NICHOLS, Associate Judge.
This employment discrimination case comes before the court on plaintiffs’ appeal of the June 30, 1980, judgment of the United States District Court for the Northern District of Illinois. The Honorable George N. Leighton held that defendants Rush-Presbyterian-St. Luke’s Medical Center, James A. Campbell, and Charles A. Freeman were entitled to judgment in their favor against plaintiffs as to all individual and class claims. We affirm.
In the district court plaintiffs Magdalena Garcia, Fernando Romero, and Victoria Perez, who were Latinos, sued on their own behalf and on behalf of other Latinos similarly situated in a class action pursuant to Rule 23(a) and (b)(2) of the Federal Rules of Civil Procedure. Plaintiffs’ complaint alleged claims for relief under Title VII of the Civil Rights Act of 1964, 42 U.S.C. §§ 2000e et seq., and 42 U.S.C. § 1981.
The district court determined that the evidence did not establish by the standard of proof required by law that the plaintiffs individually or as a class were subjected to any disparate treatment or were the object of any disparate impact because of their race, color, national origin, or because any of them were Latinos. The district court’s definition of “Latino” was “any Spanish surnamed person or individual of Hispanic ancestry.” But he excluded Filipinos. It is asserted and denied that the district court was confused about the composition of the aggrieved class. This was potentially important because of the importance of statistical evidence and the noninclusion of blacks, the largest racial minority in the community, as members of the grievant cláss. In light of our other conclusions, clearing up any semantic confusion in this regard is not needed.
The district court issued a detailed and well reasoned decision in which it determined that defendant Rush’s hiring and transfer practices were and had always been racially neutral. In rendering its decision the court considered extensive evidence submitted during 18 days of trial without a jury, from May 27 to June 23, 1980. At trial, plaintiffs called 13 witnesses, used excerpts from four depositions, and offered and had received in evidence 70 exhibits. Defendants called 11 witnesses, offered and received in evidence 102 exhibits and then on rebuttal plaintiffs offered three additional exhibits. The reasoning of the court will be set forth herein where it has been challenged by plaintiffs.
Succinctly, plaintiffs claims were and are for alleged employment discrimination. More specifically the claims of the individual plaintiffs against defendants below were:
1. Discrimination against plaintiff Magdalena Garcia by harassment, discipline, and discharge because of her race or national origin in violation of Section 703 of Title VII, 42 U.S.C. § 2000e-2.
2. Discrimination against plaintiff Victoria Perez in denying her transfer to the position of Lab Liaison Technician at Rush in violation of Title VII and 42 U.S.C. § 1981. She also alleged discrimination and retaliation by her discharge on August 17, 1976, in violation of Sections 703 and 704(a) of Title VII, 42 U.S.C. § 2000e-2 and § 2000e-3(a) and in violation of 42 U.S.C. § 1981, but at the conclusion of plaintiffs’ evidence, plaintiffs’ counsel withdrew this discharge claim under both statutes.
3. Purposeful and intentional discrimination against plaintiff Fernando Romero by Rush’s refusal to hire him because he was a Latino, in violation of 42 U.S.C. § 1981.
Plaintiffs do not appeal from the trial court’s findings of fact and conclusions of [1220]*1220law concerning any of the individual claims of plaintiffs Garcia and Perez, and no issues respecting them remain in the case. Only the legal standard applied to the facts regarding plaintiff Romero are challenged.
The class claims against defendants under Title VII of the Civil Rights Act of 1964 and under 42 U.S.C. § 1981 were:
1. Discrimination in refusing to hire qualified Latinos.
2. Making discriminatory assignments to Latino employees.
3. Utilizing discriminatory performance standards for Latino employees with respect to promotion, assignment, and tenure.
4. Discriminating against Latino employees by restraining and coercing them in the exercise of their rights to complain of discriminatory employment practices.
Plaintiffs do not challenge the trial court’s findings and conclusions that defendant did not discriminate against plaintiffs in assignments, or in utilizing performance standards for promotion, assignment, and tenure, or by restraining and coercing plaintiffs in the exercise of the rights to complain of discriminatory practices. The issues brought forward on appeal are comparatively few. Plaintiffs challenge the district court findings regarding only defendant’s hiring practices, and plaintiff Romero’s claim under 42 U.S.C. § 1981.
I
Initially though, plaintiffs argue that this case should be reversed and remanded because the trial judge adopted the proposed findings and conclusions submitted by defendants. It does appear he used in haec verba many proposed findings that defendants submitted, or altered and rearranged them only in rather immaterial particulars. But a party cannot be penalized for the persuasive nature of his submissions. We do prefer to see findings that reflect an effort of composition by the trial judge. They furnish evidence that he really worked over and analyzed the fact issues. He must avoid giving the impression that he first decided who should win, and then made the findings that would best support the favored litigant on appellate review. Here, however, the record does contain other evidence of the concern Judge Leighton displayed with the fact issues. For instance, he requested additional evidence to verify the probativeness of defendants’ statistical evidence. See infra.
As plaintiffs correctly point out, this court has criticized the practice of adopting the prevailing party’s findings verbatim and without change. But such adoption does not invalidate the findings, although they may therefore be more critically examined. Photovest Corp. v. Fotomat Corp., 606 F.2d 704, 731 (7th Cir. 1979); and FS Services, Inc. v. Custom Farm Services, Inc., 471 F.2d 671, 676 (7th Cir. 1972). The proper standard of appellate review of the trial court’s findings of fact is that the findings will not be disturbed unless “clearly erroneous.” Fed.R.Civ.P. 52(a), Title 28 U.S.C.; Hanock v. Eck, 183 F.2d 632, 635 (7th Cir. 1950) and Tornello v. Deligiannis Brothers, Inc., 180 F.2d 553 (7th Cir. 1950).
Plaintiffs have failed to establish any error on the part of the district court. Rather than showing that the district court findings were clearly erroneous as they must on appeal, with specifics as to what findings were erroneous and how, plaintiffs set forth summaries of the evidence they submitted below (complete with percentage statistics and a graph along with what is purported to be a summary of defendants’ data and expert evidence). But the test on appeal is not which party had the better evidence below. Plaintiffs’ position therefore cannot prevail. As this court explained in a previous employment discrimination case with respect to findings of expert statistical evidence—
On appeal the defendants seek to challenge the district court’s conclusions by relitigating the probativeness of the evi[1221]*1221dence introduced and rejected at trial. The findings of the trial court with respect to this evidence — statistical studies and the testimony of expert witnesses— are perfect examples of subsidiary facts to which the clearly erroneous standard applies. As this court has observed, “[t]he resolution of such evidentiary conflicts is the precise function for which our trial courts sit [and it] is only necessary for us to determine on review whether the findings supporting the judgment have an evidentiary basis. * * *. [citations omitted.] We should reverse the district court’s findings only if, with due deference to the trial judge’s resolution of conflicting evidence and to his determination of credibility, we are left with the definite and firm conviction that a mistake has been committed. * * * [citations omitted.]
United States v. City of Chicago, 549 F.2d 415, 429-30 (7th Cir.), cert. denied sub nom., Adams v. City of Chicago, 434 U.S. 875, 98 S.Ct. 225, 54 L.Ed.2d 155 (1977). The district court’s findings now before us are supported by substantial record evidence.
Plaintiffs do not contest the vast majority of the trial judge’s findings and conclusions, yet complain that the trial judge did not consider the case with a “disinterested mind.” This argument is weak. Plaintiffs cite two cases as “similar” to their own, Equal Employment Opportunity Commission v. United Virginia Bank/Seaboard National, 555 F.2d 403 (4th Cir. 1977); United States v. Commonwealth of Virginia, 569 F.2d 1300 (4th Cir. 1978). But these cases do not support plaintiffs’ position. In Equal Employment Opportunity Commission v. United Virginia Bank, supra, for instance, the appeals court reversed the district court because the district court’s findings of fact were phrased in “broad conclusory terms and did not include any subsidiary findings which would give appropriate support to the court’s conclusory findings.” 555 F.2d at 405. In addition, the district court in that case made no analysis of statistical information or the weight to be accorded it, it engaged in no inquiry into the racial pattern of the community population, and it did not review the employment records of defendant. In United States v. Commonwealth of Virginia, supra, the district court issued no written findings of fact at all. As will be shown herein, there is not even a remote similarity between the findings and conclusions written and adopted by Judge Leighton and those in the two Fourth Circuit cases cited by plaintiffs.
Plaintiffs argue also that the clearly erroneous standard is not workable here where the trial court’s findings are conclusory and argumentative. The excerpted findings of the trial court that follow, demonstrate conclusively that the trial judge carefully analyzed the statistical information submitted to him, critically reviewed the expert reports of both sides, and then made subsidiary findings to support his final determination.
II A
The relevant findings below respecting class claims which are challenged on appeal follow.
Rush is a not-for-profit corporation which operates both a medical university and an 864-bed tertiary care hospital. A “tertiary care hospital” is a highly specialized medical facility [which provides] * * * —sophisticated diagnostic services and treatment which normally can[not] be provided by secondary and primary care facilities. [It is located in Chicago.]
Rush’s delivery of health care, research activities, and education of physicians, nurses and health scientists and administrators requires a highly specialized level of skill and training among most employees in its workforce.
Rejection by Rush of those Latinos who, the evidence shows, were not hired, resulted from the two most common [non-discriminatory] reasons on which an employer can rely to reject a job applicant: an absolute [1222]*1222or relative lack of qualifications. [In 1979 the faculty of the various Rush Colleges and its registered nurses were 1,800 persons and 28 percent of Rush’s workforce. The managerial, professional, and technical employees that year were 3,994 persons, 62 percent of the workforce.]
Rush requires employees in nearly all job classifications to speak and read English in some fashion. Plaintiffs contend that this requirement results in unlawful discrimination against Latinos. They did not produce evidence, nor is there any in the record to show that Latinos were excluded from Rush’s workforce at a greater rate than persons of other national origins, by virtue of such requirement standing alone, or in connection with other employee selection procedures. The requirement that an employee speak and read English in some fashion did not have an adverse impact on Latinos; nor were Latinos treated differently than all other persons with regard to the requirement. The ability to speak and read some English is a necessary, job-related requirement for virtually every job in this highly sophisticated medical care institution.
This court further comments: Thus some facility in English is at Rush a Bona Fide Occupational Qualification (BFOQ). The question of language in a large modern hospital in an urban area is, if we may interject some judicial notice, difficult, and anyone who has been so lucky or unlucky as to be a patient in such an institution probably has had some personal experience with it. Such an institution is apt to be somewhat of a Tower of Babel. We would suppose that English is most likely to be the common language of a majority of patients and staff alike, and, therefore, a deficiency in English is the language deficiency most likely to be troublesome with an employee of a hospital located well in the interior of a supposedly English speaking nation. There would be many patients not fluent in English, who would need someone able to converse with them in their own native language. Plaintiffs allude to this question in their brief without drawing any clear conclusion from it. A random introduction of non-English speaking persons into the hospital staff would probably not contribute to meeting this need, and it is not suggested that it is required by Title VII, or that in general Title VII requires hospitals to be indifferent to the communication needs of their patients, or any of them. At any rate, the language qualification is clearly within the discretion of the hospital authorities and plaintiffs do not argue otherwise. It impinges equally on other foreign language users as on the Hispanies.
Returning to the findings—
Rush has had, and has maintained, an equal employment opportunity policy in form and substance. Long before this controversy with the plaintiffs, it has at all relevant times, publicized its equal employment opportunity policy to the public generally, to its supervisory executives and all employees through employee handbooks, bulletins, and personnel policy and procedure manuals. This policy has been published and disseminated for many years in the Medical Center’s employee newsletters, orientation programs and seminars. Rush has publicized and implemented its grievance procedures for both union and nonunion personnel as a means of correcting any supervisory action shown to be unfair or inconsistent with the Center’s declared equal employment opportunity policy. Union contracts covering approximately 1,000 employees contain clauses prohibiting discrimination on the basis of race, color, national origin, sex, and religion as covered by Title VII. Latino employees, including those mentioned by Garcia and the witness Cruz, used the grievance procedures; and, in several cases, obtained favorable dispositions thereby. These facts, shown by unrebutted evidence, undermine the unproved allegations in plaintiffs’ complaint that Rush engaged in a pattern or practice of discrimination against Latinos employed at the Medical Center.
[1223]*1223The affirmative action program is consistent with the declared policy of equal employment opportunity that Rush has administered for many years. Consistent with its declared policy, and its affirmative action programs, Rush has made efforts to recruit and increase the availability of Latinos and other minorities for its workforce, both before and after this suit was filed. These efforts include the Inroads Program in 1973 and 1974, the Boy Scout Explorer Program from 1971 to 1974, the Youth Motivation Program from 1974 to the date of the trial in this case, Health Care Career Opportunity Program from 1978 to the date of trial, the Secretarial and Clerical Opportunities Program from 1975 to the date of trial, and other continuing CETA programs for students and adults. Some participants in these programs have become full time employees of Rush. For many years prior to and after the filing of this complaint, Rush recruited employees through predominantly Latino referral agencies such as Gads Hill and Ser. After this suit was filed, Rush voluntarily participated in a formal recruitment program specifically directed to Latinos. Rush also has extensively recruited at schools and universities whose student bodies are predominantly Latinos and Negroes. Rush advertises its equal opportunity policy, and its job openings in Spanish periodicals. It maintains a tuition reimbursement and continuing education program for all employees which includes courses in English and Spanish. At all relevant times, Rush has provided health care programs for the benefit of Latinos and other minorities in the surrounding west side Chicago area; and it has published a Spanish language patient handbook and a brochure on its community relations program in Spanish. Rush has actively responded to the special needs of Latinos in its workforce and in the community; this response is inconsistent with the allegations contained in plaintiffs’ complaint that Rush’s employment practices were motivated by discriminatory purpose.
The testimony of the individual plaintiffs does not support the class claims because none identified or was able to testify about any hiring, employment, assignment, promotion, transfer, or discharge policy or practice of Rush which disparately treated or disparately impacted Latinos because of their national origin under Title VII, or their race under § 1981, or because they had engaged in any protected activity in opposition to allegedly discriminatory employment practice by Rush.
Plaintiffs called two expert witnesses: Dr. William G. Fischer, a clinical psychologist; and Dr. Wayne J. Villemez, a University of Illinois Assistant Professor in the Department of Sociology. It was Dr. Fischer’s opinion that the qualifications or requirements set forth in certain Rush job descriptions were deficient because some were too vague, others were too high for the job, and still others permitted subjectivity and potential for bias because no qualifications as such were contained thereon. Dr. Fischer could not cite any authority, or authorities in the context of a Title VII or § 1981 discrimination issue to support any of his opinions or analyses with respect to Rush’s job descriptions. He reviewed 950 to 1,000 Rush job descriptions and compared them to four published works, at least two of which contained narrative descriptions of potentially similar jobs. He admitted on cross-examination that he had not consulted any of the three most current sources until after he had already formed his opinions and made his characterizations based on a 1966 supplement to the United States Department of Labor’s Dictionary of Occupational Titles, a publication which contains no comparative descriptions at all. Dr. Fischer’s testimony was neither meaningful nor probative of plaintiffs’ claims.
Dr. Villemez failed to establish any foundation that the definitions of availability used in his reports measured the availability of Latinos who were qualified for the jobs in question at Rush. He conceded that his availability estimates were based on personal assumptions only, and that he did not know whether the census data he used in [1224]*1224estimating Latino availability for jobs at Rush were based on comparable jobs in terms of skill requirements and qualifications. Dr. Villemez chose national employment data for the hospital industry as the most appropriate “proxy population” for the professional and technical jobs at Rush; he did not know whether the jobs designated as professional and technical at Rush were comparable in terms of skill requirements and qualifications.
Defendants’ expert was Dr. George R. Neumann, Assistant Professor, Graduate School of Business, University of Chicago. Dr. Neumann’s report, prepared by him, and described in his testimony, is a more probative study.
First, he compared the representation of Latinos in specific skilled and licensed jobs at Rush with their availability as measured by their representation in identical jobs in the national health field labor force. Second, he divided the Rush workforce into three groupings according to broad definitions of skill requirements for the various jobs within each group, and compared the Latino representation in these occupational groups with their availability as measured by a weighted average of the City of Chicago and the Chicago Standard Metropolitan Statistical Area (SMSA) census data. In all comparisons, Dr. Neumann found no evidence of any statistically significant underemployment of Latinos or adverse hiring of Latinos at Rush. •
There is no direct or statistical evidence to prove plaintiffs’ class allegations. Even if plaintiffs made out a prima facie case, the [district] court finds that Dr. Neumann’s report and testimony rebut any inference of discrimination which can be drawn from the evidence. The burden of persuasion in proving classwide discrimination remains on the plaintiffs; and from all the evidence adduced at trial, the court finds plaintiffs, have failed to establish any such discrimination by defendants.
A critical examination of the findings reveal that they are not argumentative or conclusory and are not clearly erroneous.
II B
In the proceedings below, both sides submitted statistical evidence to support their respective positions. The statistical evidence relied upon by plaintiffs were “EEO-1” Reports, which are reports submitted by defendant Rush every year to the Equal Employment Opportunity Commission. EEO-1 Reports are workforce data summaries that contain a breakdown by sex and ethnic group of defendant Rush’s employees. These summaries had been prepared from records made in course of hiring employees, and represented the employees’ own representations or the conclusions of personnel officers, as to the employees’ race. Defendants relied on COMSHARE summaries which were prepared by Comshare Inc., a time-sharing computer services company. The COMSHARE summaries were workforce data taken from defendant Rush’s year-end employment information for 1972 to 1979, with adjustments attacked by plaintiffs as tainted by subjective considerations in the minds of defendants’ officers. COMSHARE data was coded to show the Latino composition of defendant Rush’s workforce.
The court’s findings set forth above show that the trial judge reviewed both sets of evidence and then found defendants’ testimony and statistical information more probative. Plaintiffs now challenge that determination and argue that their EEO-1 Reports should have been accepted and that defendants’ COMSHARE data was erroneous.
By inference from the EEO-1 Report submitted by plaintiffs, there was a statistical disparity between the number of Latinos in defendant Rush’s workforce and their availability based on 1970 census data. But the district judge found as a fact that the EEO-1 Reports did not accurately portray the number of Latinos in the workforce. This statistical analysis is not sufficiently probative for the further reason that it fails to compare the number of Latinos in defendant Rush’s workforce with the [1225]*1225availability of Latinos who were qualified for jobs at Rush, which delivers specialized health care, and where English is a BFOQ.
Next plaintiffs assert that defendants’ COMSHARE data is erroneous because it includes persons with Spanish surnames who are not Hispanic and thereby inflates the number of persons designated as Latinos in Rush’s workforce. Plaintiffs complain that “Latino” was not defined by the district court as referring to a person of “Hispanic ancestry or ethnic background.” He would allow, they say, persons of no such ancestry or background to be classed as Latinos. Yet, in deciding whether the suit should proceed as a class action the court, “over defendants’ objections to class certification accepted plaintiffs’ definition of ‘Latino’ as being ‘any Spanish surnamed person or individual of Hispanic ancestry residing in the City of Chicago.’ ” Thus the COMSHARE data followed the district court’s class definition of Latino by including persons either of Spanish surname or Hispanic ancestry. If this was a trap for plaintiffs, they laid it themselves.
The district court was persuaded by defendants’ expert Dr. Neumann who noted that while not all Spanish surnamed persons are Hispanic, many persons who do not have Spanish surnames are of Hispanic ancestry and that the net effect of the two would cancel out the over or under count. To test the accuracy of the COMSHARE summaries, the district court requested that defendant Rush submit a sampling of the employment files of persons designated as Latinos. All 12 files examined met both the Spanish surname and Hispanic ancestry measures of being a Latino. We therefore hold that the properly admitted COMSHARE data supported the conclusions reached and was correctly relied upon by the district court.
We find very damaging to plaintiffs’ position the fact that not only was their statistical evidence insufficient, but that they failed completely to come forward with any direct or anecdotal evidence of discriminatory employment practices by defendants. Plaintiffs did not present in evidence even one specific instance of discrimination. There was no individual to testify how defendant discriminated against him. A case often cited by plaintiffs to support their position is International Brotherhood of Teamsters v. United States, 431 U.S. 324, 97 S.Ct. 1843, 52 L.Ed.2d 396 (1977). In that employment discrimination case, the United States Supreme Court found that the plaintiff had the initial burden of making out a prima facie case of discrimination by the preponderance of the evidence. And in meeting their burden, the plaintiff in Teamsters presented in addition to reliable statistics, the “testimony of individuals who recounted over 40 specific instances of discrimination,” to bolster their statistical evidence. Id. at 338, 97 S.Ct. at 1856. There was also no showing how Rush discriminated, if it did, comparable to the use made of the police examinations in United States v. City of Chicago, supra. There were only petty grievances, such as that personnel officers did not return the phone calls of employment agencies. Since plaintiffs here did not make an adequate showing of either statistical or direct evidence, they failed to make out a prima facie case and so the district court correctly found that plaintiffs failed to prove any of their class allegations. Plaintiffs’ arguments about when the burden shifts to defendants are irrelevant since plaintiffs never met their initial burden.
We may remark in passing that this appeal would be close to the frivolous and probably never would have been filed, except for the strange fact that in the community, in the decade before action brought, the percentage of Latinos in the population rapidly grew, while according to the EEO reports, their percentage in the employ of the hospital actually declined. If the EEO statistics understated, the error might be expected to be consistent from year to year. This is about all plaintiffs have, and naturally they harp on it. It [1226]*1226proves, according to them, that the COM-SHARE data is false and should have been excluded. There are, however, so many innocent possible explanations it is impossible to disregard the other relevant evidence in the case when these explanations are not persuasively eliminated. For example, there could have been a change in the Latino’s perception of how it would be advantageous to him to characterize his race. Though not Filipino, a Latino could be of Mexican or of Caribbean origin, or a Spaniard from Spain, very disparate ethnic and cultural groups though lumped together by the EEOC. They constitute a minority among all the minorities. Certainly, it must be far easier to establish discrimination against all minorities than it is to accuse the employer of singling out one, or here rather three minorities, for discrimination from which other minorities are exempted. How could a great nonprofit public service body be shown to have a motive for such discriminatory discrimination? It is clear the plaintiffs undertook a difficult task, for which the simplistic use of statistics, acceptable perhaps in ordinary discrimination litigation, is insufficient here. To mulct in damages a body such as Rush, it is not sufficient to raise unanswered questions and they do not constitute in and of themselves a prima facie case. Defendants’ officers certainly thought the EEOC statistics misleading as to the true extent of Latino employment, and that was why they resorted to COMSHARES. There was no evidence they entertained that thought in bad faith or for self-serving reasons, and the district judge entertained it too.
The district judge thought the various government agencies had promulgated “conflicting and confusing definitions of Hispanic persons.” An amicus brief by the Mexican American Legal Defense and Educational Fund, Inc., disputes this and argues that the government definitions are consistent. It is not necessary to pass on this issue and we do not do so except to remark that much that is said in this field of discourse is Orwellian.
III
Finally, plaintiffs argue that the trial court used an incorrect legal standard in deciding against the individual claim of plaintiff Fernando Romero. Romero applied but was not hired as a maintenance mechanic in Rush’s physical plant department, and a white applicant was hired. Plaintiffs say that Romero was discriminated against and not hired because he is Latino. The statistics were against Romero inasmuch as the workforce in the physical plant department was well integrated, and the hires at about the date of Romero’s application included one Latino, two blacks, and seven whites.
Romero claims he established that “he was a member of a minority, that he applied for the position of maintenance mechanic, that he was qualified for the job * * *, and that a white man * * * was hired instead” and thereby proved his prima facie case. But that showing is insufficient unless the circumstances give rise to an inference of unlawful discrimination. Texas Department of Community Affairs v. Burdine, 450 U.S. 248, 101 S.Ct. 1089, 67 L.Ed.2d 207 (1981). The district judge found no such inference possible in view of several factors: the above statistics, Rush’s general policy, and the unusual courtesy, and consideration shown Romero by his own testimony.
The next erroneous argument made by plaintiffs was that defendants were required to show that the person hired was more qualified than Romero and absent such a showing, that plaintiff Romero is entitled to a finding in his favor. Burdine rejects the requirement that the employer who hired a nonminority applicant instead of a minority one, show the former was better qualified.
IV
Therefore, for the reasons above stated, the district court’s decision is
AFFIRMED.