judgment as a matter of law on alternative measures and alternatively, for a new trial. Because Oakland’s motion for judgment as a matter of law was made for the first time after the jury had rendered its verdict and was not a renewal of a FRCP 50 motion made at the close of the evidence, the district court could grant it only if the record showed that there was no evidence to support the verdict. See Patel v. Penman, 103 F.3d 868, 878 (9th Cir. 1996). In this case, Zottola did present some evidence of possible alternatives. Therefore, the district court’s decision to deny Oakland’s motion for judgment as a matter of law was proper. We hold, however, that its decision to grant a new trial was proper.

A trial court may grant a new trial if the verdict is contrary to the clear weight of the evidence, is based on false evidence, or constitutes a miscarriage of justice. Gasperini v. Center for Humanities, 518 U.S. 415, 433, 116 S.Ct. 2211, 135 L.Ed.2d 659 (1996); Hanson v. Shell Oil Co., 541 F.2d 1352, 1359 (9th Cir. 1976). In this case, although Zottola’s expert testified at the first trial about some possible alternatives, he did not discuss how effective any of them would be in measuring the knowledge, skills, and abilities that were essential to the entry-level firefighter position. Nor did he discuss whether they would be less discriminatory than the oral interview process. Thus, the district court did not abuse its discretion when it ruled that the first jury’s decision, that there were equally effective alternative tests that would have had a less discriminatory impact, was contrary to the clear weight of the evidence. Accordingly, we affirm its decision to grant a new trial.

ft) Limiting the Issues at the Second Trial

Zottola contends that there is too much overlap in the evidence necessary to show business necessity and reasonable alternatives, and the issues are too inter*313woven, to permit the district court to split the issues and hold a re-trial on alternative measures alone. We disagree. Courts have broad discretion to grant new trials on all or only some issues. Gasoline Prods. Co. v. Champlin Refining Co., 283 U.S. 494, 51 S.Ct. 513, 75 L.Ed. 1188 (1931); Hasbrouck v. Texaco, 663 F.2d 930, 933 (9th Cir. 1981). The items to be retried must, however, be distinct and separable from the issues not to be retried. Hasbrouck, 663 F.2d at 933.

The burden of persuasion with respect to business necessity is on defendants, but the ensuing burden of showing reasonable alternatives shifts to the plaintiff. See 42 U.S.C. § 2000e-2(k). Because the shifting burdens of proof in a disparate impact case constitute distinct and separable stages of litigation, we hold that the district judge did not abuse her discretion when she permitted re-trial solely on the issue of alternative measures.

(5) Evidentiary Rulings at the Second Trial

Zottola contends that the district court erred in the second trial when it prevented a lay witness from testifying that her company presented Oakland with an alternative test that Oakland then rejected. We disagree. The district court’s decision to exclude evidence of this alternative was consistent with Federal Rule of Evidence 701’s mandate that lay witnesses not be permitted to offer expert opinions.3 Because the witness was not an expert witness, she could not testify about whether the alternative exam was equally effective or would result in a less discriminatory outcome. Any statement that she might offer would have required expertise that, under this record, she did not possess about test validation and statistical analyses. Although Zottola did provide the testimony of an expert, that witness did not testify as to whether the alternative test offered by the lay witness’s company was as effective as or less discriminatory than the oral interview, and was not prepared to do so. Accordingly, the district court properly excluded as irrelevant the lay witness’s proffered evidence that her company’s alternative test had been offered to Oakland.4

Zottola finally contends that it was error for the district court to prevent its expert at the second trial from testifying that one alternative available to Oakland was to modify the oral interview process so as to eliminate a panel rater’s racial bias. The district court excluded this testimony stating that it would be prejudicial if the jury heard speculative evidence suggesting that panel rater bias caused the disparate impact. Because we agree that the testimony was more prejudicial than probative, we affirm.

For all of the above reasons, we affirm the district court’s entry of judgment in favor of Oakland.

AFFIRMED.