OPINION OF THE COURT
BARRY, Circuit Judge.In Lanning v. Southeastern Pennsylvania Transp. Auth., 181 F.3d 478 (3d Cir.1999) (hereafter “Lanning /”), we held that “under the Civil Rights Act of 1991,1 a discriminatory cutoff score on an entry level employment examination must be shown to measure the minimum qualifications necessary for successful performance of the job in question in order to survive a disparate impact challenge.” 181 F.3d at 481. We found that the District Court did *288not employ this standard, which was implicit in Griggs v. Dulce Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971), and incorporated by the Act, and, thus, vacated the judgment of the District Court and remanded the appeal for the Court to determine whether the employer, the Southeastern Pennsylvania Transportation Authority (“SEPTA”), had carried its burden of establishing that its 1.5 mile run within twelve minutes measures the minimum aerobic capacity necessary to perform successfully the job of a SEPTA transit police officer. Id. at 494, 91 S.Ct. 849. We left it to the discretion of the District Court to allow the parties to expand the record in keeping with our newly-announced standard.
Because we conclude that SEPTA produced more than sufficient competent evidence to support the finding that a pre-hire, pre-academy training aerobic capacity of 42.5 mL/kgdnin measures the minimum qualifications necessary for successful performance as a SEPTA transit police officer and has, thus, justified the conceded disparate impact on female candidates by showing business necessity, we will affirm the judgment of the District Court in favor of SEPTA. We have jurisdiction pursuant to 28 U.S.C. § 1291.2
We clearly do not write on a clean slate. The District Court conducted a twelve-day bench trial in January of 1998 after which it rendered a 162-page opinion detailing 878 findings of fact and 107 conclusions of law. Lanning v. Southeastern Transp. Auth., 1998 WL 341605 (E.D.Pa. June 25, 1998). On appeal, we, too, rendered a lengthy opinion with a lengthy dissent. To be sure, the majority opinion spent much time explaining how the standard announced therein came to be, but that opinion, and the dissent, discussed much more, including why SEPTA’s concern over public safety caused it to modify its hiring requirements, the history of this litigation, and key pieces of evidence. On remand, the District Court conducted a five-day hearing, after which it rendered a 69-page decision detailing yet another 153 findings of fact and 34 conclusions of law. Lanning v. SEPTA, 2000 WL 1790125 (E.D.Pa. Dec.7, 2000).
So much has been written and so little remains for determination that we do not believe it necessary to repeat what has been said before or, as does the dissent here, poke a hole here or there in one or more of the District Court’s extensive findings of fact and conclusions of law. There is, however, one undisputed fact which bears repetition because it sets the stage for what is to follow: it is undisputed that SEPTA management wanted to improve the crime fighting ability of SEPTA’s force and the fitness of its officers. As Judge Weis explained in Lanning I:
Concerned about its inability to control crime on its property, SEPTA instituted a three-pronged attack on the problem. It added a substantial number of officers, implemented a zone method of patrol, and adopted standards to improve the generally poor physical condition of its officers. Unlike many metropolitan police departments, SEPTA officers are deployed alone and on foot, engaging in physical activities more frequently than other law enforcement agencies.
*289The patrol zones present significant variations in conditions that affect the physical exertion of officers in the performance of their duties. Zone One, for example, has a climb of 30 to 50 steps from street level. Zone Three, a mixture of above and below-ground locations, borders a large shopping mall, featuring retail theft and pursuits that lead into the SEPTA transit system. Zone Five, which includes sports complexes, is characterized by long distances between stations. Zone Six includes the Temple University area, a scene of frequent crimes against students.
SEPTA officers must occasionally ask for assistance from their comrades in other zones. These calls are divided into two categories, “officer assists” and “officer backups.” An “assist” requires officers to respond immediately. Often the only method available to get to the scene quickly is a run of five to eight city blocks. An officer responding to an “assist” must preserve enough energy to deal effectively with a situation once arriving on the scene. SEPTA averages about 380 running assists per year. “Backups” are not as critical as “assists,” so officers generally use a “paced jog.” SEPTA averages about 1,920 “backups” annually.
181 F.3d at 494-95 (Weis, J., dissenting).
We also reiterate what we explicitly stated in Lanning I: the business necessity standard takes public safety into consideration. 181 F.3d at 490 n. 16. We observed, in Lanning I, that Congress viewed the “more liberal test for business necessity” adopted in Wards Cove Packing Co., Inc. v. Atonio, 490 U.S. 642, 109 S.Ct. 2115, 104 L.Ed.2d 733 (1989), as a significant departure from Griggs and intended, when it enacted the Civil Rights Act of 1991, to endorse the business necessity standard enunciated in Griggs and not the Wards Cove interpretation of that standard. We, thus, considered, in Lanning /, only Griggs and its pre-’Wards Cove progeny. Nonetheless, we noted:
[T]o the limited extent that the Supreme Court’s pre-Wards Cove jurisprudence instructs that the public safety is a legitimate consideration, application of the business necessity standard to SEPTA is consistent with that jurisprudence because the standard itself takes public safety into consideration. If, for example, SEPTA can show on remand that the inability of a SEPTA transit officer to meet a certain aerobic level would significantly jeopardize public safety, this showing would be relevant to determine if that level is necessary for the successful performance of the job. Clearly a SEPTA officer who poses a significant risk to public safety could not be considered to be performing his job successfully. We are accordingly confident that application of the business necessity standard to SEPTA is fully consistent with the Supreme Court’s pre-Wards Cove jurisprudence as required by the Act.
181 F.3d at 491 n. 16.
It is against this backdrop that we assess the sole issue we caused to be resolved on remand: whether or not SEPTA has proven that its 42.5 mL/kg/min aerobic capacity standard measures the minimum qualifications necessary for the successful performance of the job of SEPTA transit police officers. The District Court concluded that the answer was “yes,” and that any lesser standard “would result in officers ... who were a danger to themselves, other officers, and the public at large, [and] unable to effectively fight and deter crime.” Lanning 2000 WL 1790125, at *28 (Conclusion # 32). In the course of our concluding that the Court’s conclusion *290was correct, we have reviewed the relevant findings of fact under the clearly erroneous standard and the District Court’s conclusions of law under the plenary review standard. Newark Branch, NAACP v. City of Bayonne, 134 F.3d 113, 120 (3d Cir.1998); Banning I, 181 F.3d at 484-85.
We start by noting that when this case was last before us, we expressed concerns as to certain aspects of the statistical studies upon which the District Court relied, and we “encourage[d] the District Court to take a critical look at these studies, if necessary, on remand.” 181 F.3d at 493 n. 21. In its second memorandum opinion, the District Court noted that it had “indeed taken a second critical look at these studies as suggested and once again reaffirm[ed] their validity consistent with its prior extensive 160 plus page memorandum opinion.” Lanning, 2000 WL 1790125, at *1 n. 1. Significantly, after the five-day hearing on remand, the District Court made additional findings of fact that bolster its conclusion. As the following brief discussion demonstrates, it is evident that the District Court’s findings regarding the studies in this case were not clearly erroneous.
In Banning I, we were concerned that “the absolute number of arrests or ‘arrest rates’ [in certain studies did] not necessarily correlate with successful job performance.” 181 F.3d at 492 n. 21. While we continue to believe that a “SEPTA officer should generally attempt to control a situation without having to make an arrest,” id., the District Court found, based upon its review of the extensive body of evidence before it, that “lost arrests have a significant impact on the public safety” in the SEPTA transit system. Lanning, 2000 WL 1790125, at *26. This finding indicates that arrests do, in fact, correlate to successful performance as a SEPTA transit police officer.
We also questioned whether certain studies overemphasized the role of aerobic capacity in making arrests, and whether other studies placed too much emphasis on arrest rates for “serious crimes” given that, historically, the bulk of SEPTA arrests has been for lesser crimes. While not all SEPTA arrests are aerobic contests,3 nor are they always effectuated to apprehend “serious” criminals,4 the District Court found that “[a]n inability to proficiently perform any ... task[ ] would compromise the effectiveness of the SEPTA transit police.” 2000 WL 1790125, at *24 (emphasis added). In essence, the Court concluded what, to us, is now evident: a SEPTA transit police officer must be ready and able to apprehend not just the numerous sedentary, petty criminals, but also the fleet-footed few who, from time to time, wreak serious harm on the people of Philadelphia.5
A final concern mentioned in Lanning I involved the statistical methods relied upon in some of SEPTA’s studies. We *291noted, for example, that a study on arrest rates seemed to be based on “disproportionately large numbers of officers with an aerobic capacity over 42 mL/kg/min” and stated that, if the correlation coefficient of the study was low, the study would “be subject to close review.” 181 F.3d at 492-93 n. 21 (citing 29 C.F.R. § 1607.14(B)(6)). Upon further review, we are satisfied that the correlation coefficient was sufficiently high to be statistically significant.
We have also examined the other relevant studies in this case for possible deficiencies and conclude that they were sufficiently reliable and in keeping with sound principles of statistical analysis. As scientists well know, it is not uncommon for competent statisticians to interpret data in different ways, sometimes reaching different results. While we concede that there may be some room for disagreement — as there often is in the realm of statistical analysis — the District Court carefully considered the relevant studies and credited them. Under the standard that governs our review, we conclude that the District Court’s findings of fact were not clearly erroneous.
And so we move more directly to the critical issue before us — the minimum qualifications necessary in terms of aerobic capacity to successfully perform as a SEPTA transit police officer. Neither the District Court nor the parties have explicitly defined the key phrase “minimum qualifications necessary,” but a definition is implicit in the parties’ respective arguments and the District Court’s acceptance of that of SEPTA. SEPTA argued that the run test measures the “minimum qualifications necessary” because the relevant studies indicate that individuals who fail the test will be much less likely to successfully execute critical policing tasks. For example, the District Court credited a study that evaluated the correlation between a successful run time and performance on 12 job standards. The study found that individuals who passed the run test had a success rate on the job standards ranging from 70% to 90%. The success rate of the individuals who failed the run test ranged from 5% to 20%.6 The District Court found that such a low rate of success was unacceptable for employees who are regularly called upon to protect the public. In so doing, the District Court implicitly defined “minimum .qualifications necessary” as meaning “likely to be able to do the job.”
Plaintiffs argued, however, that within the group that failed the run test, significant numbers of individuals would still be able to perform at least certain critical job tasks. They argued that as long as some of those failing the run test can do the job, the standard cannot be classified as a “minimum.” In essence, plaintiffs proposed that the phrase “minimum qualifications necessary” means “some chance of being able to do the job.” Under this *292logic, even if those failing the test had a 1% chance of successfully completing critical job tasks, the test would be too stringent.
We are not saying, as our distinguished brother in dissent suggests we are saying, that “more is better.” While, of course, a higher aerobic capacity will translate into better field performance — at least as to many job tasks which entail physical capability — to set an unnecessarily high cutoff score would contravene Griggs. It would clearly be unreasonable to require SEPTA applicants to score so highly on the run test that their predicted rate of success be 100%. It is perfectly reasonable, however, to demand a chance of success that is better than 5% to 20%.7 In sum, SEPTA transit police officers and the public they serve should not be required to engage in high-stakes gambling when it comes to public safety and law enforcement. SEPTA has demonstrated that the cutoff score it established measures the minimum qualifications necessary for successful performance as a SEPTA officer.
The dissent concedes that SEPTA has the right to improve its workforce and does not suggest that that is not being done. Instead, the dissent concentrates its efforts on why, in its view, the 42.5 mL/kg/ min aerobic capacity cutoff score as an application requirement is wanting, concluding that “[a]fter all has been said and done, ... one unassailable fact remains. The 42.5 mL/kg/min aerobic capacity [cutoff] is not required of transit officers before or after they begin policing.” Dissent at 306. As for the “before,” we reject without more the argument that applicants — male and female — should not be tested until they have graduated from the police academy, perhaps two and one-half years after they first applied to SEPTA; indeed, the dissent recognizes but relegates to a footnote the increase in SEPTA’S costs and the uncertainty in planning and recruitment this would occasion. As for the “after,” all incumbents — male and female — are now required to take a physical fitness test every six months, another step toward improving the workforce.8 In this connection, it bears mention that SEPTA is unable to discipline incumbents who do not pass the test only because of the patrol officers’ union’s challenge, sustained by an arbitrator. With the union’s blessing, however, SEPTA offers financial incentives to those officers who do pass.
One final note. While it is undisputed that SEPTA’s 1.5 mile run test has a disparate impact on women, it is also undisputed that, in addition to those women who could pass the test without training, nearly all the women who trained were able to pass after only a moderate amount of training. It is not, we think, unreasonable to expect that women — and men — who wish to become SEPTA transit officers, and are committed to dealing with issues of public safety on a day-to-day basis, would take this necessary step. Moreover, we do not consider it unreasonable for SEPTA to require applicants, who wish to train to meet the job requirements, to do so before applying in order to demonstrate their commitment to physical fitness. The poor physical condition of SEPTA officers prior to 1989 demonstrates that not every officer is willing to make that commitment once he or she is hired. In any event, the multi-agency training which SEPTA candi*293dates receive does not provide sufficient physical fitness training to bring an unqualified candidate up to the physical standard requirements.
The judgment of the District Court will be affirmed.9
. 42 U.S.C. § 2000e-2(k).
. This case was consolidated before the District Court with United States v. Southeastern Pennsylvania Transportation Authority (D.C.Civ. No. 97-cv1161). The United States filed a notice of appeal from the final judgment now under review, with the appeal docketed at No. 01-1297. On October 1, 2001, the United States moved to dismiss its appeal pursuant to Fed. R.App. P. 42(b). That motion was granted on October 12, 2001.
. Although there is some dispute over the precise percentage of arrests that have required substantial aerobic exertion, the evidence shows that, at the very least, around one in ten arrests, or ten percent, have so required.
. Arrests for "serious crimes” have similarly accounted for around one in ten arrests.
. The District Court also credited studies which indicate that once SEPTA officers reach suspects, they often must engage in physical struggles which require substantial aerobic capacity. The Court further noted that SEPTA officers "must engage in at least one aerobic encounter during the course of his or her duties every month, either as an emergency assist or running backup.” 2000 WL 1790125, at *24. These findings support the conclusion that aerobic capacity is an important factor in successfully making arrests.
. The District Court cited numerous other studies that offer similar results. .In one such study, 80% of those passing SEPTA's run test met minimum job standards, while only 33% of those failing did. Another study showed that 84% of those passing the test could carry out an "emergency assist,” while only 14% of the failing group were able to do so. The consideration that the District Court gave to these studies lays to rest plaintiffs’ claim that the cutoff time was merely the product of the judgment of SEPTA experts. As we noted in Lanning I, a "business necessity standard that wholly defers to an employer's judgment as to what is desirable in an employee ... is completely inadequate." Id. at 489 (emphasis added). The factual record here, however, clearly demonstrates that SEPTA experts set the run time cutoff at 12 minutes for objective reasons, with the studies showing that the projected rate of success of job applicants dropped off markedly for those who ran 1.5 miles in over 12 minutes.
. Even in other studies where those failing the test had higher predicted rates of success on job tasks, the rates were still only around 33%.
. Of course, yet another step in improving the performance of incumbents would be to require a physical fitness test not only upon application but also immediately prior to entry on duty.
. We have considered the parties' other assertions of error and find them unavailing.