Plaintiffs in this employment discrimination action appeal from the district court’s order granting them preliminary injunctive relief. Plaintiffs also appeal from the district court’s ruling on a motion in limine involving a statute of limitations question and a related evidentiary issue. For the reasons set forth below, we affirm in part and reverse in part.
This appeal involves two complaints that were filed by overlapping groups of black police officers of the City of Akron challenging the promotion policy of the City’s police department. On September 21,1984, the Black Law Enforcement Officers Association (“BLEOA”); Ed Irvine, the sole black lieutenant on the police force; and Harold Craig, the sole black sergeant on the force, filed a class action suit against the defendants: the City of Akron; the Fraternal Order of Police, Lodge No. 7; Joseph P. Wheeler, President of the Civil Service Commission; Phillip Barnes, Akron’s Chief of Police; and Sidney Foster and Virgil Collins, members of the Civil Service Commission. Plaintiffs alleged that the police department’s promotional policies racially discriminated against blacks in violation of 42 U.S.C. §§ 1981 and 1983 (1982), and the equal protection and due process clauses of the Fourteenth Amendment.1 Plaintiffs sought declaratory and injunctive relief, as well as compensatory and punitive damages.
In May 1985, plaintiffs filed a motion seeking a preliminary injunction preventing the administration of a proposed promotional test for sergeant that was scheduled to be given in June 1985. The district court allowed the test to be given, but stayed any promotions pending a hearing on the merits.
On September 23, 1985, plaintiffs William W. Ellison, Donnie W. Whitworth, Dennis Johnson Jr., Leonard Mitchell Jr., O’Dell Daniels, and Douglas Prade, individual black police officers in the Akron Police Department who had sat for the June 1985 promotional exam, also filed suit against *477the same defendants. These plaintiffs alleged the same claims set forth by the plaintiffs in the BLEOA case, as well as a claim under Title VII of the 1964 Civil Rights Act, 42 U.S.C. § 2000e et seq. (1982). These plaintiffs sought the same relief as the BLEOA plaintiffs. The two cases were consolidated by the district court on October 4, 1985.
The contested promotional test was administered as scheduled on June 3, 1985. One hundred and sixty Akron patrolmen sat for the exam, including eighteen blacks. The exam consisted of five weighted components. The components and their respective weights are as follows:
Component Maximum Weight
1) Job knowledge test 42%
2) Test battery 14%
3) Performance appraisal 14%
4) Seniority 10%
5) Service rating 20%
The job knowledge component involved a written examination that required candidates to answer questions regarding the necessary knowledge, skill, and ability needed by sergeants to perform specific tasks and duties. The test battery involved a written examination designed to measure the following factors: practical reasoning ability, memory, leadership, cognitive structure, and lack of impulsiveness — all personality traits that mark an individual’s ability to make careful decisions. The performance appraisal aspect of the test involved the evaluation of individual candidates by their superiors, who were required to rate the performance and abilities of the individual patrolmen under their supervision. The service rating aspect of the examination involved rating reports completed twice per year for each officer by supervisory personnel. The report is designed to measure an officer’s quality and quantity of work, attendance, work habits, relationship with others, and supervisory skills. The ratings given are outstanding, very good, satisfactory, improvement needed, or unsatisfactory. Service ratings have been used in promotional examinations by the City for approximately fifty years. The final component of the examination, seniority, measured credit given to patrolmen based upon length of continuous service with the City up to a maximum of fourteen years. A candidate received one point for each of his first four years of service and six-tenths of a point for each of the next ten years of service.
The motion for preliminary injunction was heard over eight days from October 6, 1985, to December 17, 1985. The issue at the hearing was whether the City should be enjoined from promoting officers based on the results of the June 3, 1985 exam. The court heard extensive expert testimony from both sides as to the development of the promotional procedures used by the City and their potential adverse impact. Individual plaintiffs testified regarding the damage they would sustain if promotions were made based upon the June 3d exam. Chief Barnes also testified as to the City’s immediate need for sergeants and the adverse impact that preventing promotions would have on the City.
During the course of the proceedings the court ordered two separate rankings of candidates based on the June 3d exam. Ranking # 1 was based on the scores of all five components of the examination. Ranking # 2 excluded the scores from the performance appraisal component of the examination.
The court issued a memorandum opinion on January 23, 1986, in which it reviewed each of the components of the examination and concluded that plaintiffs had failed to establish that they were likely to prove that the test battery, job knowledge test, service rating, or seniority components of the promotional exam were discriminatory. The court concluded, however, that plaintiffs had shown that they could likely prove that the performance appraisal component of the test is discriminatory. The court noted that:
[T]he plaintiffs have made a showing of disparate impact. Additionally, plaintiffs and intervenors have jointly established that the performance appraisal review was so arbitrary, subjective, capricious, and subject to potential and actual manipulation as to constitute a violation of plaintiffs’ and intervenors’ rights to *478equal protection and due process of the law. The Court finds that together, this constitutes a showing of discriminatory intent on the basis of race sufficient to meet the requisite standard under §§ 1981, 1983 and the constitutional claims.
App. 77 (footnote omitted). The court next addressed the validity of the two court-ordered rankings based on the June 3d exam and found that ranking # 1, which included scores from the performance appraisal component, was suspect. The court also found that plaintiffs would suffer irreparable injury if permanent promotions were made from a list based upon a scoring of the promotional exam involving all five components. The court reasoned that if the plaintiffs had a legitimate right to promotions, they would undoubtedly suffer irreparable harm if white officers were permanently promoted to positions to which black officers are ultimately entitled. The City was therefore enjoined from making promotions based on this ranking.
The court found, however, that ranking #2, which excluded the performance appraisal component, was not discriminatory. The court found that plaintiffs had not been able to establish disparate impact with regard to ranking # 2 through the use of statistics and observed that the number of minorities who took the exam was so small that any assessment of adverse impact was speculative. The court also noted that plaintiffs had not established adverse impact through the EEOC’s “four fifths” or “80% rule-of-thumb” rule.2 The court observed that plaintiffs’ expert witness had testified that “[w]hen the performance appraisal was eliminated from the total ranking, the percentage who were passed among whites was 46.5%, and among blacks 38.9%. The proportion of those ratios to one another is 84%, which meets the four fifths rule.” App. 101. When the performance appraisal component was maintained, 50 percent of the blacks and 75 percent of the whites passed.
Having found that plaintiffs would suffer irreparable injury if permanent promotions were made based on the June 3d exam, the court next addressed the questions of whether substantial harm would be caused to the City by issuance of the preliminary injunction and whether the public interest would be served by issuance of the injunction. The court first found that a large number of sergeants’ vacancies existed in the ranks of the police department. There are approximately sixty-five sergeants’ positions in the Akron police department. At the time of the hearing, there were fourteen to fifteen openings for the position of sergeant. Relying in large part upon the testimony of the chief of police that the presence of the requisite number of sergeants was crucial to the safety of citizens, the court found that the currently existing shortage of sergeants would constitute a substantial degree of harm to the City if a preliminary injunction blocking promotions were issued. The court therefore concluded that the interests of the City would best be served by the prompt promotion of officers to sergeant.
The court, relying on traditional equitable principles, balanced the interests involved and concluded that “the best solution to the problem ... is a temporary one, which allows for the temporary promotion of sergeants based upon Court Ordered ranking No. 2, the eligibility list without the discriminatory performance appraisal included.” App. 114. The court therefore enjoined the City from making permanent promotions but allowed the temporary promotion of twenty-two officers to the rank of sergeant. These promotions were subject to revocation in the event that plaintiffs were ultimately successful on the merits of the case. The City promoted twenty-two officers to sergeant. Two of these officers are black.
*479The City subsequently made a motion in limine asking the court to limit evidence presented in the case to events that occurred within a one-year statute of limitations period. The court granted the motion, finding that Wilson v. Garcia, 471 U.S. 261, 105 S.Ct. 1938, 85 L.Ed.2d 254 (1985), and Mulligan v. Hazard, 777 F.2d 340 (6th Cir.1985), cert. denied, — U.S. -, 106 S.Ct. 2902, 90 L.Ed.2d 988 (1986), required the application of Ohio’s one-year personal injury statute of limitations to plaintiffs’ § 1981 and § 1983 claims.
Trial was scheduled to begin in May 1986. However, plaintiffs appealed the preliminary injunction order. The district court also issued an order pursuant to 28 U.S.C. § 1292(b) (1982) certifying the statute of limitations issue to this court. The appeals were consolidated and are now before this court.
I.
On appeal, plaintiffs challenge the nature of the relief afforded by the district court in the preliminary injunction. A court of appeals may review a district court’s denial or grant of a preliminary injunction only to determine whether the court has abused its discretion. American Motors Sales Corp. v. Runke, 708 F.2d 202, 205 (6th Cir.1983). “A district court abuses its discretion when it relies on clearly erroneous findings of fact, or when it improperly applies the law or uses an erroneous legal standard.” Christian Schmidt Brewing Co. v. G. Heileman Brewing Co., 753 F.2d 1354, 1356 (6th Cir.), cert. dismissed, 469 U.S. 1200, 105 S.Ct. 1155, 84 L.Ed.2d 309 (1985) (citations omitted). There are four factors that must be considered in determining whether the district court has abused its discretion in granting or denying a preliminary injunction: (1) the likelihood of success on the merits of the action; (2) the irreparable harm that could result without the relief requested; (3) the impact on the public interest; and (4) the possibility of substantial harm to others. Id. at 1356; Tate v. Frey, 735 F.2d 986, 990 (6th Cir.1984).
Plaintiffs contend that the district court’s order should be reversed for the following reasons. First, the court’s findings of fact are clearly erroneous. Second, black policemen will be injured and white policemen unjustly benefitted by the denial of the relief being sought. Third, if one aspect of the promotional test is suspect, then the entire test is tainted and no promotions should be made based on any aspect of the test.
Our review of a district court’s findings of fact is limited to determining whether the findings are clearly erroneous. Fed.R.Civ.P. 52(a). A district court’s findings of fact will be deemed clearly erroneous on review only if, after reviewing the entire record, the appellate court “is left with a definite and firm conviction that a mistake has been committed.” Jackson v. RKO Bottlers of Toledo, Inc., 743 F.2d 370, 374 (6th Cir.1984). The record on appeal is voluminous. The district court made extensive factual findings regarding each of the five components of the test. The court examined a great deal of evidence and heard the testimony of expert witnesses, several police officers, and representatives of the City. Based on this evidence, the court concluded in a well-reasoned, 72-page opinion that plaintiffs were likely to prove that only one of the five components of the promotional exam was discriminatory. We have thoroughly reviewed the record, and we are convinced that the district court’s findings of fact are not clearly erroneous.
Plaintiffs presented a great deal of evidence regarding the discriminatory impact of the performance appraisal component of the test. This evidence included statistical proof that white officers regularly received higher ratings than black officers. The mean rating for white officers was 79.28, while the mean for blacks was 72.96. This difference is statistically significant and is the greatest difference between the mean scores of blacks and whites on any of the five components of the test. Additionally, considerable testimony was elicited from several officers involved in the appraisal process that strongly suggested that the *480rating guidelines were not being followed and that outright manipulation of ratings had occurred. Furthermore, officers in certain divisions of the police department, such as the investigative division, which has no black officers, consistently received higher ratings than officers in other divisions. From this, the district court properly concluded that there was a strong likelihood that plaintiffs would be able to prove that the performance appraisal component of the test had a discriminatory impact on black officers.
We now turn to an examination of the four remaining components.3 We will examine the evidence regarding the discriminatory impact of the job knowledge test and the test battery in tandem. Title VII does not forbid the use of employment tests. Griggs v. Duke Power Co., 401 U.S. 424, 436, 91 S.Ct. 849, 856, 28 L.Ed.2d 158 (1971). However, Title VII does forbid the use of employment tests that have a discriminatory impact and are not job-related. Id.
The Supreme Court has set forth a three-step analysis to determine whether such procedures are discriminatory. First, the complaining party must establish a pri-ma facie case that the test is discriminatory. This may be done by showing that the challenged procedure has a disproportionate impact on minorities. Albemarle Paper Co. v. Moody, 422 U.S. 405, 425, 95 S.Ct. 2362, 2375, 45 L.Ed.2d 280 (1975). If the complaining party succeeds in carrying this burden, the defendant must then demonstrate that the procedure has a “manifest relationship to the employment in question.” Griggs, 401 U.S. at 432, 91 S.Ct. at 854; see also Connecticut v. Teal, 457 U.S. 440, 446-47, 102 S.Ct. 2525, 2530, 73 L.Ed.2d 130 (1982). The defendant may meet this burden by establishing that the procedure used measures important skills, abilities, and knowledge that are necessary for the successful performance of the job. See 29 C.F.R. § 1607.5 (1986); see also Contreras v. City of Los Angeles, 656 F.2d 1267, 1283 (9th Cir.1981), cert. denied, 455 U.S. 1021, 102 S.Ct. 1719, 72 L.Ed.2d 140 (1982). In other words, the defendant must show, “by professionally acceptable methods, [that the test is] ‘predictive of or significantly correlated with important elements of work behavior which comprise or are relevant to the job or jobs for which candidates are being evaluated.’ ” Albemarle Paper Co., 422 U.S. at 431, 95 S.Ct. at 2378 (quoting 29 C.F.R. § 1607.4(c) (1974)). If the defendant successfully carries this burden, the complaining party must show that “other tests or selection devices without a similarly undesirable racial effect, would also serve the employer’s legitimate interest in ‘efficient and trustworthy workmanship.’ ” Id. at 425, 95 S.Ct. at 2375 (quoting McDonnell Douglas Corp. v. Green, 411 U.S. 792, 801, 93 S.Ct. 1817, 1823, 36 L.Ed.2d 668 (1973)). Such a showing is akin to presenting evidence that the employer is using the test as a pretext for discrimination. Id.
The district court applied this analysis to the challenged tests and found that although plaintiffs had established a prima facie case of discrimination through evidence showing that white candidates fared better than black candidates, the defendants had successfully shown that the tests were job-related. The job-relatedness of the tests was established primarily through the testimony of Dr. Gerald Barrett, who was involved in the development of both tests, and Dr. Frank J. Landy, defendants’ expert witness. Dr. Barrett testified that in-depth research was done to determine which qualities and traits are related to the successful performance of the job of sergeant. Dr. Barrett’s testimony was supported by a validation study conducted to determine the job-relatedness of the test battery. This study indicated that the elements chosen to be included in the test *481were positively correlated with job performance. Additionally, Dr. Landy testified that the tests:
represents] knowledges, abilities and personal characteristics that would be implicated in the supervisory positions in law enforcement agencies. Further, the available technical documentation and statistical analysis suggest measures of suitable psychometric integrity. Finally, administrative procedures related to these components seem prudent and represent generally acceptable practice.
App. 70. We are in agreement with the district court that the defendants successfully carried their burden of showing that the tests were job-related. The only remaining question is whether plaintiffs successfully rebutted this showing. Again, we agree with the district court that plaintiffs failed to do this. Plaintiffs failed to present adequate evidence rebutting or contradicting the testimony of Dr. Barrett and Dr. Landy.
We next consider the seniority component of the promotional procedure. In International Brotherhood of Teamsters v. United States, 431 U.S. 324, 97 S.Ct. 1843, 52 L.Ed.2d 396 (1977), the United States Supreme Court held that a bona fide seniority system does not violate Title VII even if it perpetuates past discrimination. The Court reasoned that § 703 (h) of the Civil Rights Act, 42 U.S.C. § 2000e-2(h) (1982), dictated such a result.4 The Court concluded that the
purpose of § 703(h) was to make clear that the routine application of a bona fide seniority system would not be unlawful under Title VII.... [T]he congressional judgment was that Title VII should not outlaw the use of existing seniority lists and thereby destroy or water down the vested seniority rights of employees simply because their employer had engaged in discrimination prior to passage of the Act.
Id. at 352-53, 97 S.Ct. at 1863. A seniority system will be valid so long as an intent to discriminate did not enter into its adoption and it has been maintained free from any illegal purpose. Alexander v. Aero Lodge No. 735, 565 F.2d 1364, 1378 (6th Cir.1977), cert. denied, 436 U.S. 946, 98 S.Ct. 2849, 56 L.Ed.2d 787 (1978). In the instant case, the burden fell on plaintiffs to show the existence of such a discriminatory intent or illegal purpose. The district court concluded that plaintiffs failed to carry this burden. We agree. The seniority system used in the promotion process is neutral and applies equally to all officers, regardless of their race.
The final component we examine is the service rating component. Having reviewed the service rating form that is filled out by supervisory personnel, we conclude that the district court was correct when it held that the service rating system did not constitute an unlawful bar for blacks in the promotional process. The service rating process is clearly neutral on its face, and plaintiffs failed to present sufficient evidence that it has a disparate impact on black officers. Having examined these four components of the examination, we conclude that the district court’s finding that plaintiffs were unlikely to prove that any of them were discriminatory is not clearly erroneous.
Plaintiffs next argue that if, as the district court found, one component of the test is discriminatory, then the entire examination process is tainted, and no portion of it should be used for promotional purposes. Plaintiffs further contend that once the district court found that one component of the test was discriminatory, it should have completely enjoined all promotions based on the test, and erred by allowing temporary promotions to be based on ranking # 2. Plaintiffs raise a valid point and under different circumstances the position advanced by them might be preferable. However, in this instance the district court specifically found that a failure to promote any officers to sergeant prior to the com-*482píete resolution of the case would be potentially harmful to the City. Consequently, some method of filling the vacancies, even if only temporarily, had to be established. The court balanced the' equities of the situation and decided in favor of allowing temporary promotions. We hold that the district court did not abuse its discretion by allowing temporary promotions to be made from ranking #2.
Plaintiffs finally argue that the remedy proposed by the district court does not comport with the dictates of Title VII because it unjustly benefits white officers at the expense of black officers.5 Specifically, plaintiffs contend that since the end result of the court’s order will be the promotion of twenty white officers — persons who have not been discriminated against — the order does not comport with Title VII, which is designed to protect the victims of discrimination. Plaintiffs further argue that the temporary promotions will give white officers an unfair advantage over black officers because those whites promoted will receive on-the-job training, and will thus be more capable and better prepared to take a nondiscriminatory examination at a later date in the event that the court ultimately finds that the examination procedure currently being used is discriminatory-
Plaintiffs would apparently prefer that no promotions be made pending resolution of this suit. However, this position ignores the very real need for immediate promotions. This suit has been pending for nearly three years. It is impossible to predict exactly how much longer it will be before the suit is completely resolved. If no promotions are allowed pending resolution of this action, it would appear incontestable that the vacancies in the ranks of sergeant will continue to grow, as will the potential harm to the City. In light of this, the solution offered by plaintiffs — no promotions pending resolution of the suit — is unreasonable. In contrast, the solution offered by the district court is reasonable. Thus, although we find some merit in the arguments raised by plaintiffs, we hold that the district court’s order was not an abuse of discretion and the order is AFFIRMED.
II.
Plaintiffs also challenge the district court’s order granting defendants’ motion in limine preventing plaintiffs from presenting evidence as to any alleged discriminatory actions of defendants that occurred more than one year prior to the date plaintiffs filed their complaints. The district court granted the motion after holding that § 1981 and § 1983 actions commenced in Ohio are governed by Ohio’s one-year statute of limitations for tort actions.6 See Ohio Rev.Code Ann. § 2305.11. (Baldwin 1984). The district court gave no explanation for its decision to use the statute of limitations as a bar to the admission of evidence.
It is clear that the district court erred in using the statute of limitations to bar the admission of evidence. The func*483tion of a statute of limitations is to bar stale claims. American Pipe & Constr. v. Utah, 414 U.S. 538, 554, 94 S.Ct. 756, 766, 38 L.Ed.2d 713 (1974). “The statute of limitations is a defense ..., not a rule of evidence. Therefore, ... [it] has no bearing on the admissibility of evidence.” United States v. Ashdown, 509 F.2d 793, 798 (5th Cir.), cert. denied, 423 U.S. 829, 96 S.Ct. 48, 46 L.Ed.2d 47 (1975). The decision whether to admit evidence is based on its relevancy and probativeness, see Fed.R. Evid. 401 and 403, not on whether the evidence is derived from events that occurred prior to a certain time period.
Plaintiffs accurately point out that if their presentation of evidence were limited to discriminatory actions that occurred one year prior to the filing of the action, it would be impossible to prove the pattern and practice of discrimination that is essential to their case. Plaintiffs’ complaints relate to what they perceive to be a long history of discrimination on the part of defendants in promotional decisions involving black officers. In order to prove this, plaintiffs will certainly have to present evidence of events that occurred more than one year prior to the commencement of their suit. See United States v. Garvin, 565 F.2d 519, 523 (8th Cir.1977) (evidence of events extending beyond the statute of limitations admissible to show motive, intent, or continuing scheme).
Accordingly, the district court’s order limiting relevant evidence to a period coextensive with the applicable statute of limitations is REVERSED.
. As of the time this complaint was filed, Akron had 99 supervisory positions in the police department. Only two of these supervisory positions were held by blacks.
. The rule is set forth at 29 C.F.R. § 1607.4(d) (1986):
A selection rate for any race, sex, or ethnic group which is less than four-fifths (Vs) (or eighty percent) of the rate for the group with the highest rate will generally be regarded by the Federal enforcement agencies as evidence of adverse impact, while a greater than four-fifths rate will generally not be regarded by Federal enforcement agencies as evidence of adverse impact.
. We analyze these components pursuant to Title VII precedent. Plaintiffs raised § 1981 and § 1983 claims in addition to their Title VII claim. Disparate treatment claims brought under Title VII, § 1981 and § 1983 require proof of purposeful discrimination. "Thus, the order and allocation of proof, applicable in a disparate treatment case under Title VII, may be utilized in adjudicating ... discrimination claims arising under sections 1981 and 1983.” Daniels v. Board of Educ., 805 F.2d 203, 207 (6th Cir.1986).
. Section 703(h) reads: "it shall not be an unlawful employment practice for an employer to apply different standards of compensation, or different terms, conditions, or privileges of employment pursuant to a bona fide seniority ... system, ... provided that such differences are not the result of an intention to discriminate because of race.”
. As an alternative, plaintiffs suggest that the district court enact an affirmative action plan designed to assure the promotion of more black officers. This court has repeatedly stated its approval of affirmative action plans. See, e.g., Geier v. Alexander, 801 F.2d 799 (6th Cir.1986); Vanguards of Cleveland v. City of Cleveland, 753 F.2d 479 (6th Cir.1985). We note, however, that this litigation is at an early stage, and the district court has not yet made a final finding of discrimination on the part of defendants. Therefore, a court-enforced affirmative action plan at this point would be premature.
. At the time the district court granted the motion in limine, this court had already established that Ohio’s one-year statute of limitations for torts was applicable to § 1983 claims. See Mulligan v. Hazard, 777 F.2d at 344. We had not yet, however, made any pronouncement with regard to the statute of limitations applicable to § 1981 claims brought in Ohio. Subsequent to the filing of this appeal, the United States Supreme Court held that § 1981 actions are properly characterized as toft actions, and are subject to a state’s statute of limitations for personal injury. Goodman v. Lukens, — U.S. -, 107 S.Ct. 2617, 96 L.Ed.2d 572 (1987). Thus Ohio’s one-year statute of limitations for torts is also applicable to § 1981 actions. See Demery v. City of Youngstown, 818 F.2d 1257 (6th Cir. 1987). Consequently, the district court did not err in holding that the statute of limitations period applicable to both § 1981 and § 1983 actions is one year.