572 231 federal reporter, 3d series · 2018. 1. 12. · 574 231 federal reporter, 3d series sions...

572 231 FEDERAL REPORTER, 3d SERIES

such actions against a state. Id. at 604, 80S.Ct. 924. The Supreme Court reversedthe dismissal and remanded because, byvirtue of the new law, the district courtnow had jurisdiction. Id.

We followed suit in In Re ArrowheadEstates Dev. Co., 42 F.3d 1306 (9th Cir.1994). In that case, the appellant filed anotice of appeal from a bankruptcy court’soral judgment before the final judgmentwas entered on the record. Id. at 1308.At the time, a notice of appeal filed beforea judgment was entered was invalid, andthe Bankruptcy Appellate Panel (‘‘BAP’’)therefore dismissed the appeal for lack ofjurisdiction. Id. at 1310. However, therules governing notices of appeal changedwhile the case was pending before ourcourt, so that a notice of appeal filed afterthe announcement of a decision but beforethe entry of judgment was treated as filedafter the date of entry of judgment. Id.We held that the new rule applied retroac-tively, and therefore the BAP did havejurisdiction over the appeal. Id. at 1311.

[6–8] These cases persuade us that thedistrict court has jurisdiction. Under Dul-dulao, AEDPA § 440(a) applies retroac-tively. Thus, regardless of the jurisdic-tional rules that were in existence whenthe habeas petition was filed, we apply thejurisdictional rules that exist now. UnderLandgraf, Andrus, Alabama and Arrow-head, we apply new jurisdictional rules toretroactively grant jurisdiction even whenit was lacking at the time the case wasfiled. Since the new jurisdictional ruleshave eliminated the requirement thataliens such as Nakaranurack file for directreview of a BIA decision before filing ahabeas petition, we hold that the jurisdic-tional bar identified in Nakaranurack I is‘‘now of no moment.’’ Landgraf, 511 U.S.at 274, 114 S.Ct. 1483 (quoting Andrus,436 U.S. at 607–608 n. 6, 98 S.Ct. 2002).3

We therefore reverse the district court’sdismissal of this action. We hold that thedistrict court has jurisdiction over themerits of Nakaranurack’s habeas petition,and we remand for a hearing on the mer-its.

REVERSED and REMANDED

,

The ASSOCIATION OF MEXICAN–AMERICAN EDUCATORS(‘‘AMAE’’); California Associationfor Asian–Pacific Bilingual Education(‘‘CAFABE’’), on behalf of them-selves, their members, and all otherssimilarly situated; Oakland Allianceof Black Educators (‘‘OABE’’), on be-half of themselves, their members,and all others similarly situated; SaraMacNeil Boyd; Sam Genis; TouaYang; Bob Williams; Marta Le-Claire; Antoinette Williams; DianaKwan; and Agnes Haynes, on behalfof themselves and all others similarlysituated, Plaintiffs–Appellants/Cross–Appellees,

v.

STATE OF CALIFORNIA; The Califor-nia Commission on Teacher Creden-tialing, Defendants–Appellees/Cross–Appellants.

Nos. 96–17131, 97–15422.

United States Court of Appeals,Ninth Circuit.

Argued and Submitted June 20, 2000

Filed Oct. 30, 2000

Minority educators brought class ac-tion lawsuit against State of California and

3. In a parallel context, we have recognizedthat a petitioner need not exhaust administra-tive remedies when to do so would be futile.See, e.g., El Rescate Legal Serv., Inc. v. Execu-

tive Office of Immigration Review, 959 F.2d742, 747 (9th Cir.1991); Aleknagik NativesLtd. v. Andrus, 648 F.2d 496, 499 (9th Cir.1980).

573ASSOC. OF MEXICAN–AMERICAN v. STATE OF CALIFORNIACite as 231 F.3d 572 (9th Cir. 2000)

the California Commission on TeacherCredentialing (CTC), challenging use ofCalifornia Basic Educational Skills Test(CBEST) as requirement for certificationto teach and for other employment in Cali-fornia public schools as violation of TitleVI and VII of the Civil Rights Act of 1964.On summary judgment, the United StatesDistrict Court for the Northern District ofCalifornia, William H. Orrick, Jr., J., 836F.Supp. 1534, held that Title VI and TitleVII apply to California’s credentialing ofpublic school teachers, but, after trial,found no violation, 937 F.Supp. 1397, anddenied defendants’ application for costs.Plaintiffs appealed and defendants cross-appealed. The Court of Appeals, Graber,Circuit Judge, held that: (1) Title VII ap-plied to CBEST; (2) district court’s findingthat the CBEST was properly validatedwas not clearly erroneous; (3) there was noclear error in finding that setting of pass-ing score for writing test was consistentwith the Equal Employment OpportunityCommission’s (EEOC’s) Guidelines; (4) inabsence of any evidence even suggestingan impropriety on the part of the districtcourt in interaction with court-appointedtechnical adviser, Court of Appeals wouldassume that the district court did its jobproperly; (5) overruling National Info.Servs., Inc. v. TRW, Inc., punishing mis-conduct by the prevailing party is not theonly proper reason for denying costs to aprevailing party; and (6) there was noabuse of discretion in denial of costs.

Affirmed.

Opinion, 195 F.3d 465, withdrawn.

Reinhardt, Circuit Judge, with whomSchroeder and Thomas, Circuit Judgesjoined, filed an opinion concurring in partand dissenting in part.

Fernandez, Circuit Judge, with whomRymer, Circuit Judge joined, filed an opin-ion concurring in part and dissenting inpart.

Kleinfeld, Circuit Judge, with whomO’Scannlain, Circuit Judge, joined, filed an

opinion concurring in part and dissentingin part.

Gould, Circuit Judge, filed and opinionconcurring in part and dissenting in part.

Tashima, Circuit Judge, filed a dis-senting opinion.

1. Federal Courts O13.30Though Court of Appeals ultimately

held that California Basic Education SkillsTest (CBEST) for teaching credentials wasvalidated properly and did not violate TitleVII, it would not decline to decide whetherone of those titles applied, where vali-dation would not be relevant if neitherTitle VI nor Title VII applied, the statehad proceeded for years on the assumptionthat those laws applied to its administra-tion of the CBEST and had expended con-siderable effort and expense in attemptingto comply with federal law in this area, andas a matter of judicial economy, Court’sanswer to the statutory question couldavoid future litigation by other parties.Civil Rights Act of 1964, §§ 601 et seq.,701 et seq., as amended, 42 U.S.C.A.§§ 2000d et seq., 2000e et seq.

2. Civil Rights O143Although there must be some connec-

tion with an employment relationship forTitle VII protections to apply, a directemployment relationship is not a prerequi-site to Title VII liability. Civil Rights Actof 1964, § 701 et seq., as amended, 42U.S.C.A. § 2000e et seq.

3. Civil Rights O146Title VII applied to use of the Califor-

nia Basic Education Skills Test (CBEST)for public school employment credentials,though parties challenging it and stateagency administering it did not have adirect employment relationship and, in-stead, plaintiffs were employees and poten-tial employees of individual school dis-tricts, given plenary state control of localschool districts in California, such that theState is in a theoretical and practical posi-tion to interfere with the employment deci-


sions of local school districts, and the statehas ‘‘interfered’’ to a degree sufficient tobring it within the reach of Title VII byrequiring, formulating, and administeringthe CBEST. Civil Rights Act of 1964,§§ 701 et seq., 703(a)(1), as amended, 42U.S.C.A. §§ 2000e et seq., 2000e–2(a)(1).

4. Civil Rights O143An entity that is not the direct em-

ployer of a Title VII plaintiff neverthelessmay be liable if it interferes with an indi-vidual’s employment opportunities with an-other employer. Civil Rights Act of 1964,§ 703(a)(1), (b, c), as amended, 42 U.S.C.A.§ 2000e–2(a)(1), (b, c).

5. Civil Rights O150Application of Title VII to the Califor-

nia Basic Education Skills Test (CBEST)for public school employment credentialswas not precluded on grounds that CBESTis merely a licensing examination and is anexercise of the state’s police power, as theCBEST is but one aspect of pervasivestate control of local public schools andapplies only to public school employees,and is an exercise of the state’s proprie-tary power as well as its police power.Civil Rights Act of 1964, § 701 et seq., asamended, 42 U.S.C.A. § 2000e et seq.

6. Civil Rights O146, 150There is no overarching ‘‘licensing’’

exception to Title VII, but Title VII doesnot apply when the only connection amongthe licensing agency, the plaintiff, and theuniverse of prospective employers is theagency’s implementation of a general li-censing examination. Civil Rights Act of1964, § 701 et seq., as amended, 42U.S.C.A. § 2000e et seq.

7. Civil Rights O150Discriminatory tests are impermissi-

ble under Title VII unless shown, by pro-fessionally acceptable methods, to be pre-dictive of or significantly correlated withimportant elements of work behaviorwhich comprise or are relevant to the jobor jobs for which candidates are beingevaluated. Civil Rights Act of 1964, § 701

et seq., as amended, 42 U.S.C.A. § 2000eet seq.

8. Civil Rights O378In evaluating employment tests that

are alleged to have a racially disparateimpact, Court of Appeals first considerswhether the plaintiff has established a pri-ma facie case by demonstrating that thetest causes a disparate impact on the basisof race, and where employees or potentialemployees have established a prima faciecase, the burden shifts to parties adminis-tering the test to demonstrate that the testwas validated properly, and if it is estab-lished that the test is validated properly,the burden shifts back to the employeesand potential employees to show the exis-tence of other selection devices that alsowould serve the employer’s legitimate in-terest in efficient and trustworthy work-manship, but that are not discriminatory.Civil Rights Act of 1964, § 701 et seq., asamended, 42 U.S.C.A. § 2000e et seq.

9. Civil Rights O127.1, 150District court’s finding that the Cali-

fornia Basic Education Skills Test(CBEST) for public school employmentcredentials was validated properly, basedon three studies, and thus did not violateTitle VII, was not clearly erroneous, andthe same ruling would apply to Title VI, ifapplicable, in a disparate impact case.Civil Rights Act of 1964, §§ 601 et seq.,701 et seq., as amended, 42 U.S.C.A.§§ 2000d et seq., 2000e et seq.

10. Civil Rights O389 Federal Courts O858

Whether an employment test has beenvalidated properly, and thus does not vio-late Title VII, is primarily a factual ques-tion, and the district court’s finding there-on is reviewed for clear error. CivilRights Act of 1964, § 701 et seq., asamended, 42 U.S.C.A. § 2000e et seq.

11. Civil Rights O150To show that an employment test is

validated properly, parties administering it


are required to show that it has a manifestrelationship to the employment in ques-tion, and when a scored test is challenged,Court of Appeals requires that the test be‘‘job related,’’ that is, that it actually mea-sures skills, knowledge, or ability requiredfor successful performance of the job.Civil Rights Act of 1964, § 701 et seq., asamended, 42 U.S.C.A. § 2000e et seq.

See publication Words and Phras-es for other judicial constructionsand definitions.

12. Civil Rights O150In making a determination about job-

relatedness of an employment test, Courtof Appeals follows a three-step approach:(1) party administering the test must firstspecify the particular trait or characteris-tic which the selection device is being usedto identify or measure; (2) that party mustthen determine that the particular trait orcharacteristic is an important element ofwork behavior; and (3) that party mustdemonstrate by professionally acceptablemethods that the selection device is pre-dictive of or significantly correlated withthe element of work behavior identified inthe second step. Civil Rights Act of 1964,§ 701 et seq., as amended, 42 U.S.C.A.§ 2000e et seq.

13. Civil Rights O150The Equal Employment Opportunity

Commission’s (EEOC’s) Uniform Guide-lines on Employee Selection Proceduresare relevant to the inquiry as to whetheran employment test has been properlyvalidated, so as to not violate Title VII;failure to comply with the Guidelines, al-though not automatically fatal to an em-ployment test, diminishes the probativevalue of the defendants’ validation study.29 C.F.R. §§ 1607.1 et seq., 1607.5, 1607.5,subd. B, 1607.14.

14. Civil Rights O150The district court did not clearly err

in finding that a review relied on for vali-dation of the California Basic EducationSkills Test (CBEST) for public school em-ployment credentials under Title VII was

an appropriate form of a job analysis un-der the professional standards of the time,where the review comprised the pooledjudgments of knowledgeable persons aboutthe relevance of the skills which CBESTtested to the jobs for which it is required,and the reviewers consisted of 234 teach-ers, administrators, and other publicschool employees, 36 percent of whomwere members of minority groups. CivilRights Act of 1964, § 701 et seq., asamended, 42 U.S.C.A. § 2000e et seq.

15. Civil Rights O150Study relied on for validation of the

California Basic Education Skills Test(CBEST) for public school employmentcredentials under Title VII did not violatethe requirement that only ‘‘important’’work skills be measured, though the studyretained activities and skills if they re-ceived a mean importance rating of 1.5 ona scale that designated 2 as ‘‘important’’and 1 as ‘‘minor,’’ where the mean ratingof 1.5 was coupled with a criterion of 80percent endorsement by educators polled.Civil Rights Act of 1964, § 701 et seq., asamended, 42 U.S.C.A. § 2000e et seq.


in finding the job analysis supporting vali-dation of the California Basic EducationSkills Test (CBEST) for public school em-ployment credentials to be sufficiently spe-cific and particularized to determine thatthe CBEST was job related for both theteaching and nonteaching jobs for whichthe test was required, so as to satisfy TitleVII. Civil Rights Act of 1964, §§ 701 etseq., 703(k)(1)(A)(i), as amended, 42U.S.C.A. §§ 2000e et seq., 2000e–2(k)(1)(A)(i).


in concluding that the questions on theCalifornia Basic Education Skills Test(CBEST) for public school employmentcredentials had been shown by profession-ally acceptable methods to be predictive of


or significantly correlated with the elementof work behavior that they were designedto measure, so as to be properly validatedunder Title VII. Civil Rights Act of 1964,§ 701 et seq., as amended, 42 U.S.C.A.§ 2000e et seq.

18. Civil Rights O150An employer is not required by Title

VII to validate separately the selection ofparticular passing scores on an employ-ment test. Civil Rights Act of 1964, § 701et seq., as amended, 42 U.S.C.A. § 2000eet seq.; 29 C.F.R. § 1607.5, subd. H.

19. Civil Rights O150The California Superintendent of Pub-

lic Instruction was not required by TitleVII to set the passing score for the writingportion of the California Basic EducationSkills Test (CBEST) for public school em-ployment credentials at the lowest levelthat a majority of the readers consideredto be ‘‘passing’’; rather, he was required toset a cutoff that was logical, reasonable,and consistent with the data before him,and there was no clear error in findingthat setting passing score at a level that allthe readers agreed was ‘‘passing,’’ and toset an absolute minimum at a level that 80percent of the readers thought was ‘‘pass-ing,’’ was consistent with the Equal Em-ployment Opportunity Commission’s(EEOC’s) Guidelines. Civil Rights Act of1964, § 701 et seq., as amended, 42U.S.C.A. § 2000e et seq.; 29 C.F.R.§ 1607.5, subd. H.

20. Federal Civil Procedure O1951In those rare cases in which outside

technical expertise would be helpful to adistrict court, the court may appoint atechnical advisor.

21. Federal Civil Procedure O1951Rule of Evidence requiring that ex-

pert witness appointed by the court advisethe parties of findings and be subject tocross-examination did not apply to court-appointed technical advisor who was notcalled to testify, and the district court didnot err in refusing to allow cross-examina-

tion or to require an expert’s report. Fed.Rules Evid.Rule 706(a), 28 U.S.C.A.

22. Federal Courts O791In absence of any evidence even sug-

gesting an impropriety on the part of thedistrict court in interaction with court-ap-pointed technical adviser, Court of Appealswould assume that the district court did itsjob properly.

23. Federal Civil Procedure O2723,2727, 2742.1

Rule creates a presumption in favor ofawarding costs to a prevailing party, butvests in the district court discretion torefuse to award costs, but that discretionis not unlimited, and a district court mustspecify reasons for its refusal to awardcosts. Fed.Rules Civ.Proc.Rule 54(d)(1),28 U.S.C.A.

24. Federal Courts O830On appeal, Court of Appeals deter-

mines whether the reasons that the districtcourt has specified for denial of costs areappropriate and whether, consideringthose reasons, the court abused its discre-tion in denying costs. Fed.Rules Civ.Proc.Rule 54(d)(1), 28 U.S.C.A.

25. Federal Civil Procedure O2727Punishing misconduct by the prevail-

ing party is not the only proper reason fordenying costs to a prevailing party; over-ruling National Info. Servs., Inc. v. TRW,Inc., 51 F.3d 1470; abrogating Zenith Ins.Co. v. Breslaw, 108 F.3d 205. Fed.RulesCiv.Proc.Rule 54(d)(1), 28 U.S.C.A.

26. Civil Rights O291, 409The district court did not abuse its

discretion in refusing to award costs to theprevailing defendants, the State of Califor-nia and state agency, in civil rights actionchallenging the California Basic EducationSkills Test (CBEST) for public school em-ployment credentials, as this was an ex-traordinary, and extraordinarily important,case, plaintiffs were a group of individualsand nonprofit organizations with limitedresources, action presented issues of the


gravest public importance affecting tens ofthousands of Californians and the state’spublic school system as a whole, the issueswere close and complex, plaintiffs claimswere not without merit and defendantssubstantially altered the CBEST duringthe pendency of this litigation, and costswere extraordinarily high. Fed.Rules Civ.Proc.Rule 54(d)(1), 28 U.S.C.A.

John T. Affeldt, Public Advocates, Inc.,San Francisco, California, for the plain-tiffs-appellants/appellees.

R. Lawrence Ashe, Jr., and Eric J. Tay-lor, Paul, Hastings, Janofsky & Walker,LLP, Atlanta, Georgia; and William Lock-yer, Attorney General, and StephanieWald, Deputy Attorney General, SanFrancisco, California, for the defendants-appellees/appellants.

Stewart Weinberg, Van Bourg, Wein-berg, Roger & Rosenfeld, Oakland, Cali-fornia; Paul J. Dostart, Dostart ClappSterrett & Coveney, LLP, San Diego, Cali-fornia; Martin Bienstock, Assistant Attor-ney General, New York, New York; PeterRoos, San Francisco, California; and Greg-ory B. Friel and Jennifer Levin, UnitedStates Department of Justice, Washington,D.C., for the amici.

Appeals from the United States DistrictCourt for the Northern District of Califor-nia William H. Orrick, Jr., District Judge,Presiding. D.C. No. CV–92–03874–WHO

Before: HUG, Chief Judge, andSCHROEDER, REINHARDT,O’SCANNLAIN, FERNANDEZ,RYMER, KLEINFELD, TASHIMA,THOMAS, GRABER, and GOULD,Circuit Judges.

Opinion by Judge GRABER; PartialConcurrence and Partial Dissent by JudgeREINHARDT; Partial Concurrence andPartial Dissent by Judge FERNANDEZ;Partial Concurrence and Partial Dissent

by Judge KLEINFELD; PartialConcurrence and Partial Dissent by JudgeGOULD; Dissent by Judge TASHIMA.

GRABER, Circuit Judge:

Plaintiffs are a class of Mexican–Ameri-can, Asian–American, and African–Ameri-can educators and would-be educators inCalifornia. They appeal from an adversejudgment in their action against the Stateof California and its agency, challenging(1) the district court’s holding that theCalifornia Basic Education Skills Test(‘‘CBEST’’), which is a prerequisite to em-ployment in a variety of positions in theCalifornia public schools, violates neitherTitle VI nor Title VII of the Civil RightsAct of 1964, and (2) the district court’s useof a technical advisor at trial. Defendantscross-appeal from the district court’s rul-ings, on summary judgment, that Title VIand Title VII apply in this case. Defen-dants also appeal from the district court’sorder denying their request for costs. Forthe reasons that follow, we hold that TitleVII applies to the CBEST; that theCBEST was validated properly; that thedistrict court permissibly used a technicaladvisor; and that the district court did notabuse its discretion by refusing to awardcosts to Defendants. Accordingly, we af-firm both the judgment in Defendants’ fa-vor and the order denying them costs.

FACTUAL AND PROCEDURALBACKGROUND

Effective February 1, 1983, the Califor-nia legislature amended the California Ed-ucation Code to prohibit the CaliforniaCommission on Teacher Credentialing(‘‘CCTC’’) from issuing ‘‘any credential,permit, certificate, or renewal of an emer-gency credential to any person to serve inthe public schools unless the person hasdemonstrated proficiency in basic reading,writing, and mathematics skills.’’ Cal.Educ.Code § 44252(b). At the same time,the legislature authorized the state’s Su-perintendent of Public Instruction to‘‘adopt an appropriate state test to mea-


sure proficiency in these basic skills.’’ Cal.Educ.Code § 44252(c). The Superinten-dent adopted the CBEST and, in May1983, CCTC assumed responsibility for ad-ministering and revising the test.

The CBEST is a pass-fail examinationconsisting of three sections: reading, writ-ing, and mathematics.1 The reading andmathematics sections each contain 50 mul-tiple-choice questions, 40 of which arescored. The writing section consists oftwo essays. The CBEST was revised in1995. At that time, questions that tested‘‘higher order’’ mathematical skills, such asgeometry, were eliminated from themathematics section of the test.

To pass the CBEST, a candidate mustreceive a ‘‘scaled’’ score of 123. Accord-ingly, a candidate passes by averaging 41points on each of the three sections (out ofa score range of 20 to 80). A scaled scoreof 41 on the reading section translates intoa raw score of 28 out of 40 questionscorrect; on the mathematics section, ascaled score of 41 equates to a raw score of26 out of 40 correct. Each of the twoessays is graded by two readers, who giveraw scores of between one and four pointsper essay. Thus, the range of possiblescores for the writing section is betweenfour and 16 points. A raw score of 12points translates into a scaled score of 41points. The CBEST employs a ‘‘compen-satory scoring’’ model, under which a can-didate passes the test with a scaled scorelower than 41 on a particular section, solong as his or her total scaled score is atleast 123.

A passing score on the CBEST is re-quired for all public elementary and sec-ondary school teachers in California. SeeCal. Educ.Code §§ 44256, 44257, 44259. Apassing score also is required for manynonteaching employees of the Californiapublic schools, including administrators,see id. § 44270, school counselors, see id.§ 44266, and school librarians, see id.§ 44269.

Since the CBEST’s inception, minoritycandidates have disproportionately re-ceived failing scores. The named Plain-tiffs are three nonprofit organizations thatrepresent the interests of minority edu-cators, and eight individual minority candi-dates. They brought this action againstthe State of California and the CCTC tochallenge the validity of the test underTitle VI and Title VII, on behalf of them-selves and all others similarly situated.The district court certified the followingclass:

All Latinos, African–Americans andAsians who have sought or are seekingCalifornia public school credentials andcertificated positions who have been, arebeing, or will be adversely affected intheir ability to obtain credentials andcertificated positions by [CBEST] re-sults.

In their complaint, Plaintiffs sought to en-join the use of the CBEST, alleging thatthe test has a disproportionate, adverseimpact on minority candidates and thatDefendants have failed to adopt screeningprocedures with a less adverse impact.

The parties filed cross-motions for sum-mary judgment, addressing the applicabili-ty of Titles VI and VII. In August 1993,the district court granted partial summaryjudgment to Plaintiffs, concluding thatboth Titles VI and VII apply to theCBEST. See Association of Mexican–American Educators v. California, 836F.Supp. 1534 (N.D.Cal.1993) (‘‘AMAE I ’’).Following a bench trial, the district courtheld that: (1) Plaintiffs had demonstratedthat the CBEST has a disparate impact onminorities; (2) the studies that were sub-mitted at trial demonstrated that the testwas a valid measure of job-related skills;(3) the level at which the passing scoreswere set reflected reasonable professionaljudgments about minimum levels of basicknowledge; and (4) Plaintiffs had failed todemonstrate that other, equally effective

1. The CBEST is described in more detail inthe district court’s opinion. See Association

of Mexican–American Educators v. California,937 F.Supp. 1397 (N.D.Cal.1996).


screening devices existed. The court en-tered a judgment in Defendants’ favor.See Association of Mexican–AmericanEducators v. California, 937 F.Supp. 1397(N.D.Cal.1996) (‘‘AMAE II ’’).

Defendants then presented the districtcourt with a bill for taxable costs pursuantto Federal Rule of Civil Procedure54(d)(1). That cost bill totaled $216,443.67.In an order dated February 12, 1997, thedistrict court denied the cost bill in itsentirety.

On appeal, Plaintiffs argue (1) that thedistrict court erred in concluding, aftertrial, that the CBEST was validated prop-erly and (2) that the court violated FederalRule of Evidence 706 by relying on theadvice of an expert who was not subject tocross-examination and did not prepare areport. Defendants cross-appeal with re-spect to the district court’s conclusions, onsummary judgment, that Titles VI and VIIapply. Defendants also appeal from thedistrict court’s order denying costs.2

DISCUSSION

I. Title VII

A. Title VII Applies to the CBEST.3

Defendants appeal from the districtcourt’s summary judgment in favor ofPlaintiffs on the issue of the applicabilityof Titles VI and VII. We review de novothe district court’s grant of summary judg-ment. See Robi v. Reed, 173 F.3d 736, 739(9th Cir.), cert. denied, ––– U.S. ––––, 120S.Ct. 375, 145 L.Ed.2d 293 (1999).

[1] As a threshold matter, we notethat, because we ultimately hold that theCBEST was validated properly (see PartI.B., below), we could decline to decidewhether Titles VI and VII apply. Werewe to do so, we simply would assume for

the sake of argument that the statutesapply and move immediately to the ques-tion of validation. Although that mightappear to be an expedient approach, wedecline to follow it for three reasons.First, as a matter of logic, the applicabilityof Title VI or Title VII is a predicate toany discussion of validation. Validationwould not be required, and indeed wouldnot even be relevant, if neither Title VInor Title VII applies. Second, as a matterof fairness, these parties deserve an an-swer not only to the bare question of whowins this case, but also to the underlyingquestion of the applicability of federal civilrights law to the CBEST. The state, inparticular, has proceeded for years on theassumption that those laws apply to itsadministration of the CBEST and has ex-pended considerable effort and expense inattempting to comply with federal law inthis area. If that effort was unnecessary,the state deserves to know, so that it mayact accordingly in the future. Third, as amatter of judicial economy, our answer tothe statutory question can avoid futurelitigation by other parties. We turn, then,to a discussion of Title VII’s application.

Title VII of the Civil Rights Act of 1964provides:

It shall be an unlawful employmentpractice for an employer-

(1) to fail or refuse to hire or to dis-charge any individual, or otherwise todiscriminate against any individual withrespect to his compensation, terms, con-ditions, or privileges of employment, be-cause of such individual’s race, color,religion, sex, or national origin[.]

42 U.S.C. § 2000e–2(a)(1). Title VII ap-plies ‘‘to governmental and private employ-ers alike.’’ Dothard v. Rawlinson, 433

2. This appeal originally was heard by a three-judge panel of this court; the court lateragreed to rehear the case en banc and with-drew the panel’s opinion. See Association ofMexican–American Educators v. California,195 F.3d 465 (9th Cir.1999), withdrawn, 208F.3d 786 (9th Cir.2000).

3. Chief Judge Hug, and Judges Schroeder,Reinhardt, Fernandez, Rymer, and Thomasjoin in this part of the majority opinion.


U.S. 321, 332 n. 14, 97 S.Ct. 2720, 53L.Ed.2d 786 (1977).

[2, 3] Plaintiffs and Defendants do nothave a direct employment relationship.Rather, Plaintiffs are employees and po-tential employees of individual school dis-tricts in California. That fact does not endour inquiry, however. A direct employ-ment relationship is not a prerequisite toTitle VII liability. Although ‘‘there mustbe some connection with an employmentrelationship for Title VII protections toapply,’’ that ‘‘connection with employmentneed not necessarily be direct.’’ Lutcherv. Musicians Union Local 47, 633 F.2d880, 883 (9th Cir.1980).

[4] Among other things, we have heldthat an entity that is not the direct em-ployer of a Title VII plaintiff neverthelessmay be liable if it ‘‘ ‘interferes with anindividual’s employment opportunities withanother employer.’ ’’ Gomez v. AlexianBros. Hosp., 698 F.2d 1019, 1021 (9th Cir.1983) (quoting Lutcher, 633 F.2d at 883 n.3). In Gomez, we held that the defendanthospital could be held liable under TitleVII for its discriminatory treatment of theplaintiff, notwithstanding the fact that theplaintiff was employed by a third party, ifthe defendant had interfered with theplaintiff’s employment by that third party.See id. at 1021.

In so holding, we followed the opinion ofthe District of Columbia Circuit in SibleyMemorial Hospital v. Wilson, 488 F.2d1338, 1340–41 (D.C.Cir.1973).4 In Sibley,the plaintiff was a male private-duty nurse.When a patient in the defendant hospitalrequested a private nurse, the hospital ar-ranged through a registry service to havea private nurse provided. That nurse at-tended the patient at the hospital, but waspaid directly by the patient. The plaintiffalleged that the defendant allowed malenurses like him to attend male patients

only, but allowed female nurses to attendboth male and female patients. See id. at1339–40.

The D.C. Circuit concluded that the de-fendant could be held liable under TitleVII even though it was not the plaintiff’sdirect employer. The court reasoned that,although the defendant did not employ theplaintiff, it exercised considerable powerover his ability to form employment rela-tionships with third parties. The courtnoted that Congress intended, through Ti-tle VII, to prohibit entities that possessedsuch power from ‘‘foreclos[ing], on invidi-ous grounds, access by any individual toemployment opportunities otherwise avail-able to him.’’ Id. at 1341. The courtfurther stated:

To permit a covered employer to exploitcircumstances particularly affording itthe capability of discriminatorily inter-fering with an individual’s employmentopportunities with another employer,while it could not do so with respect toemployment in its own service, would beto condone continued use of the verycriteria for employment that Congresshas prohibited.

Id. Finally, the court held that the defen-dant’s control over the premises on whichthe plaintiff provided his services, as wellas its control over the plaintiff’s access topatients, created a ‘‘highly visible nexuswith the creation and continuance of directemployment relationships between thirdparties’’ that brought the defendant’s ac-tions within the scope of Title VII. Id. at1342.

The D.C. Circuit’s holding in Sibley wasrooted in the text of Title VII. The Courtreasoned that, although Title VII appliesto ‘‘employees,’’ Congress extended theprotections of the statute to ‘‘any individu-al’’ who suffers discrimination: ‘‘nowhereare there words of limitation that restrict

4. Other cases following the ‘‘interference’’model from Sibley include: Charlton v. Para-mus Bd. of Educ., 25 F.3d 194, 202 (3d Cir.1994); Christopher v. Stouder Mem’l Hosp.,936 F.2d 870, 876–77 (6th Cir.1991); Pardazi

v. Cullman Med. Ctr., 838 F.2d 1155, 1156(11th Cir.1988); and Shehadeh v. Chesapeake& Potomac Tel. Co., 595 F.2d 711, 722(D.C.Cir.1978).


references in the Act to ‘any individual’ ascomprehending only an employee of anemployer.’’ Id. at 1341. As we did inGomez, we agree that the D.C. Circuit’sinterpretation of the statutory text is theproper one in view of ‘‘Congress’ directiveto read Title VII broadly so as to besteffectuate its remedial purposes.’’ Duf-field v. Robertson Stephens & Co., 144F.3d 1182, 1192 (9th Cir.), cert. denied, 525U.S. 982, 119 S.Ct. 445, 142 L.Ed.2d 399(1998).

We also note, as did the Sibley court,see 488 F.2d at 1342, that Congress explic-itly made Title VII applicable outside the‘‘direct employment’’ context by includingemployment agencies and labor organiza-tions in the statute’s coverage. See 42U.S.C. § 2000e–2(b), (c). In particular,employment agencies—defined as parties‘‘regularly undertaking with or withoutcompensation to procure employees for anemployer or to procure for employees op-portunities to work for an employer,’’ 42U.S.C. § 2000e(c)—may not ‘‘refuse to re-fer TTT or otherwise discriminate against’’any individual under Title VII. 42 U.S.C.§ 2000e–2(b). This provision of Title VII,like the others, applies to states. See Do-thard, 433 U.S. at 332 n. 14, 97 S.Ct. 2720;Dumas v. Town of Mt. Vernon, 612 F.2d974, 980 (5th Cir.1980), overruled on othergrounds by Larkin v. Pullman–StandardDiv., Pullman Inc., 854 F.2d 1549 (11thCir.1988), rev’d sub nom. Pullman–Stan-dard, Inc. v. Swint, 493 U.S. 929, 110 S.Ct.316, 107 L.Ed.2d 307 (1989). We do notsuggest that Defendants are, strictlyspeaking, an ‘‘employment agency’’ underTitle VII, although they perform an analo-gous function. Rather, we mention thisstatutory provision as evidence that Con-gress intended to close any loopholes inTitle VII’s coverage and to extend thestatute’s coverage to entities with actual‘‘[c]ontrol over access to the job market,’’Sibley, 488 F.2d at 1341, whether or notthey are direct employers.

In concluding that Title VII applies inthis case, the district court held that De-

fendants ‘‘interfere[d]’’ with Plaintiffs’ em-ployment opportunities with local schooldistricts in California by requiring, imple-menting, and administering the CBEST.See AMAE I, 836 F.Supp. at 1551. Weagree.

Our conclusion is dictated by the pecu-liar degree of control that the State ofCalifornia exercises over local school dis-tricts. In California, public schools are ‘‘amatter of statewide rather than local ormunicipal concern; their establishment,regulation and operation are covered bythe [state] constitution and the state Leg-islature is given comprehensive powers inrelation thereto.’’ Hall v. City of Taft, 47Cal.2d 177, 179, 302 P.2d 574, 576 (1956).The California legislature ‘‘has plenary au-thority over the education of California’syouth.’’ San Francisco NAACP v. SanFrancisco Unified Sch. Dist., 484 F.Supp.657, 662 (N.D.Cal.1979). It is ‘‘well settledthat the California Constitution makespublic education uniquely a fundamentalconcern of the state’’ and that ‘‘the degreeof supervision TTT retained by the Stateover the common school system is highindeed.’’ Butt v. California, 4 Cal.4th 668,685, 689, 15 Cal.Rptr.2d 480, 842 P.2d1240, 1251, 1254 (1992).

The state’s involvement is not limited togeneral legislative oversight but, rather,affects the day-to-day operations of localpublic schools. ‘‘Unlike most states, Cali-fornia school districts have budgets thatare controlled and funded by the stategovernment rather than the local dis-tricts.’’ Belanger v. Madera Unified Sch.Dist., 963 F.2d 248, 251 (9th Cir.1992). Asthe California Supreme Court noted inButt, California statutes regulate ‘‘districtorganization, elections, and governance;educational programs, instructional mate-rials, and proficiency testing; sex discrimi-nation and affirmative action; admissionstandards; compulsory attendance; schoolfacilities; rights and responsibilities of stu-dents and parents; holidays; schoolhealth, safety, and nutrition; teacher cre-dentialing and certification; rights and


duties of public school employees; and thepension system for public school teachers.’’Butt, 4 Cal.4th at 689, 15 Cal.Rptr.2d 480,842 P.2d at 1254 (citations omitted). Thestate also ‘‘dictates when students may beexpelled or suspended, and TTT exerts con-trol over the textbooks that are used inpublic schools.’’ Belanger, 963 F.2d at 253(citations omitted).

Indeed, the state is so entangled withthe operation of California’s local schooldistricts that individual districts are treat-ed as ‘‘state agencies’’ for purposes of theEleventh Amendment. See Freeman v.Oakland Unified Sch. Dist., 179 F.3d 846,846 (9th Cir.1999). The fact that the dis-tricts are entitled to assert EleventhAmendment immunity underscores thestate’s unusually high degree of involve-ment in the operation of local schools.

Against that background of ‘‘plenary’’state control, we have no difficulty con-cluding that the State of California is in atheoretical and practical position to ‘‘inter-fere’’ with the employment decisions oflocal school districts. And by requiring,formulating, and administering theCBEST, the state has ‘‘interfered’’ to adegree sufficient to bring it within thereach of Title VII. Through the CBEST,the state has created a limited list of can-didates from which local public school dis-tricts may hire. Private schools may hirecandidates who have not passed theCBEST; but California’s public schools,which are under the state’s control in al-most every aspect of their operations, maynot. Thus, in addition to controlling localdistricts’ budgets and textbooks and regu-lating the duties of public school employ-ees, the state dictates whom the districtsmay and may not hire. That degree ofcontrol over districts’ hiring decisions sub-jects Defendants to the coverage of TitleVII in this case.

The relationship between the State ofCalifornia and California’s local school dis-

tricts is analogous to the relationship be-tween a corporate parent and its whollyowned subsidiaries. ‘‘In the absence ofspecial circumstances, a parent corporationis not liable for the Title VII violations ofits wholly owned subsidiary.’’ Watson v.Gulf & W. Indus., 650 F.2d 990, 993 (9thCir.1981). In Watson, this court held thatthe parent corporation was not subject toTitle VII because the case presented no‘‘special circumstances.’’ Id. But the courtwent on to explain that, ‘‘[i]f there was anyevidence that [the parent] participated inor influenced the employment policies of[the subsidiary], TTT then we would bepresented with a very different case.’’ Id.Ours is that ‘‘very different case.’’ The‘‘parent’’ state has participated extensivelyin, and influenced, the employment policiesand practices of the ‘‘subsidiary’’ localschool districts; therefore, the state is cov-ered by Title VII.

[5] Defendants contend, however, thatthey are not subject to Title VII becausethe CBEST is merely a licensing examina-tion. The administration of such examina-tions, they argue, is not covered by TitleVII. As support for that argument, theycite several cases that have held that gov-ernmental agencies are not subject to TitleVII with regard to their licensing activi-ties. See Haddock v. Board of DentalExam’rs, 777 F.2d 462 (9th Cir.1985);Fields v. Hallsville Indep. Sch. Dist., 906F.2d 1017 (5th Cir.1990); George v. NewJersey Bd. of Veterinary Med. Exam’rs,794 F.2d 113 (3d Cir.1986).5

The cases on which Defendants rely arenot controlling for two reasons. First, thestate’s high level of involvement in theoperation of local public schools distin-guishes this case from those that Defen-dants cite. In those cases, licensing wasthe entire connection between the plain-tiffs and the defendants; here, the CBESTis but one aspect of pervasive state control.

5. To the same effect, see also Woodard v.Virginia Bd. of Bar Exam’rs, 598 F.2d 1345

(4th Cir.1979).


Second, the CBEST is not merely an ordi-nary licensing examination; it applies onlyto public school employees. In otherwords, the State of California is actingpursuant to its proprietary, as well as itspolice, power.

[6] There is no overarching ‘‘licensing’’exception to Title VII. The cases that De-fendants cite stand for a related but nar-rower proposition—that Title VII does notapply when the only connection among thelicensing agency, the plaintiff, and the uni-verse of prospective employers is the agen-cy’s implementation of a general licensingexamination. In such cases, to borrow thewords of the Sibley court, the agency doesnot have a ‘‘highly visible nexus with thecreation and continuance of direct employ-ment relationships between third parties,’’such as would subject it to Title VII underan ‘‘interference’’ theory. Sibley, 488 F.2dat 1342.

In Haddock, the plaintiff conceded thathis only connection to the defendant Boardof Dental Examiners was that the Boardhad given him an examination that hefailed. This court concluded that such aconnection, by itself, was insufficient tosubject the Board to Title VII liability.See 777 F.2d at 464. The plaintiff appar-ently did not argue that the Board had‘‘interfered’’ with his employment underthe principle adopted in Gomez and Sibley;if he did, the opinion does not mention it.

In Fields, the plaintiffs argued that thedefendants had violated Title VII throughtheir administration of the Texas Exami-nation for Current Administrators andTeachers (TECAT), a compulsory certifica-tion examination. See 906 F.2d at 1019.The district court granted summary judg-ment in favor of the state defendants, con-cluding that they did not have an employ-ment relationship with the plaintiffs. Onappeal, the plaintiffs challenged that con-clusion, arguing that the state defendantsactually controlled their employment eventhough the plaintiffs nominally were em-ployed by local districts. In rejecting thatargument, the Fifth Circuit noted that

‘‘[t]he only evidence presented by [theplaintiffs] suggesting control is the TexasState Board of Education’s administrationof the TECAT exam and its ability todecertify teachers who fail the exam.’’ Id.(emphasis in original). Of particular rele-vance to this case is footnote three of theopinion, which reads:

In a footnote in [the plaintiffs’] briefon appeal, they present evidence regard-ing state funding of facilities, payment ofsalaries and selection of textbooks. Asthis evidence was not before the districtcourt, it is not part of the summaryjudgment record on appeal.

Id. at 1019 n. 3. The court suggested thatthe outcome might be different if therewere such evidence of the state’s right tocontrol the work of the teachers. See id.at 1019–20.

Finally, in George, the plaintiff allegedthat the defendant Board of VeterinaryMedical Examiners had violated Title VIIby administering a licensing examinationthat discriminated against him on the basisof national origin. See 794 F.2d at 114.The Third Circuit affirmed the districtcourt’s dismissal of the plaintiff’s action.The court distinguished Sibley, on whichthe plaintiff had relied, stating:

In the Sibley Memorial Hospital casethe relationship of the hospital to theemployment by its patients of privateduty nurses secured for them by thehospital was very close, whereas in thepresent case there was nothing even re-motely resembling an employer-employ-ee relationship between the Board andthe plaintiff.

Id.

To summarize, the circumstances heredemonstrate a level of control and interfer-ence far greater than that in the ‘‘merelicensing’’ cases on which Defendants rely.The State of California exerts a high de-gree of control over the operation of localpublic school districts. That control is evi-denced both by the record and by Califor-nia law.


Defendants cite George for the furtherproposition that state licensing examina-tions are acts of state police power, towhich Title VII does not apply. But inGeorge, the Board was acting only pursu-ant to the state’s police power to protectthe public from incompetent veterinariansand was not attempting to control the hir-ing practices of, or the performance ofwork for, any specific employer. By con-trast, the CBEST does not apply across-the-board to all who wish to teach in Cali-fornia, as (for example) a veterinary licens-ing examination applies to all who wish topractice the profession of veterinary medi-cine within the state’s borders. Rather,the CBEST applies only to those who wishto teach for the public school system—asystem over which the State of Californiaexerts plenary control, including regulationof employees’ duties.

We conclude, therefore, that administra-tion of the CBEST is not solely an exerciseof the state’s police power. Rather, it isan exercise of both the state’s police powerand its proprietary power; and it is theexercise of proprietary power that subjectsthe state to the coverage of Title VII inthis case.

We hold that the CBEST examination issubject to the provisions of Title VII. Weturn next to the question whether theCBEST violates the provisions of that Act.

B. The District Court Did Not ClearlyErr in Concluding that the CBESTWas Properly Validated.6

[7, 8] ‘‘[D]iscriminatory tests are im-permissible unless shown, by professional-ly acceptable methods, to be predictive ofor significantly correlated with importantelements of work behavior which comprise

or are relevant to the job or jobs for whichcandidates are being evaluated.’’ Albe-marle Paper Co. v. Moody, 422 U.S. 405,431, 95 S.Ct. 2362, 45 L.Ed.2d 280 (1975).In evaluating employment tests that arealleged to have a racially disparate impact,we first consider whether the plaintiff hasestablished a prima facie case by demon-strating that the test causes a disparateimpact on the basis of race. Here, thedistrict court concluded that Plaintiffs hadestablished a prima facie case. SeeAMAE II, 937 F.Supp. at 1403. Defen-dants do not challenge that conclusion onappeal.

[9] Because Plaintiffs have establisheda prima facie case, the burden shifts toDefendants to demonstrate that theCBEST was validated properly.7 See Al-bemarle Paper, 422 U.S. at 425, 95 S.Ct.2362. In its detailed and careful opinion,the district court concluded that Defen-dants had met their burden and that thetest had been validated properly based onthree studies: (1) the 1982 Wheeler andElias study; (2) the 1985 Practitioners’Review; and (3) the 1995 Lundquist study.See AMAE II, 937 F.Supp. at 1411. Plain-tiffs challenge that conclusion.

[10] Although this court has not dis-cussed in detail the appropriate standardof review for a district court’s ruling ontest validation, we have applied the ‘‘clear-ly erroneous’’ standard. Clady v. Countyof Los Angeles, 770 F.2d 1421, 1434 (9thCir.1985). The other circuits that haveaddressed this issue likewise have appliedthe ‘‘clearly erroneous’’ standard. See,e.g., Melendez v. Illinois Bell Tel. Co., 79F.3d 661, 669 (7th Cir.1996); Bernard v.Gulf Oil Corp., 890 F.2d 735, 743 (5th

6. Chief Judge Hug, and Judges O’Scannlain,Fernandez, Rymer, and Kleinfeld join in thispart of the majority opinion.

7. In cases in which a defendant establishesthat a test is validated properly, the burdenshifts back to the plaintiff to show the exis-tence of other selection devices that alsowould ‘‘serve the employer’s legitimate inter-est in efficient and trustworthy workman-

ship,’’ but that are not discriminatory. Albe-marle Paper, 422 U.S. at 425, 95 S.Ct. 2362(internal quotation marks omitted). Here,Plaintiffs’ challenge is limited to the vali-dation of the test; they do not attempt onappeal to meet their burden of demonstratingthe existence of preferable selection devices,assuming that the CBEST is validated.


Cir.1989); Hamer v. City of Atlanta, 872F.2d 1521, 1526 (11th Cir.1989). The ques-tion whether a test has been validatedproperly is primarily a factual question,which depends on underlying factual de-terminations regarding the content and re-liability of the validation studies that adefendant utilized. Consistent with Clady,we review for clear error the districtcourt’s determination in this case that theCBEST was validated properly.

[11–13] To demonstrate that theCBEST was validated properly, Defen-dants are required to ‘‘show that it has ‘amanifest relationship to the employment inquestion.’ ’’ Clady, 770 F.2d at 1427 (quot-ing Griggs v. Duke Power Co., 401 U.S.424, 432, 91 S.Ct. 849, 28 L.Ed.2d 158(1971)). In cases in which a scored test,like this one, is challenged, we require thatthe test be ‘‘job related’’—that is, ‘‘that itactually measures skills, knowledge, orability required for successful performanceof the job.’’ Contreras v. City of LosAngeles, 656 F.2d 1267, 1271 (9th Cir.1981). In making a determination aboutjob-relatedness, we follow a three-step ap-proach:

The employer must first specify theparticular trait or characteristic whichthe selection device is being used toidentify or measure. The employermust then determine that the particulartrait or characteristic is an importantelement of work behavior. Finally, theemployer must demonstrate by ‘‘profes-sionally acceptable methods’’ that the

selection device is ‘‘predictive of or sig-nificantly correlated’’ with the elementof work behavior identified in the sec-ond step.

Craig v. County of Los Angeles, 626 F.2d659, 662 (9th Cir.1980) (quoting AlbemarlePaper, 422 U.S. at 431, 95 S.Ct. 2362).8

We will analyze each of those threesteps in turn. In addition, we will considerPlaintiffs’ argument that the passing scoreon the writing component of the CBEST isset too high.9

1. Specific Traits or Characteristics

The first step of our inquiry is to identi-fy the trait or characteristic that the test isdesigned to measure. See Craig, 626 F.2dat 662. Here, the district court found thatthe test was being used to measure ‘‘basicskills in reading, writing, and mathemat-ics,’’ AMAE II, 937 F.Supp. at 1411, andPlaintiffs do not dispute that finding.

2. Important Elementsof Work Behavior

Next, we consider whether basic skills inreading, writing, and mathematics are ‘‘im-portant element[s] of work behavior,’’Craig, 626 F.2d at 662, for the publicschool jobs for which the test is required.The district court found that the testedskills were important to the jobs at issue.See AMAE II, 937 F.Supp. at 1419. Plain-tiffs challenge that finding on threegrounds. First, they argue that the 1985Practitioners’ Review failed to identify any

8. Also relevant to our inquiry are the EqualEmployment Opportunity Commission’s(‘‘EEOC’’) Uniform Guidelines on EmployeeSelection Procedures (‘‘Guidelines’’), whichare codified at 29 C.F.R. pt. 1607. Althoughthe Guidelines are not legally binding, theyare ‘‘entitled to great deference.’’ AlbemarlePaper, 422 U.S. at 431, 95 S.Ct. 2362 (internalquotation marks omitted). Failure to complywith the Guidelines, although not automati-cally fatal to an employment test, ‘‘diminishesthe probative value of the defendants’ vali-dation study.’’ Clady, 770 F.2d at 1430 (in-ternal quotation marks omitted). The studieson which Defendants rely were content validi-ty studies. See 29 C.F.R. § 1607.5. Such

studies establish whether the content of a testapproximates the knowledge, skills, or abili-ties that an applicant will use on the job. See29 C.F.R. § 1607.14. ‘‘Evidence of the validityof a test or other selection procedure by acontent validity study should consist of datashowing that the content of the selection pro-cedure is representative of important aspectsof performance on the job for which the can-didates are to be evaluated.’’ 29 C.F.R.§ 1607.5(B).

9. Our discussion of validation owes much tothe original panel majority’s excellent treat-ment of these issues.


particular work behaviors or job dutiesand thus could not be used to assesswhether the CBEST measured importantelements of work behavior. Second, theyargue that Lundquist’s 1995 study failed todistinguish ‘‘important’’ skills from skillsthat are less important. Third, they arguethat Defendants failed to demonstrate thatthe CBEST is job-related for the particu-lar positions for which it is required. Weaddress each of those arguments in turn.

[14] Plaintiffs first argue that the 1985Practitioners’ Review, conducted by Dr.Richard Watkins, was inadequate becauseit failed to identify specific job duties towhich the CBEST skills could be correlat-ed. We conclude that the district courtdid not clearly err in finding that the 1985study adequately identified the ‘‘element[s]of work behavior,’’ Craig, 626 F.2d at 662,that the CBEST is designed to measure.

The district court found that the Reviewcomprised the ‘‘pooled judgments’’ ofknowledgeable persons, such as incum-bents in the jobs, ‘‘about the relevance ofthe skills tested on the CBEST to the jobsfor which it is required, an appropriateform of a job analysis under the profes-sional standards of the time.’’ AMAE II,937 F.Supp. at 1419. Specifically, thePractitioners’ Review consulted 234 teach-ers, administrators, and other publicschool employees, 36 percent of whomwere members of minority groups. See id.at 1413. ‘‘The participants took part innine review panels, in which they judgedthe relevance of both the skills assessed bythe CBEST and the test items them-selves.’’ Id. They were asked to rate howrelevant each of the CBEST skills wouldbe to the work of four groups: (1) elemen-tary school teachers; (2) secondary schoolteachers; (3) librarians, counselors, andattendance officers; and (4) school admin-istrators. See id. The possible ratingsranged from ‘‘not relevant’’ to ‘‘very rele-vant.’’ Id.

Thus, the Practitioners’ Review wasdesigned to learn from teachers, adminis-trators, and other school employees the

categories of skills that they consideredrelevant to their own jobs. The skillsmeasured by the study tracked the cate-gories of skills measured by the CBEST,and the skills were described in some de-tail on the rating forms used by the pan-el members. For example, the broadskill category ‘‘Mathematical concepts andrelationships’’ was further described asfollows:

Questions in this category test the un-derstanding of basic concepts, such asthe meaning of certain terms (area, forexample), order among numbers, rela-tionships shown by graphs, elementaryprobability, and the like. Questions inthis category may be from arithmetic,algebra, or elementary geometry.

The study’s participants were guided bydetailed instructions relating to each skillcategory and were told to rank the impor-tance of each skill for both teaching andnonteaching jobs. The study thereforesatisfies the requirement from Craig thatthe employer determine whether a ‘‘specif-ic trait or characteristic is an importantelement of work behavior.’’ Craig, 626F.2d at 662. The district court did notclearly err in concluding that the 1985Practitioners’ Review was ‘‘an appropriateform of a job analysis under the profes-sional standards of the time.’’ AMAE II,937 F.Supp. at 1419.

We next consider Plaintiff’s second chal-lenge under the ‘‘important elements’’prong of Craig. Plaintiffs do not challengethe 1995 Lundquist study’s methodologyfor identifying job-related skills. They do,however, challenge that study’s method fordetermining which skills are ‘‘important’’to particular jobs.

[15] Dr. Lundquist polled experts andinterviewed and observed educators in or-der to develop a list of activities and skillsused by educators. See AMAE II, 937F.Supp. at 1414. She then polled 1,330teachers and administrators, asking themto rate the importance of those activitiesand skills on a four-point scale from 0


(‘‘not applicable’’) to 3 (‘‘critical’’). Activi-ties and skills were retained only if ‘‘atleast 80 percent of the survey respondentsrated the activity or skill as applicable tothe job and the mean importance ratingwas 1.5 or higher.’’ Id. at 1414 (emphasisin original). Applying those standards ledto elimination of a number of the activitiesand skills from Dr. Lundquist’s list. Afterconducting additional studies, which aredescribed in detail in the district court’sopinion, see id. at 1415–17, Dr. Lundquistthen formulated new specifications for allthree sections of the test. In response tothose specifications, Defendants revisedthe CBEST before they administered theAugust 1995 test.

Plaintiffs focus on the fact that Dr.Lundquist retained activities and skills onher list if they received a ‘‘mean impor-tance rating’’ of 1.5 on a scale that desig-nated 2 as ‘‘important’’ and 1 as ‘‘minor.’’By using a mean rating of 1.5, Plaintiffsargue, Dr. Lundquist retained skills andactivities that were rated as ‘‘less thanimportant’’ by the study’s participants.Therefore, their argument proceeds, thestudy violated the requirement from Craigthat only ‘‘important’’ work skills be mea-sured.

The district court rejected Plaintiffs’ ar-gument, finding that ‘‘Dr. Lundquist’s de-cisions reflect manifestly reasonable pro-fessional judgmentsTTTT With respect tothe 1.5 mean, as Dr. Lundquist testified attrial, a 1.5 rounds up to 2.0. It must beremembered that the mean rating of 1.5was coupled with an 80 percent endorse-ment criterion, which is quite stringent.’’AMAE II, 937 F.Supp. at 1418 n. 35.

We agree that it is theoretically possibleto imagine a circumstance that illustratesPlaintiffs’ concerns on this point. For ex-ample, suppose that 80 percent of thestudy’s participants agreed that a particu-lar skill was relevant, but 75 percent ofthem rated that skill’s importance as ‘‘mi-nor.’’ If the remaining 25 percent ratedthe skill as ‘‘critical,’’ then the skill wouldbe retained despite the fact that a majority

of the study’s participants rated its impor-tance as ‘‘minor.’’ Although that scenariois possible, such a skewed distribution ofresponses is unlikely. Plaintiffs presentonly a theoretical possibility that such‘‘highly relevant but unimportant’’ skillsremained on Dr. Lundquist’s list. Fur-ther, as the district court noted, Dr. Lund-quist conducted additional ‘‘importance’’reviews of the mathematics section of thetest.

Validation studies ‘‘are by their naturedifficult, expensive, time consuming andrarely, if ever, free of error.’’ Cleghorn v.Herrington, 813 F.2d 992, 996 (9th Cir.1987). Plaintiffs’ argument demonstrates,at most, that Dr. Lundquist’s study maynot be totally free of error. But the argu-ment does not persuade us that the districtcourt clearly erred in relying on Dr. Lund-quist’s study.

[16] Finally, Plaintiffs argue that De-fendants failed to conduct job-specificstudies to determine that the CBEST is‘‘job related for the position[s] in ques-tion.’’ 42 U.S.C. § 2000e–2(k)(1)(A)(i).The CBEST is not intended to measure allthe skills that are relevant to all the jobsfor which it is required. (Indeed, it doesnot purport to measure all the skills ofany of the jobs for which it is required.)Rather, the CBEST is intended to estab-lish only a minimum level of competence inthree areas of basic educational skills.The question is whether the validationstudies in this case have satisfied the re-quirement that those skills be ‘‘job relat-ed’’ for all the positions in question. Thedistrict court found that the validationstudies adequately analyzed the CBEST interms of both the teaching and nonteach-ing jobs for which the test is required.See AMAE II, 937 F.Supp. at 1418–19.The district court did not clearly err in sofinding.

Both the 1985 and the 1995 validationstudies contained adequate considerationof the specific positions for which theCBEST is required. The 1985 Practition-


ers’ Review defined the positions that itanalyzed as (1) elementary school teachers,(2) secondary school teachers, (3) librari-ans, counselors, and attendance officers,and (4) school administrators. All partici-pants in the study were asked to judge therelevance of the CBEST skills by categoryfor those jobs. Because the study’s partic-ipants were asked to determine the rele-vance of the basic skills measured by theCBEST to the disparate groups of posi-tions for which the test is required, wecannot say that the district court clearlyerred in finding the job analysis in thePractitioners’ Review to be sufficientlyspecific and particularized.

The 1995 Lundquist study, as noted,identified job activities through observa-tion, interviews, and reviews of specializedliterature. See id. at 1414. Dr. Lund-quist then pared her list of job skills andactivities through surveys of educatorsand arrived at a list of ‘‘common skill re-quirements’’ that were relevant for bothteachers and administrators. See id. at1414–15. Her study reports:

Basic skill ratings were examined foradministrators to determine if the sameskill sets applied to both teacher andadministrator jobs. Results showed allbut one skill item (a math item) retainedfor teachers also applied to the adminis-trator group. Thus, the basic skill re-quirements identified for teachers werefound to be job-related for administra-tors as well, and the same test specifica-tions may be used to test basic skills forteachers and administrators.

Dr. Lundquist’s study classified jobs forwhich the CBEST is required as either‘‘teacher’’ or ‘‘administrator’’ and deter-mined that the CBEST was valid for bothgroups of positions. Accordingly, the 1995study considered the validity of theCBEST across the range of jobs for whichthe test is required. The district courtaccepted the study’s conclusions and foundthat the CBEST had been validated ade-quately ‘‘with respect to teaching and non-

teaching jobs.’’ Id. at 1418. On this rec-ord, that finding is not clearly erroneous.

In sum, we hold that the district courtdid not clearly err in finding that the skillsmeasured by the CBEST are ‘‘importantelement[s] of work behavior’’ with regardto the jobs for which the test is required.Craig, 626 F.2d at 662.

3. Actual Measurement of Skills

[17] The final step in this court’sthree-step analysis from Craig is to deter-mine whether Defendants have demon-strated by ‘‘professionally acceptablemethods that the selection device is pre-dictive of or significantly correlated withthe element of work behavior’’ that it isdesigned to measure. See id. (internalquotation marks omitted). The districtcourt concluded that ‘‘the CBEST actuallymeasures TTT basic skills [in reading, writ-ing, and mathematics].’’ AMAE II, 937F.Supp. at 1411. Plaintiffs claim that thedistrict court simply accepted the ‘‘facial’’validity of the CBEST without any evi-dence that it actually measures the basicskills that it purports to test.

This court held in Contreras that ‘‘a keyrequirement of [the] third step [fromCraig ], a requirement essential to proof ofjob relatedness generally, is that the vali-dation method be professionally accept-able.’’ 656 F.2d at 1282. Here, there isevidence in the record from an expert, Dr.William A. Mehrens, that supports the dis-trict court’s findings on this issue. Dr.Mehrens reported:

ETS [Educational Testing Service] per-sonnel wrote some of the original itemsand assisted the test development com-mittees in writing other items. ETS iswell known and respected as a developerof standardized tests. They have welltrained item writers and an impressiveinternal set of guidelines they followwith respect to item writing.

When asked whether ‘‘the CBEST devel-opment [was] appropriate with respect towriting and evaluating the items,’’ he re-ported:


It has been. Many of the items camefrom an existing ETS pool. Otherswere written specifically for CBEST bymembers of the test development com-mittee in concert with ETS test develop-ment specialists. The individuals on thecommittees worked with specialists fromETS to further develop and define thecontent specifications, to review an exist-ing ETS pool test item, to write new testitems, and to review the items submittedby fellow committee members. In addi-tion, the committees studied all of thedata from the field testing, made recom-mendations for revisions as they feltnecessary, and reviewed all final testresults.

There also is additional evidence in therecord that the test questions were‘‘matched’’ to the skills that they wereintended to measure. The district courtreferred to the ‘‘Curriculum Matching Pro-ject, in which two ETS employees TTTmatched CBEST test specifications to ma-terial found in textbooks purportedly usedin the California public schools.’’ AMAEII, 937 F.Supp. at 1412 n. 21. The districtcourt was somewhat critical of this study,but noted that ‘‘the study did support theoverall conclusion that the kinds of skillstested on the CBEST can be found inelementary and secondary school text-books.’’ Id.

In short, there is evidence—even if notoverwhelming evidence—that the develop-ment and evaluation of the CBEST wereappropriate and that the test measures thetypes of skills that it was designed tomeasure. We therefore hold that the dis-trict court did not clearly err in concludingthat the test questions had been shown by

professionally acceptable methods to be‘‘predictive of or significantly correlatedwith the element of work behavior’’ thatthey were designed to measure. Craig,626 F.2d at 662 (internal quotation marksomitted).

In sum, we hold that the district courtdid not clearly err in concluding that theCBEST was validated properly.10

4. Standards for Passing Scores

Plaintiffs also argue that the 12–out–of–16 passing score on the writing section ofthe CBEST is too high. According toPlaintiffs, the 1982 Wheeler and Eliasstudy demonstrates that the proper pass-ing score is 9 or 10 out of 16.

[18] An employer is not required tovalidate separately the selection of particu-lar passing scores on an employment test.See id. at 665. Rather, the EEOC’sGuidelines more generally provide:‘‘Where cutoff scores are used, they shouldnormally be set so as to be reasonable andconsistent with normal expectations of ac-ceptable proficiency within the workforce.’’ 29 C.F.R. § 1607.5(H). This courtpreviously has applied that standard. See,e.g., Craig, 626 F.2d at 665. In analyzingthe Guidelines’ scoring requirement, theSecond Circuit has stated that an employ-er ‘‘might establish a valid cutoff score byusing a professional estimate of the requi-site ability levels, or, at the very least, byanalyzing the test results to locate a logical‘break-point’ in the distribution of scores.’’Guardians Ass’n of New York City PoliceDep’t, Inc. v. Civil Serv. Comm’n of NewYork, 630 F.2d 79, 105 (2d Cir.1980).

10. We affirm the district court’s conclusionthat the CBEST was validated properly basedon the second and third studies: the 1985Practitioners’ Review and the 1995 Lundquiststudy. Because we conclude that those testsadequately support the district court’s findingof job-relatedness, we do not discuss in detailthe 1982 Wheeler and Elias study, on whichthe district court also relied. All three studiesreviewed the same basic version of theCBEST: the version that was given between

1983 and August 1995, when the test wasrevised. There is no allegation that the testchanged significantly during that period orthat the test was invalid until 1985 but validthereafter. Accordingly, it is sufficient for usto conclude that the test was validated prop-erly by the 1985 and 1995 studies and, be-cause we so conclude, we do not decidewhether the 1982 study provides an addition-al source of validation.


Here, the district court found that ‘‘thepassing scores on the CBEST reflect rea-sonable judgments about the minimum lev-el of basic skills competence that should berequired of teachers.’’ AMAE II, 937F.Supp. at 1420. The evidence before thecourt revealed that the California Superin-tendent of Public Instruction, who was re-sponsible for establishing the cutoff scores,relied on polling data created as part ofthe Wheeler and Elias study in setting thecutoff for the writing section of the test.As part of that study, 44 readers reviewedapproximately 6,800 CBEST essays andmade recommendations regarding the cut-off between passing and failing scores.The readers unanimously agreed that araw score of 12 out of 16 was a ‘‘passing’’score. Approximately 80 percent of thereaders agreed that a score of 11 out of 16could be a ‘‘passing’’ score. On that basis,the Superintendent established a passingscore of 12 out of 16, with an absoluteminimum of 11 out of 16 under the ‘‘com-pensatory scoring’’ system.

[19] Those cutoff scores represent a‘‘logical breakpoint’’ between passing andfailing scores. Plaintiffs argue that thebreakpoint should have been set at 9 or 10out of 16, because a majority of the read-ers opined that 10 out of 16 was a ‘‘pass-ing’’ score. But the Superintendent wasnot required to set the score at the lowestlevel that a majority of the readers consid-ered to be ‘‘passing.’’ Rather, he was re-quired to set a cutoff that was logical,reasonable, and consistent with the databefore him. He chose to set the cutoff at alevel that all the readers agreed was ‘‘pass-ing,’’ and to set an absolute minimum at alevel that 80 percent of the readersthought was ‘‘passing.’’ The district courtfound that the Superintendent’s decision toset the cutoff score at that level was con-sistent with the EEOC’s Guidelines. We

conclude that the district court did notclearly err in so finding.

II. Title VI 11

Because we have concluded that TitleVII applies to the CBEST, we need notconsider whether Title VI also applies.See 42 U.S.C. § 2000d. Plaintiffs proceedon a disparate impact theory under bothTitle VI and Title VII. Thus, our discus-sion of the merits under Title VII, and ofthe validation of the CBEST, would re-solve Plaintiffs’ claims on the merits underTitle VI as well. Accordingly, we declineto issue an advisory opinion on the applica-bility of Title VI.

III. The District Court’s Appointment ofa Technical Advisor 12

Plaintiffs also argue that the proceed-ings were tainted by the influence of Dr.Stephen Klein. Dr. Klein was appointedby the district court as a technical advisor,but was not called as an expert witness,was not subject to cross-examination, anddid not furnish an expert’s report.

[20] In those rare cases in which out-side technical expertise would be helpful toa district court, the court may appoint atechnical advisor like Dr. Klein. See Reil-ly v. United States, 863 F.2d 149, 156 (1stCir.1988); see also General Elec. Co. v.Joiner, 522 U.S. 136, 149, 118 S.Ct. 512,139 L.Ed.2d 508 (1997) (Breyer, J., concur-ring) (endorsing the appointment of specialmasters and specially trained law clerks toassist district courts with scientific or tech-nical evidence). The court appointed Dr.Klein in an order that specifically identi-fied him as a technical advisor.

[21] Plaintiffs argue that the courtcommitted legal error under Federal Ruleof Evidence 706(a) by neither requiringDr. Klein to submit a report nor allowinghim to be cross-examined. The short an-

11. Chief Judge Hug, and Judges Schroeder,Reinhardt, O’Scannlain, Fernandez, Rymer,Kleinfeld, Thomas, and Gould join in this partof the majority opinion.

12. Chief Judge Hug, and Judges Schroeder,Reinhardt, O’Scannlain, Fernandez, Rymer,Kleinfeld, Thomas, and Gould join in this partof the majority opinion.


swer to Plaintiffs’ argument is that Rule706 applies to court-appointed expert wit-nesses, but not to technical advisors likeDr. Klein. See Reilly, 863 F.2d at 155.

At one point in the proceedings, thedistrict court told the parties that it in-tended to call Dr. Klein to testify near theconclusion of the trial and that the courtwould permit cross-examination at thattime. Had the court called him, Dr. Kleinwould have testified as an expert witness,and Federal Rule of Evidence 706 wouldhave applied. But the court never calledDr. Klein, and there is no indication in therecord that the court relied on Dr. Klein asa source of evidence. On this record, Dr.Klein acted only as a court-appointed tech-nical advisor, and the district court did noterr in refusing to allow cross-examinationor to require an expert’s report.

In his dissent, Judge Tashima agreeswith us that district courts retain inherentauthority to appoint technical advisors inappropriate cases; that this was an appro-priate case for a technical advisor; thatRule 706(a) does not apply; and that weare reviewing for an abuse of discretion.

[22] His disagreement rests on hisanalysis of how we should respond to therelative paucity of information in the rec-ord about Dr. Klein’s interaction with thedistrict court. In our view, the absence ofany evidence even suggesting an impro-priety on the part of the district courtmilitates against a conclusion that thecourt abused its discretion. Although it isat least possible, as Judge Tashima sug-gests, that ‘‘Dr. Klein may have impermis-

sibly influenced the court’s ultimate find-ing,’’ diss. op. at 611, we instead assumethat the district court did its job properlywhen we lack evidence to the contrary.13

Judge Tashima also proposes a list ofprocedures for district courts to followwhen appointing technical advisors. Evenassuming that those procedures are appro-priate, the district court did not have thebenefit of Judge Tashima’s dissent beforethis trial, and we will not fault the courtfor failing to foresee his recommendations.We are not willing to find an abuse ofdiscretion and to undo this entire trialbecause the district court did not follow aset of guidelines that are required nowherein the rules or relevant case law.

IV. The District Court’s Refusal toAward Costs 14

[23] Defendants appeal from the dis-trict court’s order denying their cost bill inthe amount of $216,443.67. Federal Ruleof Civil Procedure 54(d)(1) provides that‘‘costs other than attorneys’ fees shall beallowed as of course to the prevailing partyunless the court otherwise directs.’’ By itsterms, the rule creates a presumption infavor of awarding costs to a prevailingparty, but vests in the district court discre-tion to refuse to award costs. See Nation-al Info. Servs., Inc. v. TRW, Inc., 51 F.3d1470, 1471 (9th Cir.1995).

[24] That discretion is not unlimited.A district court must ‘‘specify reasons’’ forits refusal to award costs. SubscriptionTelevision, Inc. v. Southern Cal. TheatreOwners Ass’n, 576 F.2d 230, 234 (9th Cir.

13. The present case is distinguishable fromthose that Judge Tashima cites, see dissentingop. at 619 n. 9. In all but the first and last ofthe cited cases, the lower tribunal failed tomake findings of fact that were required byexisting law. The first cited case was a death-penalty habeas corpus case in which the courtremanded for an evidentiary hearing on themerits of several claims that had been dis-missed, improperly, on procedural grounds.The panel’s decision to remand was dictatedby its application of the specific rules andprocedures that govern habeas corpus in capi-

tal cases. In the last case, the district courtwholly failed to address a potentially disposi-tive legal issue.

By contrast, here the district court had be-fore it a nonhabeas civil case, in which itdecided all issues presented and made allrequired findings of fact. The only questionbefore us is whether the district court’s use ofa technical advisor was an abuse of discretion.

14. Chief Judge Hug, and Judges Schroeder,Reinhardt, Thomas, and Gould join in thispart of the majority opinion.


1978). On appeal, we determine whetherthe reasons that the district court hasspecified are appropriate and whether,considering those reasons, the courtabused its discretion in denying costs. SeeNational Info. Servs., 51 F.3d at 1471–72.

In past cases, this court has approvedthe following reasons for refusing to awardcosts to a prevailing party: the losing par-ty’s limited financial resources, see Nation-al Org. for Women v. Bank of Cal., 680F.2d 1291, 1294 (9th Cir.1982); see alsoWrighten v. Metropolitan Hosps., Inc., 726F.2d 1346, 1358 (9th Cir.1984); Moore v.Hughes Helicopters, Inc., 708 F.2d 475,486 (9th Cir.1983); and misconduct on thepart of the prevailing party, see NationalInfo. Servs., 51 F.3d at 1472. Further, inStanley v. University of Southern Califor-nia, 178 F.3d 1069, 1079–80 (9th Cir.), cert.denied, ––– U.S. ––––, 120 S.Ct. 533, 145L.Ed.2d 413 (1999), we held that the dis-trict court abused its discretion in denyinga losing civil rights plaintiff’s motion to re-tax costs without considering (1) the plain-tiff’s limited financial resources; and (2)‘‘the chilling effect of imposing such highcosts on future civil rights litigants.’’

Here, the district court gave four rea-sons for denying costs to Defendants: (1)the case ‘‘involve[s] issues of substantialpublic importance,’’ specifically ‘‘education-al quality, interracial disparities in eco-nomic opportunity, and access to positionsof social influence’’; (2) there is great eco-nomic disparity between Plaintiffs, whoare individuals and ‘‘small nonprofit edu-cational organizations,’’ and the State ofCalifornia; (3) the issues in the case are

close and difficult; 15 and (4) Plaintiffs’case, although unsuccessful, had somemerit, as evidenced by the 1995 modifica-tion of the CBEST to eliminate ‘‘higherorder’’ mathematics questions.

[25] Defendants argue that the districtcourt’s reasons for denying costs were im-proper. According to Defendants, thiscourt’s opinion in National InformationServices establishes that the only properreason for denying costs to a prevailingparty is to punish misconduct by that par-ty. We disagree.

As noted, this court previously has heldthat district courts may consider other,nonpunitive reasons for denying costs to aprevailing party. National InformationServices does appear to suggest that sucha denial is proper only as a means ofpunishing a prevailing but undeserving liti-gant. See also Zenith Ins. Co. v. Breslaw,108 F.3d 205, 207–08 (9th Cir.1997) (follow-ing National Information Services ). Butthat suggestion is inconsistent with earlieropinions of this court, see, e.g., NationalOrg. for Women, 680 F.2d at 1294, opin-ions that National Information Services,as a panel’s opinion, could not (and did notpurport to) overrule.

More importantly, we see no basis forlimiting district courts’ discretion in themanner that Defendants suggest. Therule itself contains no such limitation; itprovides simply that costs shall be allowedto the prevailing party unless the districtcourt ‘‘otherwise directs.’’ The require-ment that district courts give reasons fordenying costs flows logically from the pre-

15. Although we have not previously approvedthat reason for denying costs, other circuitshave. See Teague v. Bakker, 35 F.3d 978, 997(4th Cir.1994); White & White, Inc. v. Ameri-can Hosp. Supply Corp., 786 F.2d 728, 733(6th Cir.1986). Other reasons for denyingcosts that circuit courts have approved in-clude: the ‘‘nominal’’ or partial nature of theprevailing party’s recovery, see Richmond v.Southwire Co., 980 F.2d 518, 520 (8th Cir.1992); Howell Petroleum Corp. v. Samson Re-sources Co., 903 F.2d 778, 783 (10th Cir.1990); and the good faith of the losing party,

see Teague, 35 F.3d at 997; White, 786 F.2d at730. The Seventh Circuit also has suggested,in dictum, that the denial of costs might beappropriate in cases that present ‘‘landmarkissues of national importance.’’ Popeil Bros.v. Schick Elec., Inc., 516 F.2d 772, 776 (7thCir.1975); see also Delta Air Lines, Inc. v.Colbert, 692 F.2d 489, 490 (7th Cir.1982).That dictum is similar to the district court’sfirst reason for denying costs in this case,namely, the exceptional public importance ofthe issues presented.


sumption in favor of costs that is embodiedin the text of the rule; if a district courtwishes to depart from that presumption, itmust explain why ‘‘so that the appellatecourt will be able to determine whether ornot the trial court abused its discretion.’’Subscription Television, 576 F.2d at 234.But the limitation on district courts’ dis-cretion that Defendants advocate curtailsthat discretion in a manner and to a de-gree that are inappropriate in view of thebroad wording of the rule. We now over-rule National Information Services to theextent that it held that only misconductmay support the denial of costs to a pre-vailing party.

[26] Federal Rule of Civil Procedure54(d)(1) establishes that costs are to beawarded as a matter of course in the ordi-nary case. Our requirement that a districtcourt give reasons for denying costs is, inessence, a requirement that the court ex-plain why a case is not ‘‘ordinary’’ andwhy, in the circumstances, it would beinappropriate or inequitable to awardcosts. Misconduct on the part of the pre-vailing party is one factor that might ren-der a case ‘‘extraordinary.’’ But it is notthe only such factor. Here, the reasonsthat the district court gave for denyingcosts reflect what is clear at a glance: Thisis an extraordinary, and extraordinarilyimportant, case. Plaintiffs are a group ofindividuals and nonprofit organizations.The record demonstrates that their re-sources are limited. They have broughtan action that presents issues of the gra-vest public importance, and the action af-fects tens of thousands of Californians andthe state’s public school system as a whole.The issues in the case are close and com-plex. Although Plaintiffs have not pre-vailed in this action, their claims are notwithout merit. Indeed, as the districtcourt pointed out, Defendants substantiallyaltered the CBEST during the pendencyof this litigation. Finally, costs in this caseare extraordinarily high. In keeping withour decision in Stanley, we note that di-vesting district courts of discretion to limit

or to refuse such overwhelming costs inimportant, close, but ultimately unsuccess-ful civil rights cases like this one mighthave the regrettable effect of discouragingpotential plaintiffs from bringing suchcases at all.

We do not mean to suggest that thepresumption in favor of awarding costs toprevailing parties does not apply to defen-dants in civil rights actions. Nor are weattempting to create an exhaustive list of‘‘good reasons’’ for declining to awardcosts. We simply hold that the reasonsthat the district court gave for refusing toaward costs in this case were appropriateunder Rule 54(d)(1) and that, consideringthose reasons, the court did not abuse itsdiscretion in refusing to award costs toDefendants.

CONCLUSION

For the reasons stated, we hold thatTitle VII applies to the CBEST; that theCBEST was validated properly; that thedistrict court permissibly used a technicaladvisor; and that the district court did notabuse its discretion by refusing to awardcosts to Defendants. Accordingly, we af-firm both the judgment in Defendants’ fa-vor and the order denying them costs.

AFFIRMED.

REINHARDT, Circuit Judge, withwhom Circuit Judges SCHROEDER andTHOMAS join, concurring in part anddissenting in part:

The result we reach is most unfortunate.Not only is it wrong as a matter of law,but it may lead some to conclude that weare insensitive to the needs and problemsof minority educators and minority stu-dents. While that may not be so, onething is certain. The loser in the casebefore us is clearly the already beleag-uered public education system of the stateof California.

Although I concur in Section IA andParts II, III, and IV of the majority opin-ion, I cannot agree with the majority’s


holding that the CBEST was properly vali-dated under Title VII. As a result of thisruling, qualified minority educators—teachers, administrators, librarians, andother officials—will be denied the opportu-nity to work in California’s severely under-staffed public schools, simply because theyfailed to pass a test that concededly has adisparate impact on minority group mem-bers. Despite Title VII’s clear require-ment that a test that has a disparate im-pact on minorities must be validated byreference to a particular job, the majorityaccepts validation studies that fail to dif-ferentiate among different school jobs,ranging from bilingual education teachersto mathematics teachers to physical edu-cation teachers. Further, even though Ti-tle VII demands that the defendants dem-onstrate ‘‘by professionally acceptablemethods’’ that the test items are predictiveof or significantly correlated to job-relatedskills, the majority finds th

572 231 federal reporter, 3d series · 2018. 1. 12. · 574 231 federal reporter, 3d series sions...

Documents