Isabel v. City of Memphis

claims Finally, despite Trammel’s ISABEL; Parker; Richard Sharon discharge the conditional contrary,

to the Sanders; Williams, Gregory Walter jus constitute a criminal this case does Jr., Plaintiffs-Appellees/Cross-Appel the Guidelines. Under tice sentence under lants, Sentencing Guide the 2002 version

lines, are added to a defen points two history “if the defen dant’s criminal score while dant committed the instant offense MEMPHIS, Defendant- CITY OF sentence, in justice criminal under Appellant/Cross-Appellee. cluding probation, parole, supervised re 03-6011, 03-5912, 03-5914, 03-6231. Nos. release, lease, or es imprisonment, work 4Al.l(d). This cape status.” U.S.S.G. Appeals, States United conditional that a term of Court has held Sixth Circuit. imposed in connec discharge, even when suspended rather than a tion with fíne Argued: Nov. 2004. sentence, jail properly considered a term April in Decided and Filed: probation and therefore should be history un cluded criminal calculations

der the Guidelines. See United States v. (6th

Rollins, 539-40 Cir.

2004) (holding two-year that a conditional

discharge imposed Kentucky under law properly purposes

was considered for the calculating history criminal defendant’s

score). Thus, Trammel’s conditional dis justice a criminal

charge does constitute Because

sentence under Guidelines. un discharge

the conditional effect 22, 2001, July beginning

til activities of underlying conspiracy

Trammel’s offense and, probationary period

fell within this

thus, two-point increase under section

4Al.l(d) appropriate. of the Guidelines was

IV. reasons,

For the we con- above-stated not err in

clude the district did

calculating history Trammel’s criminal However,

score. we VACATE Trammel’s for resentencing

sentence and REMAND opinion with the

consistent with decision Booker.

Supreme Court’s III, Britt

ARGUED: Louis P. Thomas Walsh, Jr., Harrison, Memphis, Ford & J. Tennessee, for David M. Appellant. Sulli- van, Tennessee, Memphis, for Appellees. III, BRIEF: P. Britt Thomas ON Louis J. Walsh, Jr., Harrison, Memphis, Ford & Tennessee, Stephens, J. Matthew Law- Russell, Tennessee, Memphis, rence & for Sullivan, Appellant. Memphis, David M. Tennessee, Appellees. for BATCHELDER, Before: MARTIN and O’MEARA, Judges; District Circuit Judge.* * O'Meara, designation. Michigan, sitting by The Honorable John Corbett United Judge States District for the Eastern District scores, J., and would make

MARTIN, opinion delivered the candidates’ total O’MEARA, D.J., court, in which promotions ranking on the BATCHELDER, J., 416-418), (pp. joined. total scores. dissenting opinion. separate

delivered developed by test was written OPINION private managed by an in- consultant Jones, Dr. Mark psychologist, dustrial who MARTIN, JR., F. Circuit BOYCE City’s also written for developed test Judge. In promotions. developing the 1996 sergeants African-American Dr. relied on Jones the assis- Memphis Department brought Police suit police tance of officers Memphis who City Memphis engaging against the for having expertise identified pro- their discriminatory promotions violation provided fession. Dr. Jones officers Rights Title Act of 1964. VII of Civil found “training good The district court on how to construct a specifically, the cut-off item,” test’s and the officers drafted items test — against African- score—discriminated questions identifying what knowl- remedy American candidates. edge necessary performing identi- phase, promoted sergeants the court fied set of tasks as a lieutenant. Dr. pay. lieutenant and awarded back The questions then had Jones created and attorney also awarded fees critique group of officers them for stayed judgment pend- of its enforcement reliability. Together, they plausibility and ing City The appeal. appeals this percentage minimally estimated the discrimination, finding of district court’s qualified could candidates who answer the remedy, attorney choice and award to questions correctly impose order a fees; sergeants, on con- cross-appeal, cutoff score. stay. test the This case arises from use of I. score, the cutoff which has some relevant twenty In July of one hundred history. Long before the 2000 test written sergeants competed promotion for to lieu- developed, was entered into a those, sixty-three tenant. black Of were Understanding Memorandum of with the white. in- fifty-seven were The police department, in which it agreed promotion- formed the candidates eligible that in for order to be al process compo- consist of four would lieutenant, to a candidate must a achieve (2) (1) test, nents: practical a written passing seventy job score of on written (3) performance exercise evaluations knowledge agreement, Despite test. years, for previous two seniori- Dr. using Jones did not use or recommend ty points. components These four would a cutoff in the 1996 written test. score twenty percent, fifty account percent, Consequently, after the exam was twenty percent, respec- and ten percent, administered, union police officers’ tively, of each candidate’s total score. Jones, Dr. grievance against alleg- filed a Those the written passed who test would ing he violated the Memorandum of proceed components; to the next three on his Understanding. experience Based those pass not would not be invit- who did cutoff Dr. used the score Jones process. ed to continue in failed all candidates who scored also informed the candidates that it would list, order, test. descending seventy create a of the below on the written result, minority minority candi- enough passed not while candidates at a rate aAs nonminority candi- of Dr. Deshon 74.6%. testified that proportion dates in violated is statistically significant. the test. This result 15.4% difference passed dates Thus, Employment Opportunity though Com- even the written not Equal test did rule, four-fifths rule. The four-fifths violate the Commission’s mission’s four-fifths 1607.4(d), rule, pro- found 29 C.F.R. alternative showed that the suc- cess between minority vides: differential nonminority statistically candidates was sex, race, rate for A selection significant. four- group which is less than ethnic (or eighty percent) rate for

fifths After the written test tal- results were highest gen- with the will group rate ranked, lied and the candidates were regarded by en- erally be the Federal top ninety-eight candidates were invited of ad- agencies forcement as evidence promotional continue in the process. four- impact, greater while a than verse ranking by created order regarded generally rate will fifths scores served as the initial de order facto as evi- by agencies Federal enforcement in which candidates passing promot- impact. dence of adverse ed to lieutenant. rate of African-Americans passing ranking Problems resulted from candi- passing than four-fifths of

was less dates on their written test scores. To such ad- rate of nonminorities. avoid accuracy problems These concerned the Dr. Jones the cutoff impact, verse lowered in approximating performance the test points, sixty-six. four Dr. by score by pre- examples and are illustrated two ques- eliminated nine of the test Jones also at trial. exam- sented one faulty; he deemed them he tions because ple, non-minority candidate who received gave answering all credit for candidates the initial score on same the written correctly so that scores questions the plaintiffs, presumably as two of the and hundred-point on a one would remain eligible been to proceed would not have to scale. of promotion process, the next rounds the score, ninety-eight eligible pro- new cutoff was determined to be for

Under the those, passed fifty- City imposed the motion the cutoff candidates test. Of after the forty-seven gave white and were and full for sixty-six one were score of credit faulty questions. all the black. the nine Once promotional of the components process agree All that did not parties this result tallied, this candidate was ranked as were rule; violate the Commission’s four-fifths qualified promotion. the second most however, analyses other statistical —includ- only barely eligible was This candidate T-test and the show an ing the Z-test—did proceed based written test scores to on the T-test, impact. which Under the process, yet in promotion the she received the in scores measures difference mean highest ranking total score the second and nonminority minority and candi- between in candidates. pool of dates, minority was candidates’ mean score example, mean 69.17 while the white candidates’ another thirteen candidates 6.42, 75.59; difference, very order placed ranking was who low the score was retroactively is- plaintiffs’ expert, promoted Dr. Richard had to be and described Z-test, Deshon, per- large. pay which sued back because the candidates’ as Under practical vid- groups, statistical success formances on exercise—a measures across which was late passed component white candidates at a rate 90% eo administered City July to dras- motions on the 2000 tests be process required based — rankings. City required compete promotional to in a new tically adjust process, shown pay sergeant thirteen candidates back to granted those used, received, July seniority points retroactive as of and benefitted from 12, 2000, study disquali- the first of unauthorized pool the date that materials promoted competing promotion- to lieutenant. fied from new was candidates They al process. succeeded and were pro- failed the The four written vided requested relief. See Johnson v. They each scores below test. received City Memphis, Fed.Appx. 128- Sanders, 65; Isabel, 64; Williams, 63; 66— (6th Cir.2003). Parker, 57—and were ranked thus, 111,112, respectively; findings The district made similar to case. proceed primarily not allowed to the subse- the instant Based T— assessments. the statistical and Z— quent promotional Subse- tests, test, filed discrimination the court found that quently, plaintiffs City. unlawfully charges against applied, discriminated against minority candidates. The district Incidentally, this is not the first time court also relied on Dr. Deshon’s testimo- faced discrimination- has ny incapable that the cutoff score was challenges July pro- to distinguishing between candidates who can process. During motion the administra- perform and cannot lieutenant. July the City tion of the was impressed by The district court was com- practical informed exercise evidence, provided examples the two ponent promotion test for to ser- above, described the test’s contents geant compromised had been due to the were not valid and ranking the related study ma- advance release of unauthorized order unreliable. denied that the test terials. compromised trial, had component Following been the bench the court held n to evidentiary the test com- hearing continued administer its on remedies. The pletion. sergeants When members of news me- court found that the were enti- produced copies dia the test while it was tled July lieutenant as of (the administered, being pool still admitted date that the first component that the had practical promoted), compensa- exercise candidates was and City tion compromised. pay been eliminated for back and overtime the July component pro- that from the 2000 amount of the difference the between sala- process weight sergeant motional and increased the ries of from lieutenant that performance of and the the written test date to the date that the court’s order was court it evaluations. entered. The noted that “cannot say Plaintiffs, any, the if which of would Subsequently, African-American promoted the illegal been but for Hispanic-American police officers who and, discrimination,” on the “errfing] side vying sergeant filed Plaintiffs,” promoted all of them. discrimination, alleging claims of that the City intentionally against Subsequently, discriminated the court considered by eliminating practical plaintiffs’ request for attorney them exercise fees. The component increasing weight of that the argued award should be re- promotional pro- the written test after the the sergeants duced because did not suc- They completed. had ceed three of requested cess been their four claims. The sergeants rejection that the who were that its pro- awarded found some of

411 proscribes employment a common set Title VII legal prac theories on plaintiffs’ monetary judg- not form facts did affect the tices are “fair in but discrimina equitable requested. tory or in operation.” Griggs ment remedies v. Duke Power Co., sergeants gained 424, 431, 849, “excellent Because 401 91 U.S. S.Ct. 28 results,” requested (1971). all the court awarded L.Ed.2d 158 Discrimination occurs expenses. fees and an employer particular when “uses a em ployment practice that a disparate causes stayed Finally, the district court execu- race, color, impact on the religion, basis appeal to pending its order tion of sex, or national origin fails to demon “emergency sergeants filed Court. challenged practice strate is stay. appealing This Court motion” related for position question 23, 2003, September the motion on denied 42 necessity....” consistent with business considered in the traditional factors citing 2000e-2(k)(l)(A)(I). U.S.C. determining stay that the stay appeals and an abuse of was not discretion. Supreme has devised three-part burden-shifting test to deter

II. disparate mine whether an unlawful im Title VII A. any particular exists in pact case. Albe This Court’s standard of review 405, Paper Moody, marle 422 Co. v. U.S. case in a Title VII discrimination is nar 425, (1975). 2362, 95 45 L.Ed.2d S.Ct. 280 Under Federal Rule Civil Proce row. First, the must plaintiff prima establish a 52(a), findings a district court’s dure i.e., plain facie case of discrimination — clearly unless fact should be reversed impact tiff must establish that Cleveland, Zamlen erroneous. v. plaintiff succeeds, If has occurred. (6th Cir.1990), 209, 217 cert. de employer protocol must show nied, 499 S.Ct. 113 U.S. 111 question relationship has “a manifest (1991). 444 error Clear will lie L.Ed.2d employment”' so-called “business —-the court is left reviewing when 401 at 91 justification.” Griggs, U.S. definite, firm that a mistake conviction succeeds, employer If 849. S.Ct. v. has made. Anderson Bes been must then show other tests plaintiff N.C., 564, 573, City, 105 semer U.S. would protocols serve the em selection (citing 84 L.Ed.2d 518 S.Ct. un ployer’s creating interest without Gypsum States States United United discriminatory effect. Albe desirable Co., 364, 395, S.Ct. 333 U.S. marle, 425, 432, (1948)). The issue not wheth L.Ed. con er the district court reached best i. Prima Facie Case clusion, but whether evidence before A case is supports prima court’s find face established the court the district (1) specific a em Heights Congress Hilltop plaintiff v. identifies ings. Cmty. when: Cir.1985), Inc., (6th 135, challenged; be and practice 140 to Realty, ployment 774 F.2d denied, 1019, 106 through analysis statistical rt. 475 S.Ct. relevant U.S. ce (1986). Also, challenged 1206, practice that the has an proves 89 L.Ed.2d 318 a impact protected group. on on findings court’s the credi adverse district it v. Health and Hu Dept. the witnesses are entitled Johnson bility of before of Cir.1994) (6th Servs., 45, man 48 Wooldridge deference on 30 F.3d great appeal. to Co., 540, (citing Scales v. & 925 Indus. F.2d J.C. Corp., v. Marlene Bradford (6th Cir.1991)). Cir.1989). City (6th F.2d 907-08 The applicants a on eouraged disproportionately to establish plaintiffs that failed argues race, sex, origin.” grounds and ethnic case failed facie because prima 1607.4(D). § 29 C.F.R. use cutoff score re- that the show minorities impact in an adverse on sulted City The that the dis argues also Equal Employment Opportunity under complied trict court erred because the test City four-fifths rule. Commission’s rule, a com with the four-fifths and test’s only that we consider the contends should definitively rule pliance with four-fifths measure, rule, no other four-fifths and impact. the absence of adverse establishes has impact whether an adverse determine also This is incorrect. The Commission’s Thus, ques- the determinative occurred. regulations compliance make explicit may is look to in this case whether we tion four-fifths with the rule establishes T— analyses statistical alternative —the analysis will that the four-fifths statistical impact. an adverse and Z—tests—to find impact. not of an adverse serve evidence “a greater The rule states: than four-fifths that caselaw contends generally regarded rate not will be forbids reliance on alternative our Circuit agencies Federal enforcement as evidence To analyses in Title VII cases. statistical impact.” C.F.R. contrary, require only we 1607.4(D). Thus, a notwithstanding analyses “relevant.” John statistical be rule, compliance test’s with the four-fifths son, Also, Supreme may analyses other still reveal an adverse case-by- a approved has the use of Court impact. approach, recognizing that statistics case ... variety in infinite their “come Indeed, Supreme when Court depends all the sur upon usefulness plaintiff may rely solely that a stated rounding facts circumstances.” prima evidence statistical to establish States, v. 431 U.S. Teamsters United disparate Cove impact, facie case Wards 339-40, 1843, 52 L.Ed.2d 396 97 S.Ct. Atonio, Packing Co. v. 656- 490 U.S. (1977). Moreover, commentary (1989), 57, 109 S.Ct. L.Ed.2d 733 regulations excep allows say Commission’s what evi it did kind of statistical dif tions to the four-fifths rule: “Smaller dence should be on. Neither the relied may selection rate nonetheless nor ever Supreme ferences this Court has they are impact, constitute adverse where in Title VII limited choices practical analysis in both significant involving statistical cases statistical Moreover, way.1 have dis- for sta grateful terms or where user’s actions we are impact, its and this 1. We are aware that this Court limited Court did not consider analysis case of to the rule in the question rely four-fifths of whether it would on alterna- Akron, (6th Black v. 831 F.2d 131 impact in tive to find an adverse Cir.1987). City suggests analysis that our situations where four-fifths rule was not ruling dictates in this case. in Black our question presented That is violated. However, Black is the situation distinct from case, inapplicable. Black this Further- here, and, moreover, ruling case in that more, Supreme holding in Ward Court’s by subsequent rulings been offset has Cove, 656-57, 490 U.S. at Black, Supreme Court and this Circuit. Court's decision in N.A.A.C.P. size, argued sample a small claimants Cir.1989), (6th Mansfield, imply 866 F.2d 162 black white can- where fewer candidates than analyses may consid- other statistical passed a a discriminato- didates revealed ered; or, least, very at the these decisions ry impact despite comportment the test's we to the four- make clear that are not limited rule. no alter- four-fifths Black involved fifths rule. showing the adverse native statistical

413 beyond analysis four-fifths rule drawn that tistics must be the cutoff score mea- to look to prefer qualifications. because we the sum sures minimal The Guide- to make a statistical evidence decision provide lines that cutoff “where scores are v. used, they these kinds of cases. See N.A.A.C.P. should normally be set so as to (6th City Mansfield, F.2d 162 Cir. be reasonable and consistent with normal 1989) (approving finding of district court expectations of acceptable proficiency the sum of evi plaintiffs’ that statistical within the work force.” 29 C.F.R. 1607.5(H). a marginally § dence “demonstrated statisti The district court found that cally significant disproportionate impact nothing “the cutoff score was more than blacks”). arbitrary decision did not measure Here, qualifications.” minimal the district

ii. Business Justification emphasized court that the cutoff score was by never validated the methods Supreme has established suggested in the Uniform Guidelines. plaintiff presented facie prima once has case, a production burden of lies with the Dr. was Jones asked whether the cutoff employer justifi to demonstrate business following score was validated pas- challenged cation for employment sage: Cove, 669-60, practice. 490 U.S. at Wards Q: to, didn’t attempt quote, “You vali- Equal Employment 109 S.Ct. 2116. The score, you?” the cut date did Opportunity developed has Commission A: “No.” Guidelines, 1607.1-.16, Uniform C.F.R. Dr. Jones also testified he did not establishing validity for tests these want to use a cutoff score and saw no procedures. According to the Guide one; value in implementing explained he lines, employers may use types three only reason he used one was employee studies validate selection the union to. because forced him content-, construct-, procedure: or criteri validity on-related studies. 29 C.F.R. The court also the test for faulted meas- 1607.5(A); see also Zamlen v. only (job uring component one knowledge) (6th Cleveland, Cir. failing job to test the entire domain. 1990). Jones, by This fact was confirmed Dr. who designed testified he the written test City attempts justify job knowledge, sought to test by the written test content-related validi major the “other measure elements of the ty. validity A if test will content job” components remaining relationship there is a direct between the process. district court job test contents and the contents. effect, ranking found that orders Co., v. Ford Motor Williams F.3d based on had no value (6th Cir.1999). 538-43 To establish wheth because could not trusted to be employment job- er contents of an test are job performance. related to actual related, employer must show that the “predictive significantly particularly contents are of or im- The district important by pressed correlated with elements of the two examples illustrated at comprise Referring work behavior which or are rele trial. to the nonmi- jobs vant to the candi nority which candidate who failed the written Williams, being are was ultimately dates evaluated.” test but ranked second (internal omitted). highest promotion, F.3d citations in line for the court score, validate a piece clearly To cutoff the inference stated that of evidence “[t]his 414 (1867). 99, 18 L.Ed. The Congression that the written test 752 any inference

dispels Record, (1972), job performance.” Cong. al Rec. 7168 approximated 118 clear: makes there is clear evidence agree that We did from the written test that the scores of are [remedy provisions VII] The Title potential a candidate’s approximate to give intended the courts wide discre- did not The district court performance. exercising equitable powers tion their to finding that the test its abuse discretion complete possi- the fashion most relief justification. had no business dealing present ble. with section to City fails meet its Because the burden that the 706(g) courts have stressed the Title VII production, we conclude scope relief under section of analysis in case. The district Act is intended to make the victims correctly ruled that whole, unlawful and that discrimination Title violated VII. objective of this rests not attainment particu- the elimination of the upon Pay and Back B. Promotion practice lar employment unlawful com- employer finds that an If a district court of, plained requires per- also but employment unlawful engaged in an has aggrieved by sons consequences authorizes award practice, Title VII employment prac- effects the unlawful See other retroactive relief. pay back be, possible, tice so far as restored to a 5(g)(1); see also U.S.C. they Loef position 2000e— where would have been 549, 558, Frank, 108 S.Ct. v. 486 U.S. it not unlawful fler for the discrimina- (1988). 1965, 100 As the Su L.Ed.2d tion. Loeffler, preme explained Congress’ vesting variety purpose by authorized Title VII “is pay back award “discretionary” powers the courts was Congress’ intent a manifestation of possible “to make ‘fashion(ing)(of) ‘persons injuries make whole for suffered ” ” complete possible.’ Albe most relief (quot through past discrimination.’ Id. (cit marle, U.S. at S.Ct. 1733 Albemarle, 422 U.S. at 95 S.Ct. ing 7168). ing Cong. Rec. 2362). The court found that the district racial is Where discrimination complete possible most relief includes (district) found, merely “the court has not pay both award of back duty a decree power but render 12, 2000, July retroactive because as possible as which will far eliminate so promoted plaintiffs would have been discriminatory past effects of the as well received lieutenants’ salaries but for the like in the future.” as bar discrimination agree unlawful discrimination. We States, Louisiana v. United 380 U.S. court, despite City’s argu the district (1965). 154, 85 13 L.Ed.2d 709 contrary. City argues ments to the is, general wrong rule that when a “The pro are not entitled to done, gives has been and the law reme pay motion or back because there is no shall to the dy, equal be compensation assurance, sergeants passed had the even injury. latter the standard promotional process, would which the is to be measured. former promoted have been to lieutenant because injured placed near as party is be, police operates often be may department situation he would budget may if had com not have hired all occupied wrong not been low However, Hoppock, passing mitted.” candidates. Wicker Wall.

415 support accepted plaintiffs’ lawyer’s evidence to court the present not does rate, hour, contention, and, hourly failing per thus to meet his contri- $250 this bution, hours, as burden, Wooldridge, at 590.7 reasonable. The see 875 F.2d its City argues appeal on that the award was (holding that defendant bears burden 547 unreasonable and should be reduced. “by of preponderance demonstrating of factors other than the evidence that the We review a district court’s award deci- discrimination caused the condemned attorney of fees for abuse of discretion. dis- complains”), claimant we sion which Fikse, (6th 633, v. Cramblit 634 argument. miss its Cir.1994). affirm the We unless court’s that none also contends ruling is on an erroneous view of the based high enough have ranked plaintiffs would clearly law or on erroneous assessment even if no cutoff score eligibility on its list of the Id. record. Because same stan However, as the had been used. district “generally applicable dards are in all cases stated, correctly “ranking court Congress which has authorized an ” only job knowledge had invalid because fees ‘prevailing party,’ award of to a Eckerhart, in the not been tested Hensley v. 461 U.S. 433 n. domain, ... job orderings (1983), the rank entire 40 L.Ed.2d we and should not rely precedents involving attorney were indeed unreliable on fees regard consider- have been used without whether involved or- Title or some federal Relying ations. on same rank VII other statute. juncture list at would be incon- dered point The trial initial court’s judgment with the sistent Court’s when departure, calculating reasonable at liability phase.” fees, torney is the determination of the fee Third, contends that the statis- “lodestar,” applicant’s proven which is the by on tical relied district reasonably expended number of hours on only two of four court showed that by attorney, multiplied by the case eligible the group ranked within plaintiffs hourly Hensley, rate. reasonable a promotion. Because the district 103 S.Ct. 1933. The reasonable reasonably concluded earlier that here, ness of the 590.7—and rate— hours — could approximate perform- by hour—-is determined consider per $250 justification, the (1) ance and had no business factors: and labor re ing twelve time successfully argue cannot (2) now novelty difficulty of the quired; eligible (3) two of four could be questions presented; the skill needed (4) on the tests results. properly; perform legal service preclusion employment the attor Attorney C. Fees (5) case; ney acceptance due to of the (6) fee; court, in provides customary Title that “the its whether the fee is fixed VII (7) discretion, party time and limitations im may prevailing contingent; allow circumstances; litigation posed by Title ... a the client or the VII] under reason- [in (8) ... attorney’s part of the the amount involved and the results able fee (9) 2000e-5(k). obtained; § experience, reputation, The ser- costs.” U.S.C. (10) attorney attorneys; of the the “un geants requested ability award of (11) fees, case; costs, desirability” to 42 expenses pursuant nature (1994). 2000e-5(k) length professional relationship U.S.C. district client; and in “similar requested: court awarded amount fees with the awards Reed, $184,461 $36,786. 471-72 179 F.3d at n. 3 and costs cases.” Georgia just arguments Ex success one of the Highway (citing Johnson (5th Inc., Cir. This hoped-for F.2d 717-19 would achieve the results. press, 1974)). rare exceptional is not one of those justified. in which a reduced award is cases argues the rate City

The that on adjusted downward based should be Stay D. have which could numerous tasks listed City’s mo- granted The district court hourly a rate. performed at lower been stay judgment tion to of the enforcement however, practi a lawyer, Plaintiffs’ is solo sergeants pending appeal. appeal operate not the assis tioner who does with stay, upon entry which this deci- expires support paralegals staff. tance sion, they arguing being irrepara- that are aver that their counsel’s City argues also bly harmed in the meantime. hourly is reasonable age rate of more $150 panel previ- Another of this Court has However, counsel’s rate. plaintiffs’ than ously appeal stay. plaintiffs’ denied City’s hourly counsel’s rate was lead adopt and its panel’s We that decision $270; make no that City argument can reasoning Septem- in its stated order of comparison. on is based 23, 2003. ber “ in de most critical factor’ ‘[T]he of a fee termining the reasonableness III. ” degree ‘is of success obtained.’ award reasons, For the we affirm the foregoing 114, U.S. Hobby, Farrar v. respects. in judgment district court’s all (quoting 121 L.Ed.2d S.Ct. 1933). at Hensley, 461 U.S. 103 S.Ct. BATCHELDER, Judge, Circuit plaintiff has obtained excellent Where dissenting. results, attorney his should recover full I plaintiffs dissent because fee; plaintiff if a obtains compensatory failed to that use of demonstrate success, “limited the district court should score, required applicants a cut-off which only is award that amount of fees questions to answer the 100 on the 66 of to the success ob reasonable relation portion written of the examination correct- 435, 103 Hensley, tained.” at ly, impact had on African- 1933. This has stated At the Americans in violation of Title VII. attorney applied fees is to be reduction Union, insistence of the Police exceptional cases where rare and agreed to a cut-off on the impose score requires it. specific evidence record examination, part ap- which Adcoch-Ladd, 349-50. plicants required would be to achieve to be contends that the award order considered for by fifty-three percent administering lieutenant. should be reduced Before Jones, plaintiffs psychologist not on three Dr. Mark an industrial because did succeed by the City of their The district court hired to create the promotional four claims. exam, appropriate set cut-off score at With a found no reduction was Afri- pass obtained excellent cut-off score of rate of because the than Indeed, results. not have can-American candidates was less could achieved A of alter 45% of order to achieve proffer better results. whites. arguments justify reducing compliance native “four-fifths does with EEOC’s award, rule,” arguments says which that a rate especially where the selection made facts and race which is more than four-fifths on a common set of group rate for the highest tion. Id. at 134. The Majority distin- Black, generally regarded rate will not be as evi- guishes which says it “did not con- impact, dence of adverse see 29 C.F.R. question sider the of whether it rely would (2004), 1607.4D adopted Dr. analyses alternative to find an adverse Jones’s recommendation that the cut-off impact.” That Black did not consider al- score be lowered to 66. With a cut-off ternative forms of analyses statistical pass score of rate of African-Amer- exactly point: courts in this Circuit do icans was 83.4% of that of whites. not consider alternative when the employment practice in question satisfies fact Despite the that the written exam rule. Majority’s opin- four-fifths yielded a pass rate that satisfied the four- cite, ion my does not research failed to rule, Majority, relying fifths on alter- uncover, any cases from this Circuit where analyses native forms of statistical such as the court looked to other analy- statistical Z-Tests, the T— and concludes that ses to find an adverse impact when the of a minimum use cut-off score cre- employment results of the ques- action in impact ates an adverse in violation of Title tion satisfied the four-fifths rule. view, however, In my VII. the law of this requires Circuit that when the results of The Majority’s misunderstanding employment action in question satisfy Black’s rule is evidenced its citation to rule, the four-fifths courts are not to look City Mansfield, N.A.A.C.P v. 866 F.2d 162 (6th Cir.1989), for other analyses might statistical which it says “implies] that reveal an impact. adverse Black v. other statistical may be consid- Akron, (6th Cir.1987) ered,” pub- a support in proposition lished Circuit Sixth case that makes a Black “has been by subsequent offset rul- appearance cameo in the first footnote of ings.” court did of Mansfield Majority’s opinion, analyzed whether indeed consider analy- alternative forms of an ses, examination that satisfied the which, four- such “chi-square,” as the ac- fifths rule still created an impact, cording to plaintiffs witness, expert and addressed the concerns of suggested some com- “differing examination mentators and courts that pass the four-fifths ratios for whites and blacks on the rule is not a useful indicator of adverse police test were not attributable to impact when a small sample employees chance.” Id. at 167. But the results of being is tested. The court stated that “the the test at issue in that satisfy case did not small sample might rule, size of the abe reason City Mansfield, four-fifths to therefore, rule that there was no prima case permit does not the conclusion facie violated, even if the rule were it but is that we are free rely 4/5 to seek out and on certainly not a finding prima reason for alternative forms of analysis statistical case when the rule is not violat- impact find an adverse in cases where the 4/5 facie ed” and held that the plaintiffs’ statistical four-fifths rule has otherwise been satis- evidence not suggestive of discrimina- fied.1 tion, Despite Majority the fact that the notes that Wards Cove does not "offset” Black's Co., Atonio, Packing Wards Cove Inc. v. holding. City Mansfield, As in the Court in 642, 656-57, 109 S.Ct. plaintiffs’ Wards Cove held that the statistical say L.Ed.2d 733 "did not what kind of disparate impact evidence did not establish a on,” statistical evidence should be relied it in violation of Title VII. 490 U.S. at argues "implfies] that Wards Cove that other S.Ct. 2115. Wards Cove is not a case where analyses may statistical be considered.” the Court relied on alternative statistical anal- assuming supports Even proposi- that it this (74.6%) and 47 out of 63 African-Ameri- past years, employers like

For the passed. cans I do not think that this small City Memphis relied terms,” disparity significant “practical rule wheth- line four-fifths to assess bright 1607.4(D) (“Smaller practices comport 29 C.F.R. differ- their see employment er may As of to- ences rate nonetheless requirements. Title VII’s selection however, impact, no constitute adverse where are day, employers Circuit significant practical that their scru- in both statistical and longer have that assurance ”) therefore, terms ... I do not believe adherence to four-fifths rule pulous impact cognizable it is under insulate them from adverse Title VII. What’s will more, opinion absolutely nothing in Majority’s opens there is suit. City purposeful- suggest to file Title VII suit record door ly against flavor of discriminated on the newest statistical African-Ameri- month, ignored absolutely no cans risk that the written test provides but To impact. tests would cause an adverse guidance as to which are be used contrary, retained an industrial assessing employment prac- whether psychologist to create a that would not impact, tice results in an adverse how impact cause an adverse and even lowered are to or to deter- applied, tests be how the cut-off more score ensure that racial significance. mine their statistical minorities, have who otherwise would considering Even the alternative statisti- passed remained consideration on, Majority cal relies evidence to lieutenant. disparity between scores white I Accordingly, would reverse the deci- applicants, African-American while sion of the the Ap- district because statistically does not rise to significant, pellees failed to that the City’s have show level of a Title VII infraction. Watson of a use cut-off score created an adverse Trust, v. Bank & 487 U.S. Forth Worth impact violation of Title VII. 977, 994-95, 2777, 101 L.Ed.2d 108 S.Ct. Supreme Court indicated disparity that a must be substantial: formulations, which have never been

“[o]ur any rigid mathematical

framed terms of

formula, consistently stressed that sufficiently must disparities

statistical Jeffrey MCKINLEY, Plaintiff- raise substantial that such an infer- Appellant, in Albe- ence causation.” v. that, Moody Paper marle Co. stated al., MANSFIELD, CITY et OF impact, complain- establish Defendants-Appellees. party “that tests in ing must show No. 03-4258. pro- question applicants select for hire pattern motion in a dif- significantly racial Appeals, United States Court of pool applicants.” ferent from that of Sixth Circuit. Argued: Oct. 2004. (1975). bar, In the L.Ed.2d 280 case and Filed: April Decided (81.7%) passed applicants out when a cut-off score of 66 (89.5%) of 57 imposed; 51 out whites yses impact to find when the four- fifths rule was otherwise satisfied. an adverse

Case Details

Case Name: Isabel v. City of Memphis

Court Name: Court of Appeals for the Sixth Circuit

Date Published: Apr 11, 2005

Citation: 404 F.3d 404

Docket Number: 03-5912, 03-5914, 03-6011, 03-6231

Court Abbreviation: 6th Cir.