Plaintiffs Silvester Trevino, Jr., Elíseo Carbajal, and Elida Alonzo brought an employment discrimination class action based on Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq., the Civil Rights Act of 1866, 42 U.S.C. § 1981, and section 301 of the Labor Management Relations Act (LMRA), 29 U.S.C. § 185, against their employer Holly Sugar Corporation (Holly) and their union, Local 321 of the American Federation of Grain Millers International Union, AFL-CIO (the Union).1 After denying class certification, the district court conducted a bench trial on plaintiffs’ individual claims. The court denied all of plaintiffs’ claims and entered judgment for defendants. Plaintiffs Trevino and Carbajal2 appeal the district court’s denial of *899class certification, denial of relief in their individual actions, and award of costs to defendants. (No. 85-1502). In a consolidated action, Holly appeals from the district court’s denial of its motion for attorneys’ fees against plaintiffs and their attorneys. (No. 85-1727). We affirm.
Facts and Proceedings Below
Holly is engaged in the business of producing sugar from sugar beets, with refineries in several states. The employees at Holly’s Hereford, Texas refinery are covered by a collective bargaining agreement signed by Holly, the American Federation of Grain Millers International, and its affiliated local unions. Plaintiff Trevino is an employee at Holly’s Hereford plant and is a member of the collective bargaining unit represented by Local 321. Plaintiff Carbajal was employed by Holly and a member of that collective bargaining unit until his discharge in September 1982.
The production of sugar at Holly occurs on a seasonal basis. The production season, called the “campaign,” usually runs from October until January or February at the Hereford facility. During campaign season, when the sugar is refined, the factory operates twenty-four hours a day, seven days a week. During “intercampaign” seasons, Holly repairs its production machinery, performs capital improvement projects, and packages and ships stored sugar. Due to the seasonal nature of the industry, the size and composition of the work force fluctuates over the course of each year.
The collective bargaining agreement divides employees into two main categories— campaign employees and year-round employees. Holly selects its year-round employees at the end of each campaign. Employees on the year-round list receive greater benefits than campaign employees, such as paid vacation leave arid health insurance.3 Paragraph 14.19 of the collective bargaining agreement states: “When hiring employees for year-round employment, the Company agrees to give due consideration to seniority, knowledge, ability, skill, together with the requirements of the job.” Employees are selected for the year-round list based largely on their skills, because Holly selects those who perform jobs needed during the intercampaign or who are crucial to starting the next production period. Of those qualified for each needed position, the most senior employee usually is chosen, even if this means not selecting a more qualified employee with less seniority. In addition, the agreement divides all the jobs into five levels — Technician I, Technician II, Station Group A, Station Group B, and Station Group C — in decreasing order of wages. Employees move into the Technician and Station Group A levels only by bidding for vacancies, which are posted. These posted jobs are awarded “on the basis of seniority, knowledge, ability, skill, together with the requirements of the job.” Agreement 1114.12. Employees holding Technician jobs possess the skills used most during intercampaign and thus are most likely to be selected for the year-round list.4
In April 1983, plaintiffs filed an across-the-board employment discrimination class action against Holly and the Union, alleging discrimination based on race, national *900origin, and/or sex.5 Plaintiffs asserted that Holly and the Union discriminated against the members of the purported class “at every step of the employment process, including recruitment, hiring, job assignment, discipline, promotion, demotion, wages, training, seniority, the allocation of seasonal versus full-time employment, layoff, recall, and termination.” Plaintiffs initially sought class certification for themselves and all past, present, and future Hispanic employees at Holly’s Hereford factory.® By the close of their five-day class certification hearing in April 1984, plaintiffs had narrowed the purported class to Hispanics adversely affected by the year-round employee provision of the collective bargaining agreement at any Holly facility subject to the agreement. At the hearing, plaintiffs also stated that they did not seek class certification against the Union and that they did not present Carbajal as a class representative. On May 11, 1984, the district court denied plaintiffs’ motion for class certification.
In March 1985, the district court conducted a ten-day bench trial on the merits of the individual claims.6 7 Plaintiffs’ main contention was that they had been injured by Holly’s pattern or practice of discrimination against Hispanics. Plaintiffs argued that the disparity between the time it takes Hispanic employees to achieve year-round status as compared to the time for non-Hispanic employees demonstrates Holly's discrimination against Hispanic workers.8 Plaintiffs also asserted that it takes Hispanics longer than non-Hispanics to attain seniority and to rise from lower to higher job levels. Trevino specifically claimed that he was discriminated against in not being placed on the 1982 and 1988 year-round lists. He also averred that his temporary demotion in 1982 from locomotive engineer to yard and factory worker was discriminatory, and that Holly illegally retaliated against him for attempting to enforce rights guaranteed by Title VII. Carbajal claimed that his discharge was discriminatory and that the Union failed to properly pursue his grievance.
The district court found that plaintiffs failed to prove discrimination in each of the discrete occasions asserted, and that plaintiffs had not suffered from a pattern or practice of discrimination. The district court analyzed the statistical evidence presented by each side and concluded that Holly’s statistics presented “a more accurate depiction of reality” than the plaintiffs’ statistics. The court observed that, even if it were to credit plaintiffs’ statistics, it would not benefit the individual claimants in this case. Therefore, the court denied plaintiffs’ relief and awarded costs to defendants. Plaintiffs do not appeal the district court’s determination of their demotion, retaliation, or discharge claims; they do appeal the district court’s finding that they failed to prove that defendants discriminated against them in the application of the year-round provision of the agreement.
In July 1985, Holly filed a motion seeking attorneys’ fees against plaintiffs, their attorneys, Texas Rural Legal Aid, Inc. (TRLA), and the Legal Services Corpora*901tion. Holly appeals the district court’s denial of this motion.
Discussion
Individual Claims
Plaintiffs appeal the district court's finding that defendants have not discriminated against them based on their national origin and race in violation of Title VII, section 1981, and section 301 of the LMRA. Plaintiffs claim to have suffered injury from a pattern or practice by defendants of discriminating against Hispanics in their application of the year-round employment status provision of the collective bargaining agreement.9 The thrust of plaintiffs’ complaint is that Hispanics do not have an equal opportunity to accrue the seniority or job skills that are important factors in selection to the year-round list and in successful bidding for job promotions (which in turn facilitates selection to the year-round list). Plaintiffs argue that it takes Hispanics longer than non-Hispanics to achieve year-round status and thus Hispanics are harmed in not receiving the economic benefits of year-round status on an equal basis.
The nature of Holly’s bidding system and year-round list selection has the effect of multiplying the advantages of seniority. An employee with little or no seniority is greatly disadvantaged, because an employee with any significant seniority has a much better chance to accrue increased seniority. Plaintiffs contend that Hispanics do not have an equal opportunity to accrue seniority and thus are disadvantaged in their attempts to get on the year-round list. Plaintiffs also argue that Hispanics lack seniority because they are not placed on the year-round list in adequate numbers considering their proportion of Holly’s employees. It seems that a significant aspect of plaintiffs’ complaint is their contention that Hispanics are not proportionally represented among those employees recalled for intercampaign work. Campaign employees who are recalled for intercampaign work accrue more seniority for the year and are more likely to learn job skills important to the year-round operation of the plant than campaign employees who work only during the campaign.
The district court carefully analyzed the statistics presented by plaintiffs and by Holly on the issue of a pattern or practice of discrimination. The court found flaws in each presentation, but found more significant deficiencies in plaintiffs’ statistics. Therefore, the court credited defendants’ statistics and concluded that plaintiffs had failed to demonstrate a pattern or practice of national origin (or racial) discrimination by defendants. The court held, alternatively, that even if it had credited plaintiffs’ statistics, they would not establish discrimination against the individual plaintiffs in this case. The evidence demonstrates that the only time Trevino was denied a job for which he bid was when he simultaneously bid for two positions. He received one job, and the other was also awarded to an Hispanic. Carbajal never bid for a job during his employment at Holly. The court concluded that, because the plaintiffs here were not harmed in the job-bidding process, they did not suffer from any discriminatory practices.10
Plaintiffs contend, however, that a failure to exercise in job bidding the seniority *902and skills the plaintiffs had is not relevant to the pattern or practice issue, because the plaintiffs were harmed in their ability to accrue the seniority and skills that would have assisted them in other bids. Furthermore, plaintiffs argue that it took them longer than the average non-Hispanic to achieve year-round status. We believe that plaintiffs’ failure to use the bidding system more effectively is relevant to the pattern or practice issue, because use of the bidding system is an important aspect of an employee’s ability to acquire seniority and skills. Nevertheless, we need not determine whether plaintiffs’ statistics would have benefited the individual plaintiffs, because we hold that the district court’s finding that plaintiffs failed to establish a pattern or practice of discrimination is not clearly erroneous.
We review the district court’s determination that plaintiffs have failed to prove discrimination under the clearly erroneous standard of Fed.R.Civ.P. 52(a). Anderson v. City of Bessemer City, 470 U.S. 564, 105 S.Ct. 1504, 1511, 84 L.Ed.2d 518 (1985); Pullman-Standard v. Swint, 456 U.S. 273, 102 S.Ct. 1781, 72 L.Ed.2d 66 (1982); accord Bazemore v. Friday, — U.S. —, 106 S.Ct. 3000, 3008, 92 L.Ed.2d 315 (1986). As plaintiffs acknowledge, challenges to employment practices that rely on subjective selection criteria must be analyzed under the disparate treatment theory of discrimination, which requires a finding of intentional discrimination. Page v. United States Industries, Inc., 726 F.2d 1038, 1045 (5th Cir.1984); Carroll v. Sears Roebuck & Company, 708 F.2d 183, 188 (5th Cir.1983). Under the disparate treatment theory, the plaintiff may prove the existence of a pattern or practice of discrimination by establishing by direct or circumstantial evidence “that racial discrimination was the company’s standard operating procedure — the regular rather than the unusual practice.” International Brotherhood of Teamsters v. United States, 431 U.S. 324, 97 S.Ct. 1843, 1855, 52 L.Ed.2d 396 (1977). The plaintiff must prove more than the mere occurrence of “isolated” or “sporadic” acts of discrimination. Id. The plaintiff may establish a prima facie case of disparate treatment using statistics alone if the statistics show a “gross disparity” in the treatment of workers based on discriminatory factors. Page, 726 F.2d at 1046. However, if the plaintiffs’ statistical evidence is insufficient alone to establish a prima facie case of disparate treatment, the plaintiff may combine the statistical evidence with “historical, individual, or circumstantial evidence.” Id. (quoting Payne v. Travenol Laboratories, Inc., 673 F.2d 798, 817 (5th Cir.), cert. denied, 459 U.S. 1038, 103 S.Ct. 451, 74 L.Ed.2d 605 (1982)). To rebut a prima facie case of disparate treatment, the defendant must either discredit the plaintiffs’ evidence or provide a nondiscriminatory reason for the apparently discriminatory results. Page, 726 F.2d at 1046; Carroll, 708 F.2d at 190.
Much of plaintiffs’ statistical evidence was devoted to an effort to demonstrate that it takes Hispanics longer than non-Hispanics to move through the promotion system and to be selected for the year-round list. Plaintiffs contend that there is a statistically significant disparity between Hispanics and non-Hispanics for the total elapsed time from the date of hire during the year the employee began working continuous campaigns until an award of year-round status.11 Plaintiffs' expert, Dr. Bernard Yancy, testified that the average non-Hispanic employee was awarded year-round status after oniy 132 days, while the average Hispanic worker was awarded year-round status after 2,017 days. This figure is not compellingly persuasive evidence of discrimination, however, because plaintiffs’ expert failed to take into account the effects of accrued seniority, job skills, or bidding activity. The evidence offered by both parties established that historically *903employees are selected for the year-round list primarily from the Technician II and Technician I job levels. In fact, in the years 1981 to 1983, most of the Technician I employees were on the year-round list, and thus not eligible for the liability period year-round lists. Therefore, plaintiffs attempted to demonstrate that it takes longer for Hispanics to move into the Technician positions than it takes non-Hispanics, and that it takes longer for Hispanic Technician employees to be selected to the year-round list than for non-Hispanic Technician employees. Plaintiff’s expert analyzed movement during the discovery period from the Station Group levels to Technician II and suggested that there was a statistically significant disparity between the time it took Hispanics and the time for non-Hispanics, at least those who moved from Station Groups B and C.,12 However, as plaintiffs’ own expert demonstrated, most of the Hispanics who move into Technician II jobs are promoted from within the factory, while the majority of non-Hispanics who move into Technician II jobs are either new hires or rehires who thus have zero days between date of hire and date of promotion. Therefore, the probative value of these figures would depend upon whether Holly discriminated in hiring from outside directly into the Technician II position. This was not shown. Plaintiffs also attempted to prove that it takes longer for Hispanic Technician employees to be selected to the year-round list than it takes for non-Hispanic Technician employees. The main problem with Dr. Yancy’s figures in this area is that they reflect the total elapsed time from the employee’s date of hire (before continuous campaign employment) to the date of selection to the year-round list and do not reflect any difference between elapsed time in the Technician levels for Hispanics and non-Hispanics selected to the year-round list. As the district court commented: “[Tjhese statistics do not indicate whether any problem is in moving into the Technician II level, or in moving onto the year-round list from the Technician II level.” In addition, these statistics again fail to take into account any differences in accrued seniority between Hispanics and non-Hispanics.
Plaintiffs’ assertions that it takes longer for Hispanics to gain year-round status are undercut also by the failure of plaintiffs’ expert to consider the distribution between Hispanics and non-Hispanics of those job skills that Holly considers in selection to the year-round list. Holly’s expert, Dr. Joan Haworth, analyzed selection to the year-round list among those in “key craft positions,” those with skills usually needed during intercampaign. Her figures indicated that minorities with these skills were added to the year-round list in the same proportion in which these skilled workers appeared among the campaign workers. Her statistics also suggested that minorities not in key craft positions were added to the year-round list in the same proportion in which minorities without these skills appeared in the campaign work force. In addition, Dr. Haworth’s analysis indicated that, considering the additions segregated by job skills, it did not take Hispanics longer to achieve year-round employee status. In fact, she testified that during the 1981 to 1983 liability period it took Hispanics less time than non-Hispanics to get on the year-round list whether added in a key position or another position, although the difference was not statistically significant.
Dr. Haworth's statistics did not account for every variable material to selection for the year-round list. Although these statistics gave some consideration to the employee’s skills, they were not segregated by job levels. Moreover, these figures considered all minorities in the same group. On the other hand, the evidence demonstrates that there were very few non-Hispanic minority employees at Holly. In general, Holly’s analysis presented a more complete picture of the year-round selection process.
*904Plaintiffs do not claim that defendants have discriminated in their application of the bidding process itself,13 which is based primarily on seniority.14 Indeed, Holly’s statistical evidence indicated that during the entire discovery period almost ninety-four percent of the posted jobs were awarded to the most senior bidder of those qualified for the job, and that during the liability period the percentages of jobs going to the most senior bidder were even higher. Moreover, Holly’s evidence demonstrates that minorities on the average were more successful than non-minorities in receiving awards when they bid for jobs.
Plaintiffs contend not that the bidding system is illegal per se, but that Hispanics are discriminated against in their attempts to acquire important job skills and seniority. Plaintiffs’ expert testified that almost half of all job bid awards went to a worker who was paid at the rate of the awarded job during the time before the vacant position was posted or before the job was awarded. Plaintiffs argue that these figures illustrate that many employees are pre-selected through temporary assignments in these positions and trained before the jobs are awarded, and that this pre-selection process adversely affects Hispanics. However, plaintiffs’ expert did not take into account lateral movement between jobs with the same rate of pay and did not present any evidence that Hispanics were treated differently from non-Hispanics in making temporary (or lateral) assignments. Plaintiffs’ expert also testified that during the discovery period there was a disparity between Hispanics and non-Hispanics in the amount of seniority accrued in a given year, and that this disparity was statistically significant for the years 1981 to 1983. Dr. Yancy based this conclusion on the seniority lists provided by Holly. Using the same seniority lists, Dr. Haworth concluded that from 1978 to 1983 on the average Hispanics not on the year-round list acquired slightly more seniority per year than non-Hispanics who were not on the year-round list. Of course, all employees— Hispanic or non-Hispanic — on the year-round list accrued twelve months of seniority each year. It appears that Dr. Yancy’s failure to separate year-round employees from those not on the year-round list accounts for the difference between the experts’ conclusions.15 Because plaintiffs’ disparate treatment claim centers around Holly’s selection of campaign employees for the year-round list, the average yearly accrual of seniority by campaign employees rather than by all employees is more significant to the issue of discrimination.
We believe that the district court did not err in crediting defendants’ statistical evidence. Because plaintiffs’ analyses failed to take into account significant factors other than national origin (or race) that may have contributed to the results, their statistical evidence alone does not compel the conclusion that a pattern or practice of discrimination was demonstrated by a preponderance of all the evidence. See Bazemore, 106 S.Ct. at 3008; Page, 726 F.2d at 1049-51. Of course, plaintiffs *905may supplement their statistical evidence with other evidence of discrimination. Several Hispanic Holly employees, including plaintiffs, testified that they had been discriminated against based on their being Hispanic. Much of the anecdotal evidence involved employees’ claims that they had been given more difficult work assignment because they were Hispanic or that they were retaliated against for filing grievances or EEOC charges. This testimony did not directly relate to plaintiffs’ pattern or practice claim concerning year-round selection. There was also some employee testimony indicating that sometimes some supervisors appeared more likely to train Anglo employees than Hispanic employees, especially on matters such as the general operation of the plant. However, as noted below, this evidence did not tend to show a pervasive practice and was not of a compelling character.
Plaintiffs failed to establish that Holly acted with discriminatory intent in not selecting Trevino to the year-round lists for 1981 and 1982. In each instance, all of those selected with skills similar to Trevino’s had accrued more seniority. Nor did plaintiffs establish that Trevino suffered discrimination in his 1982 demotion. Three witnesses (two of them Hispanic) testified that, when Holly eliminated the four beet-end locomotive engineer positions because it no longer received beets by rail, Trevino was informed in Spanish and English that he would need to bid on a new position. Trevino completed the campaign (at the same pay rate) without bidding on a new position. When he returned for the next campaign, he received a job as yard and factory worker at a lower wage, but subsequently bid successfully for another locomotive engineer position.
Plaintiffs’ witnesses recounted general complaints of national origin discrimination, but offered few substantiated examples of different treatment between Hispanic and non-Hispanic workers.16 Plaintiffs’ individualized comparisons of the work history of one Hispanic employee, such as Trevino, with that of one non-Hispanic employee, such as factory manager Gene Allen, may suggest disparities in Holly’s treatment, but are not significant in light of the extensive evidence concerning the overall workforce. In any event, defendants’ evidence did more than highlight the deficiency in plaintiffs’ analyses, it also adequately rebutted any inferences of discrimination raised by plaintiff’s evidence. See Page, 726 F.2d at 1053. Defendants presented legitimate, non-discriminatory reasons to account for the disparities that plaintiffs attributed to discrimination. Because plaintiffs have failed to demonstrate a gross disparity in the treatment of Hispanics and non-Hispanics, the district court was not required to infer that Holly has discriminated against Hispanics. We therefore conclude that the district court’s finding is not clearly erroneous.
Class Certification
Plaintiffs also appeal the district court’s denial of class certification. Plaintiffs sought class certification for all Hispanics who had been adversely affected by Holly’s application of its year-round selection system. The district court concluded that the commonality and typicality requirements of Fed.R.Civ.P. 23(a)(2) and (3) were not met, because each addition to the year-round list must be considered in light of the employee’s accrued seniority and job skills. The necessity for individualized proof indicates that a class action is not “an economical or efficient way of processing the complaints of the proposed class.” We will not overturn a decision to grant or deny class certification unless the district court has abused its discretion. Redditt v. Mississippi Extended Care Centers, Inc., 718 F.2d 1381, 1387 (5th Cir.1983).
Having affirmed the district court’s denial of relief on Trevino’s individual claim, and having granted Alonzo’s mo*906tion to withdraw, see note 2, supra, we therefore affirm the district court’s denial of class certification. When the district court has denied class certification, our decision on review of whether the named plaintiffs should represent a class “is appropriately made on the full record, including the facts developed at the trial of the plaintiffs’ individual claims.” Everitt v. City of Marshall, 703 F.2d 207 (5th Cir.) (quoting East Texas Motor Freight System, Inc. v. Rodriguez, 431 U.S. 395, 97 S.Ct. 1891, 1898 n. 12, 52 L.Ed.2d 453 (1977)), cert. denied, 464 U.S. 894, 104 S.Ct. 241, 78 L.Ed.2d 231 (1983). When an employee’s individual claim of employment discrimination is properly dismissed, the employee no longer has a nexus with the membership of the purported class. Id. To represent a class, the employee must “possess the same interests and suffer the same injury” as the purported class members. Rodriguez, 97 S.Ct. at 1896. We therefore affirm the district court’s denial of class certification, because the proposed class representative may not now claim to have suffered from the injuries asserted by the proposed class.17
Defendants’ Costs
Plaintiffs also challenge the district court’s award of costs to defendants Holly and the Union pursuant to Fed.R.Civ.P. 54(d). Rule 54(d) provides that: “Costs shall be allowed as of course to the prevailing party unless the court otherwise directs.” Plaintiffs contend that the standard announced by the Supreme Court in EEOC v. Christiansburg Garment Company, 434 U.S. 412, 98 S.Ct. 694, 54 L.Ed.2d 648 (1978), that attorneys’ fees should be awarded a prevailing defendant in a Title VII case only if the plaintiffs’ action was frivolous, should be extended to an award of costs. However, we have rejected this argument in previous decisions. Lewis v. NLRB, 750 F.2d 1266, 1279 (5th Cir.1985); Hill v. J.C. Penney Co., 688 F.2d 370, 375 (5th Cir.1982). We thus affirm the district court’s award of costs to defendants.18
Attorneys' Fees
Holly appeals the district court’s denial of its motion for attorneys’ fees. Holly sought attorneys’ fees from plaintiffs and their attorneys, TRLA, and attempted to join the Legal Services Corporation (LSC). The district court did not expressly rule on Holly’s motion to join the LSC, but denied Holly’s motion for attorneys’ fees. Holly bases its attorneys’ fees claim on the following grounds: 42 U.S.C. § 1988, 42 U.S.C. § 20000e-5(k) (section 706(k) of Title VII), 28 U.S.C. § 1927, Fed.R.Civ.P. 11, and 42 U.S.C. § 2996e(f). Finding that none of these grounds requires us to reverse the denial of attorneys’ fees in this case, we affirm the district court’s ruling.
Prevailing defendants are entitled to attorneys’ fees under section 706(k) of Title VII or under section 1988 only upon a showing that the plaintiff’s action was frivolous, unreasonable, groundless, or made in bad faith, or that the plaintiff continued to litigate after it clearly became so. Christiansburg Garment, 98 S.Ct. at 700. Holly argues that the district court abused *907its discretion in denying it attorneys’ fees under this standard and in failing to evaluate its motion on a claim-by-claim basis. Holly contends under Hensley v. Eckerhart, 461 U.S. 424, 103 S.Ct. 1933, 76 L.Ed.2d 40 (1983), and Commonwealth Oil Refining Co. v. EEOC, 720 F.2d 1383 (5th Cir.1983), that the district court should have conducted a claim-by-claim analysis to determine which claims were frivolous.
In Uviedo v. Steves Sash & Door Co., 753 F.2d 369 (5th Cir.l985), cert. denied, — U.S. —, 106 S.Ct. 791, 88 L.Ed.2d 769 (1986), this Court held that
“a plaintiff’s claims which are unrelated to each other, though formally and properly joined in a single suit against a single defendant, are nevertheless to be considered for attorneys’ fees purposes as if each were brought in a separate suit, so that a party’s success or failure on one claim is not to be considered in determining that party’s entitlement (or vulnerability) to attorneys’ fees on an unrelated claim.” Id. at 370-71.
We noted that the Commonwealth “central issue” test may be applied to a suit as a whole when all the claims are related to each other, id. at 370, although there is “no certain method” for determining which claims are related, id. at 371.
Holly relies primarily on plaintiffs’ pleadings and statements in the pre-trial order alleging employment discrimination in several specific aspects of their employment at Holly in arguing that plaintiffs had unrelated claims. Plaintiffs contend that all the claims pursued at trial revolve around discrimination in selection for year-round status, and that they narrowed their claims from those originally pleaded. Nevertheless, it seems that plaintiffs pursued some claims — such as Carbajal’s discharge claim and Trevino’s Title VII retaliation claim — that are unrelated to the claims surrounding year-round employee status. We believe, however, that in the circumstances of this case, Holly should not be allowed to successfully challenge the district court’s failure to perform a claim-by-claim analysis, because Holly did not request such findings in its motion or accompanying briefs below, nor did Holly make any attempt to segregate any of plaintiffs’ claims for the district court in computing its requested fees, and instead pursued its entitlement to attorneys’ fees on the basis of the case as a whole.19
Section 1927, 28 U.S.C., provides that an attorney who “multiplies the proceedings in any case unreasonably and vexatiously” may be liable for excess attorneys’ fees resulting from such conduct. The district court found that Texas Rural *908Legal Aid did not unreasonably or vexatiously multiply the proceedings. We are unable to say that the district court abused its discretion. See Jackson Marine Corp. v. Harvey Barge Repairs, Inc., 794 F.2d 989, 992 (5th Cir.1986). See also Nilsen v. City of Moss Point, 621 F.2d 117, 122 (5th Cir.1980).
Rule 11, Fed.R.Civ.P., provides for sanctions against an attorney who signs a pleading, motion, or other paper that to the best of his knowledge after reasonable inquiry is not “well grounded in fact and ... warranted by existing law or a good faith argument for the extension, modification or reversal of existing law.” The district court declined to impose sanctions based on the pleadings, and held that Holly’s request for sanctions based on the filing of motions was not timely. See Rule 11, Notes of Advisory Committee (timing is within district court’s discretion although it is anticipated that sanctions be determined “when the motion is decided or shortly thereafter”). We conclude that the district court did not abuse its discretion in determining that Holly failed to preserve its challenge to plaintiffs’ motions and that Holly did not establish that attorneys’ fees should be imposed based on plaintiffs’ pleadings. Holly claims that it clearly challenged all of plaintiffs’ claims as frivolous since the beginning of the suit; yet, the district court did not abuse its discretion in determining that this general challenge was not adequate to preserve claims as to each motion separately considered.
Finally, defendants rely on the Legal Services Corporation Act, 42 U.S.C. § 2996e(f), which provides for attorneys’ fees against the LSC when one of its grant recipients (in this case TRLA) pursues an action “for the sole purpose of harassment of the defendant” or when “a recipient’s plaintiff maliciously abuse[s] legal process.” Again, we hold that the district court did not abuse its discretion in determining that plaintiffs’ conduct did not meet the standard required for an award based on this provision. See Jackson Marine Corp., 794 F.2d at 992.
In sum, although plaintiffs initially pleaded a wide range of violations, they eventually narrowed their focus to a few claims that clearly were not frivolous. For the reasons stated, Holly is not entitled to attorneys’ fees under the Christiansburg Garment standard. Nor is it so entitled under the statutory and rule standards which it asserts. We cannot say that the district court abused its discretion. Thus, we will not overturn the district court’s denial of attorneys’ fees. Jackson Marine Corp., 794 F.2d at 992; In re Hunt, 754 F.2d 1290, 1294 (5th Cir.1984).
Conclusion
Determining that the district court was not clearly erroneous in finding that plaintiffs failed to establish a pattern of discrimination, we affirm the district court’s denial of relief to plaintiffs Trevino and Carbajal and the denial of class certification. Furthermore, we find that the district court did not abuse its discretion in awarding costs to defendants or in denying Holly’s motion for attorneys’ fees. Accordingly, the judgment is affirmed.
AFFIRMED.
. Plaintiffs also sued the American Federation of Grain Millers International Union, AFL-CIO, and its Locals 279, 280, and 285 as defendants. Prior to trial, the parties stipulated that Local Unions 279, 280, and 285 would become necessary party defendants only pursuant to Fed.R.Civ.P. 19(a), and that plaintiffs would seek no economic relief from these locals. Local 321 and the International Union (collectively referred to as the Union) remain as primary defendants.
. Having settled her claims against defendants, plaintiff Alonzo sought withdrawal from appeal No. 85-1502. We granted Alonzo’s unopposed motion to withdraw from appeal No. 85-1502, but our action was without prejudice to the rights of Trevino and Carbajal in appeal No. *89985-1502 or to Holly’s rights against Alonzo and the other appellees in appeal No. 85-1727.
Carbajal did not seek to be a class representative and brought only an individual claim. Thus, of the appellants before us, only Trevino sought class certification.
. In addition, campaign employees who are not absent from the payroll for more than forty days during a year are "vacation-rated" campaign employees, and thus receive some of the same benefits given to year-round employees.
. The highest proportion of those selected for the year-round list in 1978 through 1983 came from the Technician II position. In those years, most employees in Technician I positions were already on the year-round list. Station Group A provided the next largest group for year-round *900selection, followed by Technician I, and then the B and C Station Groups.
. The district court dismissed plaintiff Alonzo's sex discrimination claim as untimely. Plaintiffs do not appeal that determination in this action.
. The district court denied plaintiffs’ request for discovery of information regarding the other Holly facilities, including its corporate headquarters. However, plaintiffs were permitted to depose corporate executives concerning decisions made at corporate headquarters affecting employees at the Hereford facility.
. On plaintiffs’ Title VII actions, defendants were subject to claims of discrimination that occurred after January 1, 1982, which was 180 days before Trevino filed a charge of discrimination with the EEOC. The liability period for the section 1981 actions commenced on April 13, 1981, two years before this suit was filed on April 13, 1983. Much of the testimony, however, focuses on the five-year discovery period from July 1978 to June 1983.
. Trevino was selected for the year-round list in 1984 after 2,667 days of employment with Holly; in 1979, Carbajal was selected for the year-round list after 332 days at Holly.
. Plaintiffs initially sought in the district court to prove discrimination under both disparate treatment and disparate impact theories, arguing that the agreement as written and as applied was discriminatory. On appeal, plaintiffs assert that defendants’ application of the agreement has resulted in disparate treatment between Hispanics and non-Hispanics.
National origin discrimination is expressly forbidden by Title VII. 42 U.S.C. § 2000e-2. We assume, arguendo, that the racial basis of section 1981 would be satisfied here by a showing of intentional ethnic discrimination against Hispanics. See Vasquez v. McAllen Bag & Supply Co., 660 F.2d 686, 687 (5th Cir.1981), cert. denied, 458 U.S. 1122, 102 S.Ct. 3509, 73 L.Ed.2d 1384 (1982); Bullard v. Omi Georgia, Inc., 640 F.2d 632, 634 (5th Cir.1981); Olivares v. Martin, 555 F.2d 1192, 1196 (5th Cir.1977); Alvarado v. El Paso ISD, 445 F.2d 1011 (5th Cir.1971).
. The district court commented: "Holly cannot be faulted for failing to promote the plaintiffs to jobs they never bid for, particularly since the job bid system is required by the Master Agreement between Holly and plaintiffs’ own union.”
. The parties used a probability of random occurrence of .05 as a benchmark of statistical significance, that is, a disparity that is likely to occur at random at or less than five times out of one hundred is considered significant for the purpose of inferring that the disparity did not occur randomly.
. Dr. Yancy did not break down his findings to show how this movement affected Hispanics during the liability period only.
. Openings in Station Group A, Technician II, and Technician I positions are posted. After employees submit bids for the jobs, a committee comprised of union and company representatives makes the selections from among those who bid. Holly does not consider hiring from outside the plant unless there is no qualified bidder for a particular position.
. Based on plaintiffs’ concession, the district court found the bidding system to be a bona fide seniority system pursuant to section 703(h) of Title VII, 42 U.S.C. § 2000e-2(h). The court also concluded that to the extent that year-round list status goes to the most senior skilled employee in bidding system job classifications, it is also immunized by section 703(h).
. Dr. Yancy’s inclusion of year-round employees is reflected by the higher averages he produced. He testified that Anglos accrue between 6.7 and 7.9 months per year on the average, while the average Hispanic employee accrues between 4.2 and 6.6 months per year. Dr. Ha-worth stated that from 1978 to 1983 Hispanics not on the year-round list accrued a yearly average of 4.35, while non-Hispanics accrued an average of 4.14 months of seniority. Her statistics also demonstrate that from 1981 to 1983 Hispanics not on the year-round list accrued 4.65 months of seniority per year, while non-Hispanics had a yearly average of 4.51 months of seniority.
. We also note that the district court's rejection of the factual merits of plaintiffs’ claims of anti-Hispanic discrimination in all the specific asserted instances of demotion, discharge, and retaliation is not challenged.
. We do not, of course, suggest that the district court may not grant class certification without first ruling that the class representatives have established their respective individual cases on the merits.
. Plaintiffs also argue that our panel opinion in International Woodworkers v. Champion International Corp., 752 F.2d 163 (5th Cir.1985), on rehearing en banc, 790 F.2d 1174 (5th Cir. 1986), cert. granted, — U.S. —, 107 S.Ct. 568, 93 L.Ed.2d 573 (1986), supports by analogy their argument that costs should be awarded only upon a finding that the suit is frivolous. However, we voted to rehear the case en banc, thereby vacating the panel opinion, and issued a decision en banc. We held that costs for excess expert witness fees above the amount specified by 28 U.S.C. § 1821 may be awarded only when expressly authorized by Congress or when one of the equitable exceptions to the American rule applies. 790 F.2d 1174. Because plaintiffs do not suggest that defendants were awarded any excess costs or explain how the costs here were beyond the scope of Rule 54(d), our decision in International Woodworkers does not support plaintiffs’ contention that costs should not be awarded.
. In Holly’s Memorandum in Support of its Motion for Attorneys’ Fees, Holly argued below that plaintiffs made an unreasonably wide and unsupported list of allegations, but Holly treated all these claims collectively as frivolous and did not apply the central issue approach or segregate groups of unrelated claims. Instead, Holly contended: ”[P]laintiffs’ claims were not simply unreasonable and groundless when filed, but also pursued in bad faith for more than two years after receiving uncontrovertible evidence of their lack of merit. In its defense, Holly has incurred attorneys’ fees in the amount of $690,-155.93 and costs in the amount of $12,193.29.” Similarly, Holly averred: "Plaintiffs had no legal right to compel Holly to expend the literally thousands of hours of attorney time and the literally hundreds of thousands of dollars required to defend a case that was demonstrably frivilous [sic ] from its inception.” In its discussion of the amount of fees that should be awarded, Holly focused on the relevant factors set out in Johnson v. Georgia Highway Express, Inc., 488 F.2d 714 (5th Cir. 1974), but again did not suggest that the determination should be segregated by claims.
In its reply brief in support of its Motion for Attorneys’ Fees, Holly did make one vague reference to the fact that plaintiffs’ brief states that the district court should conduct a specific factual inquiry with regard to each unrelated claim. Holly did not challenge plaintiffs’ assertion, nor offer its agreement, nor did Holly suggest how the rule would be applied in this case. Throughout the reply brief, Holly repeated its claim that all of plaintiffs’ claims were frivolous and brought in bad faith.
On appeal, Holly again maintains that all of plaintiffs’ claims were frivolous and does not break down these claims into groups of "unrelated claims.” Instead, Holly suggests that the district court was required to sua sponte separately examine each of plaintiffs' initial allegations of discrimination (e.g., hiring, promotion, firing, etc.), although several of these allegations are clearly related.