On the Legal Front: Ground Rules for Adverse Impact
Art Gutman
Florida Institute of Technology
The Supreme Court has been silent on adverse (or disparate) impact since Wards Cove v. Atonio (1989). Furthermore, even though adverse impact is featured in the Civil Rights Act of 1991 (CRA-91), there are still residual ambiguities, as evidenced by the controversial ruling in
Lanning v. Septa (1999; see for example Sharf, 1999). The Uniform Guidelines on Employee Selection Procedures
(1978) are scheduled for revision. Given our likely role as contributors, I want to propose for discussion nine ground rules for the revision based on the following definition of adverse impact:
A statistically significant difference in actual or implied selection rates between (or among) two (or more) Title VII protected classes (race, color, religion, sex & national origin) causally connected to selection tests or other selection criteria that, themselves, are subject to psychometric scrutiny in accordance with the SIOP Principles (as revised).
Ground Rule 1Only Actual or Implied Selection Rates
Griggs v. Duke Power (1971), the landmark adverse-impact ruling, featured two causes of adverse impact: (a) cognitive tests (Bennett & Wonderlic) and (b) the high school diploma. So did
Albermarle v. Moody (1975), which together with Griggs served as the basis for the
Uniform Guidelines. The cognitive tests excluded 94% of actual Black applicants as compared to 42% of
actual White applicants. These were people with names and addresses who tried and failed. The diploma was deemed exclusionary based on demographic data revealing a 34% high-school graduation rate for Whites and a 12% graduation rate for Blacks in North Carolina at the time. Clearly, applicants need not apply if they are knowingly deficient with respect to an exclusionary rule. Therefore, while adverse impact based on standardized tests is generally evaluated using actual applicants, adverse impact based on exclusionary rules occurs by implication.
Other examples of implied adverse impact include exclusionary rules based on height and weight criteria
(Dothard v. Rawlinson, 1977), methadone use (NYC v. Beazer, 1979), facial hair
(Fitzpatrick v. Atlanta, 1993; Bradley v. Pizzaco, 1993), arrest records (e.g.,
Gregory v. Litton, 1972; Green v. Missouri Pacific, 1975), misdemeanors versus felonies
(Carter v. Gallagher, 1971; Butts v. Nichols, 1974; and Hyland v. Fukada, 1978), and credit information such as wage garnishment
(Johnson v. Pike, 1971; and Wallace v. Debron, 1974). There are others, but these will suffice. Ground Rule 1 says actual or implied selection rates are the
only viable source of adverse impact.
Ground Rule 2Title VII Classes
Ground Rule 2 says Title VII is the only viable law available for applicants or employees to sue for traditional adverse impact violations. As a starter, consider a portion of Section 1607.2(D) of the
Uniform Guidelines that states:
These Guidelines apply only to persons subject to Title VII, Executive Order 11246, or other equal employment opportunity requirements of federal law. These Guidelines do not apply to responsibilities under the Age Discrimination in Employment Act of 1967, as amended, not to discriminate on the basis of age, or under Sections 501, 503, and 504 of the Rehabilitation Act of 1973, not to discriminate on the basis of handicap.
The Supreme Court has ruled that adverse impact is not viable in constitutional claims (see
Washington v. Davis, 1976). Furthermore, despite early adverse-impact victories in the Age Discrimination in Employment Act (ADEA; see
Geller v. Markham, 1980; Leftwich v. Harris-Stowe, 1983), such claims were effectively precluded by the Supreme Court ruling in
Hazen v. Biggens (1993). Adverse impact is also irrelevant to the Equal Pay Act because those claims are scripted (i.e., proof of unequal pay for jobs equal in skill, effort, responsibility, and working conditions).
Executive Order 112146 (EO 11246) on voluntary affirmative action is clearly an important issue in the
Uniform Guidelines, but not for lawsuits by applicants or employees. EO 11246 is administered by the Office of Contract Compliance Programs (OFCCP) for nonfederal contractors and the EEOC for federal agencies. These two agencies have sole power to challenge affirmative action violations and order remedies. To both the OFCCP and EEOC, elimination of adverse impact is one of several recommended solutions for reducing underutilization of minorities and women (together with enhanced recruitment, outreach, and training).
Adverse impact is also referenced in statutory language in Section 103(a) of the Americans with Disabilities Act (ADA), but here, it is merely a prelude to the reasonable accommodation scenario. As written in the ADA:
It may be a defense to a charge of discrimination under this Act that an alleged application of standards, tests, or selection criteria that screen out or otherwise deny a job or benefit to an individual with a disability has been shown to be job-related and consistent with business necessity, and such performance cannot be accomplished by reasonable accommodation, as required under this Act.
For example, a written test with adverse impact on learning-disabled people requires proof of job relatedness, but the larger question is whether the KSAs needed to perform well on the test relate to essential functions of the job (and if so, if barriers implied by the test can be eliminated with reasonable accommodations). To illustrate, in
Stutts v. Freeman (1983), Stutts, a dyslexic, was excluded from operating heavy equipment based on low GATB scores. The 11th Circuit ruled the KSAs required to perform well on the GATB are unrelated to successful operation of heavy equipment.
In addition, adverse impact in disability claims is generally facially apparent, meaning, statistical proofs are unnecessary. For example, in
Strathie v. Dept. of Transportation (DOT; 1983), a DOT regulation excluding bus drivers with hearing aids had an obvious adverse impact on individuals with hearing impairments. So too would any rule excluding individuals with eye glasses, who take medication, who cannot sit or stand for long periods, and so on.
Ground Rule 3Focus on Selection, Not Recruitment
The reason for limiting the Uniform Guidelines to Title VII and Executive Order 11246 is transparent. It protects employers covered by EO 11246 from vulnerability under Title VII. As stated in Section 1607(C) of the
Uniform Guidelines:
These Guidelines apply only to selection procedures which are used as a basis for making employment decisions. For example, the use of recruiting procedures designed to attract members of a particular race, sex, or ethnic group, which were previously denied employment opportunities or which are currently underutilized, may be necessary to bring any employer into compliance with federal law, and is frequently an essential element of an effective affirmative action program; but recruitment practices are not considered by these Guidelines to be selection procedures.
Thus, Ground Rule 3 says recruitment is not a viable source of adverse- impact claims. Consequently, fears of adverse impact due to Internet-based recruitment are misplaced (see Sharf, 2000; Harris, 2003). There is vulnerability for employers who recruit exclusively on the Internet if doing so leads to significant underutilization of minorities. However, the vulnerability is with respect to EO 11246,
not Title VII. Of course, if employers knowingly use such practices with the motive to limit minority recruits, this makes for a viable Title VII
disparate treatment claim. But how likely is that in this day and age?
Another illustration of the connection between adverse impact and selection procedures is in comparable-worth claims. Here, jobs with 70% or more females usually pay less than jobs with 70% or more males, even when job evaluation studies show the male jobs do
not have more internal worth than the female jobs. Thus, the claim is that market forces (or prices) adversely impact jobs congregated by females, as made, for example, by female nurses in
Spaulding v. University of Washington (1984). However, the ruling in Spaulding was that market prices do
not represent a specific employment practice, and it negated the adverse impact claim. After
Spaulding, comparable-worth plaintiffs were forced to prove that employers use market forces to intentionally limit female participation, a claim that has failed repeatedly (see for example
American Federation v. Nassau County, 1996).
Ground Rule 4Cross-job and Composition Disparities are for Motive Scenarios
Compare Griggs to International Teamsters v. US (1977) and Hazelwood v. US (1977) and you see three types of statistical disparities.
Griggs featured actual or implied applicant flow data (i.e., selection rates). In comparison, Teamsters featured
cross-job disparities (minorities congregated in a less appealing job as compared to nonminorities congregated in a more-appealing job) and
Hazelwood featured composition disparities (underutilization of minorities in the workforce when compared to qualified and available minorities in the labor pool). Ground Rule 4 is a corollary to Ground Rule 1 and says cross-job and composition disparities are more suitable for disparate treatment scenarios than adverse impact scenarios.
Griggs was a prototypical adverse-impact case. In contrast, Teamsters and
Hazelwood were prototypical class-action disparate treatment (or pattern or practice) cases. As explained by this author
elsewhere,1 the Supreme Court ruled that large cross-job and composition disparities are sufficient to establish a prima facie pattern or practice claim. However the implied defense is the same as in individual claims of disparate treatment as prescribed in
McDonnell Douglass v. Green (1973). That defense, often termed a burden or production, requires only an articulation (or verbalization) of a legal reason for a selection decision, not factual proof. The plaintiff must then prove the articulation offered is a pretext for discrimination (i.e., baloney).
1 See Gutman (2000), pages 3338.
The teamsters lost because there was no plausible explanation other than facial discrimination (i.e., minorities need not apply) for an inexorable zero number of minorities in the more desirable of two bus-driving routes. On the other hand, the Hazelwood School District won because (a) the relevant composition
disparity2 was not statistically significant and (b) the defendant had
a plausible legal reason explaining why the disparity existed (they could not successfully recruit from St. Louis because it was too far from the Hazelwood
School District and the St. Louis jobs offered better pay). Neither case involved nor implied defenses amounting to job relatedness as required by Griggs. Ground Rule 4 says keep it that way.
2 In Hazelwood, the composition disparity for the entire county (including St. Louis) was statistically significant, but the composition disparity in the immediate area excluding St. Louis was not. The Supreme Court then ruled the appropriate comparison was to the labor pool that excluded St. Louis.
Why? Basically, Wards Cove contains the elements of Teamsters, not
Griggs. There were cross-job disparities such that Eskimos and Filipinos were congregated in a less-desirable job category (cannery work) and Caucasians were congregated in a more-desirable job category (professional work). Ironically, in a prior case with virtually identical facts as
Wards Cove (Domingo v. New England Fish, 1984), the 9th Circuit viewed the cross-job disparities as evidence for a pattern or practice of discrimination and concluded an adverse ruling was unnecessary. Accordingly:
In this case, plaintiffs presented sufficient evidence of discriminatory treatment...[t]he workforce statistics were not necessary to raise an inference of discrimination. They merely demonstrated the consequences of Nefcos discriminatory hiring practices.
The 9th Circuit should have done likewise in Wards Cove. Strangely, the
Wards Cove Circuit Court decided that adverse impact based on cross-job disparities is a viable claim. Ground Rule 4 says no way!
The controversy in Wards Cove focused squarely on a plurality opinion by Justice OConnor (and three others) in the prior case of
Watson v. Fort Worth Bank (1988). In Watson, OConnor proposed using the defense from
McDonnell Douglass v. Green (articulation) to replace the job-relatedness defense from
Griggs and Albermarle (and the Uniform Guidelines). Had the Supreme Court followed its own precedents from
Teamsters and Hazelwood, the logical ruling would have been (a) Wards Cove
is a pattern or practice case, not an adverse impact case and (b) in a pattern or practice case, the burden is on the plaintiff to prove the defendants articulated legal reason for the statistical disparity (cross-job or composition) is lunchmeat. In other words, there was no need to disturb the
Griggs-Albermarle tradition in Wards Cove.
Ground Rule 5Subjectivity is not Discretion
Blacks Law Dictionary (1990) defines discretionary acts as involving no hard and fast rule as to the course of conduct that one must take or not take and, if there is a clearly defined rule, such would eliminate discretion. In other words, discretionary power implies arbitrary decision making. To us,
subjective decision making implies judgments open to psychometric scrutiny. For example, we do not endorse discretionary performance appraisals. Yet, to the OConnor plurality in
Watson (including Rehnquist, Scalia, and White), the terms discretion and subjectivity were treated as synonyms, a view later endorsed by Justice Kennedy, who teamed up with the OConnor plurality to form the majority in
Wards Cove. Ground Rule 5 says subjective decisions and discretionary decisions are not necessarily the same.
Watson should have been Griggs revisited.3 Clara Watson was passed over for promotion four times by White supervisors based on subjective ratings of interview performance, experience requirements, and past job performance. The 1987
Principles for the Validation and Use of Personnel Selection Procedures identifies the selection procedures below as suitable for validation research. The causes of the cross-job disparities cited in
Wards Cove (word-of-mouth recruitment, walk-in hiring, nepotism, and vague subjective selection procedures) are conspicuously absent from this list, whereas the causes of adverse impact cited by Clara Watson are underscored.
[P]aper-and-pencil tests, performance tests, work samples, personality inventories, interest inventories, projective techniques, tests of honesty or integrity including polygraph examinations, assessment center evaluations, biographical data forms or score application blanks,
interviews, educational requirements, experience
requirements, reference checks, physical requirements such as height and weight, physical-ability tests,
appraisals of job performance, computer-based tests interpretations, estimates of advanced potentials, or any other selection instrument, whenever any one or a combination of them is used or assists in making a personnel decision.
3 At least the major part of it was. There were also charges of disparate treatment based on stock statistics that the Supreme Court ignored.
In a brief filed for the American Psychological Association (APA), Bersoff, (1988) argued that subjective tests (or devices) use the same psychometric procedures for validation as objective tests. Justice OConnor believed differently, stating:
Standardized testslike those at issue in our previous disparate impact cases, can often be justified through formal validation studies respondent warns, however, that validating subjective criteria in this way is impracticable
OConnor suggested it is because subjective decisions are discretionary that they are harder to validate than standardized tests. Accordingly:
In the context of subjective or discretionary employment decisions, the employer will often find it easier than in the case of standardized tests to produce evidence of a manifest relationship to the employment in question.
In short, the APA was unable to convince a majority of justices that in accordance with our
Principles, there is no difference in how to validate an objective test versus (say) a structured (and scored) interview.
Ground Rule 6Psychometric Scrutiny
The OConnor plurality was influenced by key lower-court rulings that subjective decision making is not a viable source of adverse impact. However, in those cases, the so-called subjective decisions were discretionary, not the types of judgments referenced by Bersoff (1988) and the
Principles as amenable to psychometric scrutiny. Ground Rule 6 says adverse impact rules apply
only to subjective decisions that are quantifiable and subject to psychometric scrutiny, and do not apply to discretionary decisions.
To illustrate, in Harris v. Ford (1981), the plaintiff claimed subjective decisions in determining discharges for poor workmanship impacts disproportionately on women. In
Talley v. US Postal (1983), the plaintiff claimed subjective decision making by the primarily White supervisory force has disproportionately affected Blacks and females. Harris was allegedly fired for poor performance and Talley (a postal worker) was allegedly fired for losing her mailbox keys a second time. Both courts reached the same conclusion. As stated by the Harris Court, subjective decision making...is not the type of practice outlawed under
Griggs. Ground Rule 6 says discretionary decisions do not fall under
Griggs, but quantifiable subjective judgments do.
The termination decisions in the Harris and Talley cases were at the discretion of supervisors. Discretionary decision making should fall within the purview of disparate treatment.
Harris and Talley should have had the opportunity to challenge the articulation that they were fired for cause, as would occur under disparate-treatment rules. Any claim here of adverse impact is frivolous without evidence of actual or implied applicant flow-rate disparities. Of course, the
Watson ruling was unanimous in concluding that subjective decision making is subject to adverse impact rules. What needs to be clarified is that claims as in
Harris and Talley do not fit the description of subjective decision making because discretionary decision making is not subject to psychometric scrutiny.
Ground Rule 7The Guidelines are not Equally Applicable to All Causes of Adverse Impact
As stated elsewhere by this author4, adverse impact is not a homogeneous phenomenon. Although the
Griggs-Albermarle rules for assessing validity are exacting, they are not equally applicable to all causes of adverse impact. They apply best to tests and other procedures that are standardized, or capable of being standardized. However, case law reveals both a lighter and a heavier standard.
4 See Gutman (2000) pages 4250 and Gutman (2003).
Biographical factors such as methadone use (for transit authority cops; New York City v. Beazer, 1979) and felony armed robbery conviction (for security guards;
Hyland v. Fukada, 1978) have been successfully defended with articulations (as in
McDonnell Douglas v. Green, 1973) of obvious reasons for exclusion. Cases featuring educational requirements have been successfully defended by citing federal studies supporting a high-school diploma for police officers
(Davis v. Dallas, 1985) and experts testifying that a 4-year college degree is necessary to cope with the exacting training demands for commercial airline pilots
(Spurlock v. United Airlines, 1972). It is arguable, of course, that the defenses above were lighter relative to
Griggs-Albermarle because each involved safety sensitive positions. However, cases such as
Lanning (and so many others) reveal that the defense to adverse impact by standardized tests is no less exacting for police departments as for power companies and paper mills.
At the other extreme, cases such as Dothard v. Rawlinson (1977) reveal more exacting standards for exclusion based on physical characteristics such as height and weight. Here, the standard is proving it is reasonably necessary to exclude all or most individuals who fail to meet the standard. If this sounds similar to the BFOQ (bona fide occupational qualification) defense in gender and age cases, it is. Indeed, in
Dothard, the Supreme Court struck down the height and weight criteria but found it was reasonably necessary to exclude all or most women from being guards in all-male maximum-security prisons. Thus, it was easier to exclude members of an entire class than it was any individual, male or female, based on height and weight.
Therefore, Ground Rule 7 says case law reveals that different types of causes of adverse impact are associated with different standards for defense, and the
Uniform Guidelines should reflect that fact.
Ground Rule 8Components Should Be Identified By the Employer
We know from Connecticut v. Teal (1982) that any component of a selection procedure that causes adverse impact must be defended, even if there is no (bottom line) adverse impact after all components are completed. In
Watson, the OConnor plurality interpreted Teal to mean the plaintiff must identify the cause(s) of adverse impact. As noted in Bersoffs (1988) brief, it was unclear if Fort Worth Bank actually scored their interviews and ratings, or how they were combined to make the promotion decision. In several
pre-Watson cases with similar facts (Gilbert v. Little Rock, 1983;
Griffen v. Carlin, 1985; Green v. USX, 1988), circuit courts demanded defense of the entire selection system if the bottom-line outcome was adverse impact and the cause(s) was unclear. CRA-91 requires plaintiffs to identify the cause(s) of adverse impact. However, Sec.105(B)(i) in CRA-91 specifies that the entire selection system must be defended if there is bottom-line adverse impact and the cause(s) are not clear. Accordingly:
[T]he complaining party shall demonstrate that each particular challenged employment practice causes disparate impact, except that if the complaining party can demonstrate to the court that the elements of a respondents decision-making processes are not capable of separation for analysis, the decision-making process may be analyzed as one employment practice.
Ground Rule 8 says that as a practical matter, it makes sense to score every selection test or criterion that is scorable and understand the contribution of each such component to the selection rate.
Ground Rule 9Job Relatedness is Not Synonymous With Business Necessity
The defense to adverse impact in CRA-91 is job relatedness and consistency with business necessity, as it is in the ADA, which was codified in 1990 (before CRA-91). As noted by Gutman (2003), taken literally business necessity implies a heavier defense than job
relatedness.5 Furthermore, there is an element of a literal meaning of business necessity in the BFOQ defense and in the
Dothard ruling on height and weight criteria. However, in the authors opinion, the 3rd Circuit wrongly interpreted this phrase in
Lanning v. Septa (1999) when it ruled a test must measure the minimal qualifications necessary for successful job performance. This runs counter to statutory language in CRA-91 that states:
5 It was noted that the literal meaning of business necessity implies something is necessary for the business to survive, whereas the meaning job relatedness does not.
The terms business necessity and job-related are intended to reflect the concepts enunciated by the Supreme Court in
Griggs v. Duke Power Co., 401 US 424 (1971), and in other Supreme Court decisions prior to
Wards Cove Packing Co. v. Atonio, 490 US 642 (1989).
Therefore, Ground Rule 9 says there is no reason to suppose the phrase job relatedness and consistency with business necessity changes any of the original meaning or any of the parallel terms in
Griggs-Albermarle or the Uniform Guidelines.
Conclusions
There are other issues, most notably, alternative procedures that produce less or no adverse impact (a requirement in the
Uniform Guidelines). For example, in Bridgeport Guardians v. Bridgeport (1991), one witness (Jim Outtz) proposed more intensive scrutiny of applicants using videotaping procedures, but the 2nd Circuit ruled the city was not obligated to incur extra expenses to do so. Similarly, Barrett (1997) reported that in the 1970s, there was no adverse impact among applicants who passed his course on legal issues before taking a police-entry exam. However the city (Akron, Ohio) stopped offering the course based on financial reasons. Obviously, there is no reasonable accommodation requirement for equally valid alternatives, but perhaps this issue should be entertained (e.g., incentives for companies that pay extra to find equally valid alternatives). Ill leave that for another time (and possibly, another author). For now, let me reemphasize my reason for proposing the ground rules. I hope we can start a dialogue and come to some agreements that we can then take to the EEOC so as to influence the revision of the
Uniform Guidelines. I have no direct proof of this, but it is my sense that we have been most successful in court cases on issues SIOP members have agreed on and less successful on issues SIOP members have bickered over.
References
Barrett, G. V. (1997). An historical perspective on the Nassau County Police entrance examination.
The Industrial-Organizational Psychologist, 35(2), 4250.
Black, H. C. (1990). Blacks law dictionary (6th ed.). St. Paul, MN: West Publishing Company.
Bersoff, D. N. (1988). Should subjective employment devices be scrutinized?
American Psychologist, 43, 10161018.
Gutman, A. (2000). EEO law and personnel practices (2nd Edition). Thousand Oaks, CA: Sage Publications.
Gutman, A. (2003). Adverse impact: Why is it so difficult to understand?
The Industrial-Organizational Psychologist, 40(3), 4250.
Harris, M. (2003). Speeding down the information highway: Internet recruitment and testing.
The Industrial-Organizational Psychologist, 41(2), 103106.
Sharf, J. (1999). Third Circuits Lanning v. SEPTA decision: Business necessity requires setting minimum standards.
The Industrial-Organizational Psychologist, 37(2), 138149.
Sharf, J. (2000). As if g-loaded adverse impact isnt bad enough, Internet recruiters can expect to be accused of e-loaded impact.
The Industrial-Organizational Psychologist, 38(2), 156.
Society for Industrial and Organizational Psychology. (1987).
Principles for the validation and use of personnel selection procedures (3rd ed.). College Park, MD: Author.
Uniform Guidelines on Employee Selection Procedures, 29 C.F. R. 1607 (1978).
Case Law Citations
Albermarle Paper Co. v. Moody (1975) 422 US 405.
American Federation etc. (AFSCME) v. Nassau County (1996) 96 F.3d 644.
Bradley v. Pizzaco of Nebraska Inc. (CA8 1993) 7 F.3d 795.
Bridgeport Guardians, Inc. v. City of Bridgeport (CA2, 1991) 933 F.2d 1140.
Butts v. Nichols (S.D. Ia. 1974) 381 F.Supp 573.
Carter v. Gallagher (CA8 1971) 452 F.2d 315.
Connecticut v. Teal (1982) 457 US 440.
Davis v. Dallas (CA5 1985) 777 F.2d 205 (No. 97-843).
Domingo v. New England Fish (CA9 1984) 727 F.2d 1429.
Dothard v. Rawlinson (1977) 433 US 321.
Fitzpatrick v. City of Atlanta (CA11 1993) 2 F.3d 1112.
Geller v. Markham (CA2 1980) 635 F.2d 1027.
Gilbert v. City of Little Rock (CA8 1983) 722 F.2d 1390.
Green v. Missouri Pacific Railroad Company (CA8 1975) 523 F.2d 1290.
Green v. USX (CA3 1988) 843 F.2d 1511.
Gregory v. Litton Systems, Inc. (CA9 1972) 472 F.2d 631.
Griffen v. Carlin (CA11 1985) 755 F.2d 1516.
Griggs v. Duke Power Co. (1971) 401 US 424.
Harris v. Ford Motor Company (CA4 1981) 651 F.2d 609.
Hazelwood School District v. United States (1977) 433 US 299.
Hazen v. Biggens (1993) 507 US 604.
Hyland v. Fukada (CA9 1978) 580 F.2d 977.
International Brotherhood of Teamsters v. United States (1977) 431 US 324.
Johnson v. Pike Corp. of America (CD Cal 1971) 332 F Supp 490.
Lanning v. SEPTA (CA3 1999) 181 F.3d 478.
Leftwich v. Harris-Stowe State College (CA8 1983) 702 F.2d 686.
McDonnell Douglas Corp. v. Green (1973) 411 US 792.
New York City v. Beazer (1979) 440 US 568.
Spaulding v. University of Washington (CA9 1984) 740 F.2d 686.
Spurlock v. United Airlines, Inc. (CA10 1972) 475 F.2d 216.
Strathie v. Dept. of Transportation (CA3 1983) 716 F.2d 227.
Stutts v. Freeman (CA11 1983) 694 F.2d 666.
Talley v. United States Postal Service (CA8 1983) 720 F.2d 505.
Wallace v. Debron Corp (CA8 1974) 494 F.2d 674.
Wards Cove Packing Company v. Atonio (1989) 490 US 642.
Washington (Mayor, DC) v. Davis (1976) 426 US229.
Watson v. Fort Worth Bank & Trust 487 US 977 (1988).
January 2004 Table
of Contents | TIP Home
| SIOP Home