Parry (1993) asserted that "Title I may be the single most important title of the Americans with Disabilities Act because employment is the key to independence" (p. 57). Previous research by disability scholars on the impact of Title I of the Americans with Disabilities Act of 1990 (hereinafter ADA) in reducing unemployment for people with disabilities has been primarily quantitative in nature (see, for example, Bruyère, Erickson, & VanLooy, 2006; Colker, 2005; Harris, Gould, Ojok, Fujiura, Jones, & Olmstead, 2014; McMahon, Hurley, West, Chan, Roessler, & Rumrill, 2008; Miller, 1998; Moss, Ullman, Ranney, & Burris, 2005; Shaw, Chan, & McMahon, 2012). In addition, the focus of much of the existing research on the ADA has been on the outcomes of litigation rather than on the process of litigation (Falk, 1994; Lens, 2003), leaving a gap in the disability literature about the possible extralegal influences, such as societal attitudes toward disability, on case outcomes and what happens to ADA cases as they make their way through the litigation process (Burris & Moss, 2000; Harris, et al., 2014).

Purpose of Study

The courts have played an important role in the implementation of the ADA (Percy, 2001). Individuals who believe they have been discriminated against in the workplace on the basis of disability are allowed, under the ADA's regulations, to bring a lawsuit in the U.S. federal court system. Judicial opinions in ADA cases are a rich source of information about the law's implementation, and therefore served as the unit of analysis for this study.

The purpose of this study was to explore the legal patterns and trends in all publically available U.S. appellate judicial opinions in disparate impact cases decided under the employment provisions (Title I) of the Americans with Disabilities Act from the ADA's July 26, 1992, effective date through July 31, 2012. Cases filed under the ADA Amendments Act of 2008, which became effective on December 31, 2008 (U.S. EEOC, Questions and answers on the final rule implementing the ADA Amendments Act of 2008, n.d.) were not included in this study.

Civil Rights Cases in the United States

Title VII of the Civil Rights Act of 1964 made it unlawful for an employer to limit, segregate, or classify his employees in any way that would deprive or tend to deprive any individual of employment opportunities or otherwise adversely affect his status as an employee, because of "a particular characteristic or characteristics like race or sex" (Grossman, 2010).

In 1971, in Griggs v. Duke Power Co., the Supreme Court interpreted this language to prohibit not just intentional discrimination (disparate treatment), but unintentional discrimination (disparate impact) as well. At issue in that particular case were the requirements that a candidate for certain low-skill jobs must have a high school diploma and must achieve a certain score on an aptitude test — requirements that had disqualified a disproportionate number of African American applicants for new positions and internal transfers, despite the requirements' possessing no obvious connection to job performance. The Supreme Court ruled in Griggs that employers could not rely on hiring criteria that imposed a demonstrable negative impact on a protected class of workers and yet were not justified by business necessity. The Court concluded that targeting discriminatory consequences — not just discriminatory motives — was part of Congress' purpose in enacting Title VII.

In 1991, in response to several Supreme Court decisions narrowing the scope of liability in cases involving disparate impact or unintentional discrimination, Congress amended Title VII to restore the initially broader scope by adding a provision that expressly prohibits disparate impact. Under the current statute, an employer is liable if "a complaining party demonstrates that a respondent uses a particular employment practice that causes a disparate impact on the basis of race, color, religion, sex, or national origin and the respondent fails to demonstrate that the challenged practice is job related for the position in question and consistent with business necessity" (Title VII).

Disparate Treatment and Disparate Impact

There are two legal theories available to plaintiffs (employees) who file a claim of employment discrimination against defendants (employers) under Title I of the ADA: disparate treatment (intentional discrimination) and disparate impact (unintentional discrimination). As mentioned, disparate treatment and disparate impact claims did not originate with the ADA, but with Title VII of the Civil Rights Act of 1964. Disparate treatment and disparate impact claims are allowed under the ADA because Congress adopted the language of previous civil rights statutes in the drafting of the ADA (Mezey, 2005).

The United States Federal Court System

Employment discrimination against employees with disabilities is covered under Title I of the ADA. The United States Equal Employment Opportunity Commission (EEOC) is charged with enforcing the ADA, so an employee must first file a complaint with the EEOC. After an investigation, the EEOC makes a determination on the merits of the case. The EEOC may issue a Notice of the Right to Sue, which allows the employee to bring a lawsuit against his/her employer in U.S. federal district court, or the EEOC may file the lawsuit on behalf of the employee (U.S. EEOC, n.d.).

The U.S. federal district courts serve as trial courts to hear federal cases in both criminal and civil suits. Cases are filed by plaintiffs or appellants (an employee in the case of an employment discrimination case) and filed against defendants or respondents (an employer in the case of an employment discrimination case). Cases are presented before a federal judge and typically are decided by a jury. There are 94 federal judicial districts. Each state has at least one district. The 94 districts are divided into 12 regions, referred to as circuits (United States Courts, n.d.).

If the plaintiff or defendant is unsatisfied with the outcome of the case at the district court level, an appeal may be filed with the U.S. court of appeals. There are 12 appellate courts, one for each region or circuit. A panel of three judges presides over the trial, and the court makes a decision on the appeal based on the case established at the U.S. district court level. The judges do not receive additional evidence or hear testimony from witnesses, and there is no jury. Attorneys from both sides present their arguments in a written format, known as a brief. In some cases, the court may ask the attorneys to present oral argument on specific issues. Oral argument is typically a short, structured conversation between the panel of judges and each attorney (United States Courts, n.d.).

Based on the argument presented, the appellate court may decide to affirm or overrule the U.S. district court decision. If the case is complex or may have important ramifications for public policy, the court may grant a request from the losing party that the case be reheard en banc, which means that the case is reheard by the full set of judges (up to 11) who preside over the appellate court that hears the case. In rare cases, the court itself may decide to hear the case for the first time en banc. In most cases, the appellate court decision is the final decision in the case. Although the losing party has the right to ask the U.S. Supreme Court to review the case, the Supreme Court is not required to grant that review. The Supreme Court may agree to hear a case if the case involves an important and significant legal principle (United States Courts, n.d.).

Rationale for Examining Title I ADA Disparate Impact Cases

Although disparate impact cases make up a small percentage of the cases decided by U.S. appellate courts under Title I of the ADA (N = 20; See Table I), they are a valuable source of information for disability scholars for several reasons:

First, the small number of ADA Title I disparate impact cases allows for an examination of judicial process in addition to case outcomes. The process of litigation, that is, the mechanism judges use to make decisions, may not be transparent to people with disabilities and disability advocates because judicial decision-making is cloaked in jargon and hidden away in legal databases, such as LexisNexis and Westlaw. Illuminating the judicial decision-making process in a subset of ADA Title I cases may improve the disability community's understanding of not only who wins and loses in ADA Title I cases, but may also increase awareness about the contextual factors of the cases that were salient to the judges in their decision making process.

Second, disparate treatment cases, which comprise the majority of ADA Title I cases and which allege intentional discrimination towards an employee or employees on the part of an employer, focus on the employer's discriminatory behavior, attitudes or motives, which may be covert and thus difficult to prove in court. In contrast, disparate impact cases allege that workplace discrimination is a result of oral or written workplace policies and procedures that have the unintentional effect of excluding a protected class of employees. Thus, evidence in disparate impact cases relies more on objective measures and statistics (e.g., how many employees of a given protected class were hired under a given policy) and less on subjective interpretations of the employee, employer, attorneys and judges about workplace behavior, attitudes, and motives. The theory of disparate impact is used to address discrimination that occurs through the application of a facially neutral policy or procedure; therefore, there is no need for the plaintiff to show intentional discrimination on the part of a covered business or government agency (Stein & Waterstone, 2006). Removing intentional discrimination from the equation provides an opportunity for scholars, rehabilitation and legal professionals, people with disabilities, and advocates to learn more about policies and procedures that may be unintentional barriers to inclusion, including employment, for people with disabilities.

Third, after the passage of Title VII of the Civil Rights Act of 1964, and more specifically after the 1971 U.S. Supreme Court decision in Griggs v. Duke Power Co., policy-based litigation has been viewed by the courts as an effective means of addressing the often "harder-to-reach embedded norms that require job and policy modifications" to ensure the participation of minority groups in society, including the workplace (Stein & Waterstone, 2006, p. 860). The drafters of the ADA incorporated the provisions and remedies of Title VII into the ADA, which included disparate impact theory and policy-based remedies. Policy-based litigation treats appellants as members of a protected class. This approach reflects the minority group approach to disability described in the sociopolitical perspective of disability.

Finally, disparate impact cases address "systemic discrimination or policies that have a disparate impact on a protected class" (EECO v. Randstad, 685 F.3d 433, 440). The idea of a protected class is similar to the minority group approach within the sociopolitical model of disability (Gill, Kewman, & Brannon, 2003; Hahn, 1984). The sociopolitical model of disability asserts that disability is socially constructed. As such, courts as social institutions are influenced by social and cultural understandings of the phenomena they consider in their decisions (Lens, 2003). Therefore, proponents of the sociopolitical model suggest that discrimination against people with disabilities can only be addressed by challenging societal assumptions about people with disabilities as individuals and as a group (Hahn; Percy, 2001). Under the sociopolitical model of disability, legislation and other legal remedies are important tools in challenging societal assumptions about disability (Hahn).

Theoretical Guidance

People with disabilities have always been with us. What have changed slowly over the centuries are the approaches and services provided to these diverse groups of individuals - from isolation within family units and segregation within institutions to inclusion in the various communities and activities of modern life (Drake, 2001; Johnston & Helms, 2008). Indeed, quality of life for people with disabilities is intertwined with public policy designed and implemented at the federal, state, and local levels. For people with disabilities, public policy determines not only where they live and work, but also how socially isolated or integrated they are within society (Drake, 2001; Percy, 2001; Schriner, 2005).

Throughout history, societies have attempted to understand and explain disability as a social construct (Drake, 2001; Liachowitz, 1988). Within the past century, three primary explanations, or models, of disability have emerged within the scholarly literature: medical, economic, and sociopolitical (Blanck, Hill, Siegal, & Waterstone, 2004; Riggar & Maki, 2004; Schriner, 1995; Scotch, 2000a;). Models of disability espouse a particular view of disability, which shapes the design and implementation of disability policy (Schriner, 2001; Scotch). For example, vocational rehabilitation incorporates an economic model of disability, with its focus on work and economic self-sufficiency (Obermann, 1965). Social Security Disability Income (SSDI) reflects a medical model of disability, which sites the problem within the individual with the disability, specifically a bodily condition (Mudrick, 1997). Therefore, SSDI requires recipients to obtain certification from a doctor as to the severity of their disability to be eligible for benefits under SSDI (Berkowitz, 1987).

Sociopolitical Model of Disability

In contrast, the sociopolitical model of disability is based on a civil rights approach to disability policy (Hahn, 1984, 1985, 1991) and was patterned after the Civil Rights Act of 1964, which advanced the civil rights of racial minorities and women. The core beliefs of the sociopolitical model of disability include the right and responsibility to have input into legislation and public policy, the right to access public and private programs and buildings, and the belief that disability is part of a diverse society and should not be treated as an abnormality or solely as a medical condition that needs to be ameliorated (Hahn).

Sample

This study examined all reported U.S. Appellate Court ADA disparate impact cases that met the study selection criteria. The cases were selected through a search of the Westlaw electronic legal research database, a commercial electronic case law reporting system owned by Thomson Reuters.

The initial study sample contained 188 cases. The search terms used to generate the sample in Westlaw were intentionally broad to ensure that no cases were missed. The researcher reviewed all cases in the initial sample and eliminated cases that fell outside the scope of the study. Cases were eliminated for the following reasons: criminal not civil action, case origin or decision date outside the study parameters, not an ADA case, or not an ADA disparate impact case, not a published case, or duplicate case. The analysis generated a sample of 52 Title I, II, and III ADA disparate impact cases (See Table 1). ADA Title I disparate impact cases were the focus of this study; Title II and Title III ADA disparate impact cases were not included in the study. There were 20 ADA Title I disparate impact cases in the sample (See Table 1).

Method

This research employed naturalistic or qualitative research methods in the design and execution of this study for several reasons: First, appellate court judicial opinions are the phenomenon under study and they would be difficult, if not impossible, to study in a controlled, or laboratory, environment. A naturalistic or qualitative approach considers the case and the contextual factors surrounding the case, thus preserving the uniqueness of each case. Second, qualitative research is useful if a researcher is examining a topic about which not much is known and the study is exploratory in nature (Patton, 2002). Third, an advantage of qualitative design and methodology is that it may produce a deeper and richer understanding of the data (Denzin & Lincoln, 2005; Patton, 2002).

Analysis

Qualitative analysis employing the constant comparative method (Strauss & Corbin, 1990) was used to identify and characterize patterns and themes of the judicial opinions in the 20 Title I disparate impact cases. The literature on the sociopolitical model and the minority group approach guided the content analysis of the cases in this study because ADA Title I disparate impact cases focus on workplace policies and practices that may unintentionally discriminate against employees with disabilities as members of a protected group or class (U. S. EEOC, 2008). The researcher developed a framework for analysis based on a review of the literature of the sociopolitical perspective of disability to guide the coding and analysis of the judicial opinions in the subset of Title I ADA disparate impact cases.

Results

The case outcomes in the 20 ADA Title I disparate impact cases that comprised the study were as follows: five cases were decided in favor of the plaintiff; one case was split among nine plaintiffs (2 won; 7 lost); twelve cases were decided in favor of the employer; two cases were indeterminate (waived). Thus, in this sample, there was a 30 percent success rate for plaintiff employees, and a 60 percent success rate for defendant employers, with 10 percent of the cases falling in an indeterminate category.

The results of the qualitative analysis of the text of the judicial opinions in the 20 ADA Title I disparate impact cases produced four themes and three relationships. The four themes were 1) accommodation(s); 2) workplace culture, norms, and policies; 3) judicial process; and, 4) policy space. The three relationships were gap-filling, weighing and balancing, and maintaining status quo versus effecting social change.

The four themes and the three relationships illuminated the process of litigation in the 20 ADA Title I disparate impact cases. Much of the previous research on the ADA focused on the outcomes of a specific case or group of cases; the results of this study, however, showed that the outcome of a case is a part of a larger process; to wit: a complex process involving multiple stakeholders, government and company policies, laws and regulations, and socio-cultural beliefs and attitudes about civil rights in general and specifically civil rights for people with disabilities. Each of the four themes describes specific inputs into the final judicial decision in a case. The process of evaluating the inputs in a specific case is the weighing and balancing relationship. In other words, judges examine the information available to them (inputs), weigh and balance the information, and issue a decision (output) in the case. Each case is comprised of a unique set of facts; therefore, judges must use case law, statutory language, and testimony to fill any gaps between the evidence in the new case with previous cases. Finally, judges may be called upon to make decisions in cases that have the potential to effect change at a broad societal level.

The Concept Map (Figure 1) displays each of the four themes, their categories and subcategories, and the relationships among them (Daley, Conceicao, Mina, Altman, Baldor, & Brown, 2010).

View extended text description of Figure 1

Theme 1: Accommodation

This theme incorporates references in the disability literature and in the text of the judicial opinions to accommodations. Theme 1 contains three categories: A) "Legal Requirement"; B) "Environment"; and, C) "Theory and Data Interaction."

Category A "Legal Requirement" includes references in the text of the judicial opinions to the term "reasonable accommodation(s)." There are two subcategories: "Rehab Act" and "ADA." The subcategories include references in the text of the judicial opinions to the requirement that governmental agencies and private businesses covered under the Rehabilitation Amendments Act of 1973 (Rehab Act) and the ADA must provide reasonable accommodations to qualified persons with disabilities.

Category B "Environment" contains two subcategories, "Inclusion" and "Access." This category and subcategory reflects the view in the disability literature that it is environment that limits inclusion and access to the workplace and other parts of society, not the disability (Hahn, 2003).

Category C ,"Theory and Data Interaction" encompassed different meanings of the word "accommodation(s)" that the researcher identified in the literature review and in the analysis of the judicial opinions in the study sample. For example, the judges in Eckles v, Consolidated, discuss accommodations as a legal requirement:

We recognize that the ADA does expressly recognize "reassignment to a vacant position" as an expected form of reasonable accommodation, thereby rejecting a line of precedent under the Rehabilitation Act holding that reassignment of a disabled employee was never required (Eckles v, Consolidated, 94 F.3d 1041, 1996).

Discussions about accommodations in the disability literature included references to the disability rights movement's goals of improving access to the built environment (e.g., curb cuts, motion-activated lighting) and increasing inclusion in the social environment, including education and work (e.g., assistive technology and job accommodations) (Huemann, 1993). Accommodations that increase access to the built and social environments in turn facilitate inclusion, self-determination, and choice for people with disabilities (Bruyère, 2000; Hahn, 2003; Johnston & Helms, 2008).

Finally, the notation "Pressing for Change" denoted the references in the disability literature to the educational and political actions taken by disability scholars and advocates to advance the view that accommodations facilitate inclusion in the workplace (Hahn, 2003; Huemann, 1993)

Theme 2: Workplace Culture, Norms, and Policies

Theme 2 included intangible factors, such as workplace culture and norms, and tangible factors, such as written policies, procedures, and legislation that govern the employer/employee relationship. The theme included references in the text of the judicial opinions to workplace rules, behavior, and expectations. This theme also incorporated the literature on the sociopolitical perspective of disability, which asserts that people with disabilities experience prejudice and discrimination in the workplace due to societal stigma and negative attitudes toward people with disabilities (Goffman, 1963; Wright, 1983). Theme 2 contained three categories: A) "Environment"; B) "Company-Specific Policies"; and, C) "Public Policies Governing the Workplace."

Category A, "Environment," included references in the text of the judicial opinions to statements and behavior of employers or coworkers about employees with disabilities. In Dalton v. Subaru, the evidence in the case showed negative attitudes toward employees with disabilities on the part of management:

For example, SIA's Manager of Safety and Environmental Affairs, Mark Siwiec, frequently expressed a negative attitude toward disabled employees, calling them "piece[s] of work," showing skepticism about their injuries, and volunteering the opinion that SIA should get rid of everyone with permanent restrictions (Dalton v. Subaru141 F.3d 667 (1998)).

And on the part of the judges:

To require him [the employer] to retain the least able because of disability would handicap the able-bodied, and that is not required by the Act. Such handicapping, such discrimination in favor of the disabled, would invite the same criticisms as "reverse" discrimination on racial and sexual grounds— especially in a RIF case, where a better worker would lose a job to a worse one merely because the better worker had the good fortune not to be disabled (Matthews v. Commonwealth Edison, 128 F.3d 1194 (1997)).

Category B "Company-Specific Policies" incorporates the policies and procedures unique to a specific employer or job. Category B contains two subcategories. The first subcategory is "Contractual (CBA, BFOQ)" which refers to legal contracts and agreements between the employer and labor unions (collective bargaining agreements) and characteristics or attributes that are necessary to the normal operations of a business (bona fide occupational qualifications). This category includes references in the text of the judicial opinions to contractual agreements or job qualifications and requirements specific to a particular business or a specific job. For example, many of the cases in this study dealt with providing accommodations in an employment setting which was governed by a collective bargaining agreement (CBA). The text of the judicial opinion in Kennedy v. Chemical Waste showed the judges' efforts to incorporate the relatively new accommodations requirements under the ADA into existing CBAs:

Kennedy's second argument takes off from the fact that Title VII as amended in 1991 allows an employee "injured by the application of the [employer's] seniority system" to measure the period of limitations from the date of that application, which in this case would be 1994, when Kennedy was laid off. 42 U.S.C. § 2000e-5(e)(2). This provision, Kennedy contends, was incorporated into the ADA along with the rest of Title VII's provisions regarding limitations periods. But the provision is applicable only to "a seniority system that has been adopted for an intentionally discriminatory purpose in violation of this subchapter [i.e., Title VII]," and Kennedy does not claim that the defendants' seniority system discriminates against disabled persons or anybody else. He argues, however, that this qualification in Title VII—the limitation to intentionally discriminatory schemes—was not taken into the ADA. Title VII contains an exemption for bona fide seniority systems. 42 U.S.C. § 2000e-2(h). The purpose is to prevent the use of the "disparate impact" approach to challenge seniority systems, and thus to confine Title VII challenges to seniority systems that intentionally discriminate on one or more of the grounds forbidden by the statute. Pullman-Standard v. Swint, 456 U.S. 273, 276-77, 102 S.Ct. 1781, 1783-84, 72 L.Ed.2d 66 (1982); Banas v. American Airlines, 969 F.2d 477, 481 n. 5 (7^th Cir.1992). The qualification limiting the new statute of limitations to intentionally discriminatory seniority systems is merely a recognition that those are the only systems that can be challenged under Title VII. There is, the argument continues, no exemption in the ADA for bona fide seniority systems. Therefore disparate impact must be a basis for challenging seniority systems under the ADA (an issue, however, on which there is as yet no appellate ruling), and so the qualifier "intentionally discriminatory" would have no function in the ADA and so should be deemed not to have been brought over into it. This is an ingenious argument, but we think it more likely that no part of the provision of Title VII relating to an extended period for challenging a loss of (or due to) seniority was taken into the ADA (Kennedy v. Chemical Waste, 79 F.3d 49 (1996)).

The second subcategory "Rules, policies, and procedures (written and unwritten)," refers to references in the text of the judicial opinions to the materials companies use to hire, train, promote, discipline, and discharge employees. For example, written and unwritten policies on absenteeism, tardiness and sick leave; job and position descriptions; and retirement and health care polices.

Category C "Public Policies Governing Workplace" includes references in the text of the judicial opinions to specific state and federal policies that govern the employer/employee relationship and the operations of the workplace. The category contains two subcategories. The first subcategory, "Local, State, and Federal," includes references in the text of the judicial opinions to specific policies at the local, state, and federal levels of government; for example, worker's compensation (state) and the Family and Medical Leave Act (federal). The second subcategory, "Legal Compliance versus Maximizing Potential" includes references in the text of the judicial opinions about the court's duty to consider the rights of each party affected by the litigation; for example, the employer, the non-disabled employees, and the qualified person with a disability. For example:

For similar reasons, the ADA does not require an employer to renegotiate a bona fide seniority system to avoid "screening out" a disabled employee who may not be able to reach the highest rung on a promotional series because of disability (Boersig v. Union Electric Co., 219 F.3d 816 (2000).

Theme 3: Judicial Process

This theme incorporates references in the text of the judicial opinions to the law and the process of litigation, including judicial decision-making. The theme includes references in the text of the judicial opinions to the content of the law (legislative history, statutes, regulations, and case law) and to the process of law (judicial decision making, testimony, argument).

Theme 3 contains two categories: The first category "What the Statute Says" includes references in the text of the judicial opinions to the ADA's legislative history, statutory and regulatory language, and case law. This category encompasses the courts' references to the language of the statutes and regulations, often referred to informally or colloquially as "the document on its face." The category has four subcategories: "Legislative history"; "EEOC Regulations (Title I)"; "Statutory language"; and "Case law (U.S. Supreme Court precedent)" which describe the types of information judges use to guide them in filling in the gaps between what the statute says and what the statute doesn't say. The four types of information include: excerpts and direct quotations from the ADA's legislative history, the ADA statutes, regulations (such as the EEOC regulations governing the implementation of Title I), and citations of prior cases that dealt with similar legal or procedural issues.

The second category "What the Statute Doesn't Say" includes references in the text of the judicial opinions to methods and strategies the courts use to make decisions about legal issues about which there is no clear direction in the statutes, existing case law, or regulations governing the implementation of the law. The category has four subcategories: "Analogy or Hypothetical"; "Evidence" "Dicta"; and "Case law (Appellate Court precedent)" which explain the types of information judges use to guide them in filling in the gaps between what the statute says and what the statute doesn't say. The four types of information include analogies and hypothetical examples that explain a legal point or a decision; evidence or testimony from the record in the district court case and appellate and defendant briefs filed in the appellate case; dicta, which is judicial commentary that has no legal weight; references to testimony by witnesses during the district court trial and briefs submitted by the parties at the appellate court trial; and citations of prior appellate cases that dealt with similar legal or procedural issues. For example, the judges in Kennedy v. Chemical Waste include dicta (in the form of exaggeration) and a hypothetical situation to explain their decision to rule against Kennedy:

But Kennedy should not feel that he has been robbed of a good suit by a technicality, and not only because statutes of limitations are not technicalities but serve important social purposes. Galloway v. General Motors Service Parts Operations, supra, 78 F.3d at 1165, and cases cited there. He has no case. There is nothing discriminatory about Chemical Waste Management's seniority system. If you are removed from a union position, for whatever reason, you lose the seniority accrued in that position; seniority does not vest. Moreover, Kennedy was not removed from his truck driver's position in 1988 in violation of the ADA; the ADA hadn't been enacted yet (it was enacted in 1990 and became effective in 1992). Nor was there anything discriminatory about his removal because of multiple sclerosis, for he does not question either the doctor's judgment that he could not continue in the job or the bona fides of Chemical Waste Management in acting on that judgment. The notion of reasonable accommodation cannot be stretched to the point of requiring the provision of superseniority to disabled employees who lost their seniority on account of disability years, perhaps decades, before the Americans With Disabilities Act was passed. Suppose Kennedy had worked for Chemical Waste Management from 1930 to 1940, and been laid off in 1940 because he was disabled. If he recovered and was rehired in 1980, and then laid off in 1995, we do not think any court would listen to his argument that the company was obliged to restore to him the 10 years of seniority that he had accrued between 1930 and 1940, any more than it would listen to an argument that he should be given a bonus that he had been denied in 1932 because of his disability. (Kennedy v. Chemical Waste, 79 F.3d 49 (1996)).

Theme 4: Policy Space

This theme incorporates references in the disability literature and in the text of the judicial opinions to policy space. Policy space refers to a group of policies that share a similar focus or purpose (Biggs & Helms, 2007). The shared policy space or focus in civil rights legislation is the design and implementation of legislation that will advance and protect the civil rights of minority group populations who have experienced societal discrimination, such as racial/ethnic groups, women, and people with disabilities.

The literature on the sociopolitical perspective of disability describes the role of grassroots disability rights organizations in raising awareness about societal discrimination against people with disabilities, and the process of rallying diverse disability groups in the United States around a common political cause (Bagenstos, 2009; Hahn, 1984, 1985) The literature details the efforts of coalitions comprised of policymakers and members of grassroots disability rights organizations in the development and passage of Section 504 of the Rehabilitation Act Amendments of 1973 and the ADA (Percy, 1989; Scotch & Berkowitz, 1990).

Theme 4 includes references in the text of the judicial opinions to the ADA as a civil rights law, definitions of disability under the ADA, and the role of accommodations as civil rights for people with disabilities under the ADA. The theme includes references in the text of the judicial opinions as to how the exercise of the right to accommodations operates in the workplace; for example, the effect of accommodations and the accommodations process on employers and non-disabled employees. Theme 4 has three subcategories:

Category A "Component of U.S. Civil Rights Legislation" includes references in the text of the judicial opinions to the ADA's relationship to other civil rights legislations, such as the Rehab Act and Title VI of the Civil Rights Act of 1964. Category A has two subcategories, "Modeled on Title VII and Rehab Act" and "Political and cultural factors" which include references in the text of the judicial opinions that compare and contrast the ADA to other civil rights legislation (Title VII and the Rehab Act) and references to socio-cultural issues related to civil rights for people with disabilities.

Category B "Definition of Disability" includes references in the text of the judicial opinions to the definition of disability within the statutes, policies, programs, and services that comprise U.S. disability policy, including disability employment policy. Category B has two subcategories. The first subcategory, "U.S. versus Global constructs" represents the fact that public policy is contextual (Yanow, 2000). For example, U.S. disability policies put in place to address disability discrimination may define disability differently than do disability discrimination policies in other countries. The second subcategory "Eligibility" includes references in the text of the judicial opinions to the definition of disability within the statutes, policies, programs and services that comprise U.S. disability employment policy:

During that 12-year period, the Social Security Administration ("SSA") reviewed Mr. Lawson's medical condition every two years and determined that he continued to meet its definition of disability, allowing Mr. Lawson continually to receive benefits (Lawson v. CSX Transp., Inc., 245 F.3d 916 (2001))

Category C) "Accommodations" has two subcategories. The first subcategory "A social good (inclusion)" incorporates the literature on the sociopolitical perspective of disability, which links accommodations to increased levels of participation in society for people with disabilities (Hahn, 2003). The second subcategory "A policy solution (legal requirement)" reflects the fact that the accommodations requirements contained in both the Rehab Act and the ADA represent a policy solution to the policy problem of disability discrimination:

The law forbids the employer to disqualify the disabled applicants on the basis of their disability unless the disability prevents them from doing the work even with a reasonable accommodation (Matthews v. Commonwealth Edison, 128 F.3d 1194 (1997)).

Relationships Among the Categories and Themes

There were three relationships identified in the analysis of the 20 ADA Title I DI cases in the study: gap-filling, weighing and balancing, and maintaining status quo versus effecting social change.

Gap-filling includes references in the text of the judicial opinions to the process courts engage in when making decisions in cases where information is missing or incomplete. For example, the court may state that there is no case law that relates to the instant case (which means the case before the court) so it has relied on the language in the legislative history of the law in making its decision:

While the text of the ADA does not provide much support for Eckles' position, neither does it decisively answer the question of whether "reasonable accommodation" can require that the otherwise valid seniority rights of other employees be trumped. Thus we look also to the background of the ADA's "reasonable accommodation" concept and the legislative history of the ADA (Eckles v, Consolidated, 94 F.3d 1041, 1996).

Weighing and balancing includes references in the text of the judicial opinions to statements courts make about the factors that influence their decision-making process. For example, courts may state that they must balance the requirement that a business covered by the ADA accommodate a qualified employee with a disability with the rights of employees covered under a collective bargaining agreement:

In either case, the plaintiff invites the court to disrupt a carefully negotiated agreement between union and employer at the expense of other union employees who hold legitimate expectations of advancement based on the governing CBA (Boersig v. Union Electric Co., 219 F.3d 816 (2000).

Maintaining status quo versus effecting social change includes references in the text of the judicial opinions to statements the courts make about their duty to allow employers to maintain "the ordinary course of business" versus their role as protecting the rights of employees in the workplace: For example, in the Matthews case, the judges stated that the rights of employees with disabilities should be protected in the case of a RIF (reduction in force) or a layoff:

A RIF is not an open sesame to discrimination against a disabled person. Christie v. Foremost Ins. Co., 785 F.2d 584, 587 (7^th Cir. 1986). Even if the employer has a compelling reason wholly unrelated to the disabilities of any of its employees to reduce the size of its work force, this does not entitle it to use the occasion as a convenient opportunity to get rid of its disabled workers (Christie v. Foremost Ins. Co., 785 F.2d 584, 587 (7^th Cir.1986))(Matthews v. Commonwealth Edison, 128 F.3d 1194 (1997)).

Implications

The results of this research has five broad implications for disability scholars, practitioners, people with disabilities, and advocates: 1) Precedent; 2) Collective Bargaining Agreements (CBA); 3) Circuit and Judge Effects; 4) EEOC Successes; and, 5) Sutton Effect.

Precedent

As mentioned, the law evolves through precedent, which means that the decisions in earlier cases influence the decisions in later cases with similar fact patterns (Mellinkoff, 1992). As the law evolves, cases whose decisions have not been overturned or remanded by a higher court (the appellate courts for the district courts; and the U.S. Supreme Court for the appellate courts) form a body of law on a specific legal issue. Precedent set by a higher court is binding on a lower court. For example, decisions by the U.S. Supreme Court are binding on U.S. Appellate and District Courts, and decisions by U.S. Appellate Courts are binding on U.S. District Courts within the appellate court's jurisdiction or district (Mellinkoff).

Three cases in the sample served as precedent for later cases in this study: Eckles v. Consolidated, Matthews v. Commonwealth, and Dalton v. Subaru. The fact patterns in all three cases involved an employee (or employees in the Dalton case) who was requesting an accommodation to be transferred or reassigned in a workplace that operated under a collective bargaining agreement (CBA). In each case, the employer asserted that the requested accommodation was unreasonable because it would violate an existing bona fide occupational qualification (BFOQ) or seniority system under a CBA. In each case, the court agreed with the employer, and cited the earlier cases to support its opinion. For example, the judges in the Dalton case, which was decided in 1998, referred to both the Matthews (1997) and Eckles (1996) cases in their decision in Dalton v. Subaru: The court first cites the Eckles case, which was decided in 1996:

We have already recognized some limitations on the employer's duty to reassign. For example, in Gile we held that an employer has no duty to "bump" an incumbent from a position just to accommodate the request of a disabled employee…Nothing in the ADA requires an employer to abandon its legitimate, nondiscriminatory company policies defining job qualifications, prerequisites, and entitlements to intra-company transfers. See, e.g., Cochrum, 102 F.3d at 912-13 (employer not required to violate the provisions of a collective bargaining agreement to reassign a disabled employee pursuant to the ADA), citing Eckles v. Consolidated Rail Corp., 94 F.3d 1041, 1051 (7^th Cir.1996), cert. denied, 520 U.S. 1146, 117 S.Ct. 1318, 137 L.Ed.2d 480 (1997); Foreman v. Babcock & Wilcox Co., 117 F.3d 800, 810 (5^th Cir.1997), cert. denied, 522 U.S. 1115, 118 S.Ct. 1050, 140 L.Ed.2d 113 (1998) (even if there were no CBA in place, the ADA does not require an employer to reassign in violation of a bona fide seniority system) (Dalton v. Subaru).

The court then cites the Matthews case, which was decided in 1997, and notes that it has found no case law to support reassignment as an accommodation for an employee with a disability if such reassignment would require violating the terms of an existing CBA:

In fact, we have been unable to find a single ADA or Rehabilitation Act case in which an employer has been required to reassign a disabled employee to a position when such a transfer would violate a legitimate, nondiscriminatory policy of the employer, see Duckett, 120 F.3d at 1225 (also failing to find such a case), and for good reason. The contrary rule would convert a nondiscrimination statute into a mandatory preference statute, a result which would be both inconsistent with the nondiscriminatory aims of the ADA and an unreasonable imposition on the employers and coworkers of disabled employees. An employer cannot, of course, convert its responsibility to look to a "broad range" of jobs into a "narrow band" simply by adopting a "no transfer" policy. Any such policy would remain subject to challenge both for any disparate impact it might impose on disabled employees, and for any unreasonable inflexibility in the face of a demand for reasonable adjustments to accommodate a disabled candidate for transfer. See 29 C.F.R. Pt. 1630, App. § 1630.10. See also Matthews, 128 F.3d at 1195-96. Rather, the "broad range" of jobs to which an employer must look when considering transfer as a reasonable accommodation for a disabled employee is bounded from above by the employer's freedom not to offer promotions and from below by its legitimate, nondiscriminatory limitations on lateral transfers and demotions. (Dalton v. Subaru)

All three cases (Eckles v. Consolidated, Matthews v. Commonwealth, and Dalton v. Subaru) were cited as precedent in one or more of seven of the 20 cases in the study.

Examining precedent is one way to determine what issues the court finds salient across a body of law. Precedent-setting decisions in appellate court cases also delimit the decisions of the lower district courts (Maltz, 1988). Once precedent has been set by a higher court, lower courts, which hear the majority of cases, are reluctant to decide cases against precedent in future cases containing the same or similar fact patterns (Mellinkoff, 1992). Thus, future appellants in cases with similar fact patterns may find that precedent set in the earlier cases narrows the scope of relief available to them (Mellinkoff). It is this narrowing, especially on the issue of the definition of disability under the ADA, that some disability rights activists, disability scholars, and legal scholars refer to as backlash against the ADA (Krieger, 2003).

Precedent in ADA Title I disparate impact cases may help or hinder employees with disabilities who believe they have been discriminated in the workplace. For example, an employee with a disability who works in a company where a CBA is in place may be denied an accommodation request for reassignment because such a request violates the terms of the CBA. That employee may still choose to bring a disparate impact suit against his or her employer under Title I of the ADA; however, the employee may find it difficult to prevail because of the precedent set in previous cases with the same or similar fact patterns if the precedent was set by the appellate court which has jurisdiction over the district court that hears the employee's case, or if the precedent was set by the U.S. Supreme Court.

In contrast, precedent may benefit an employee whose suit fits a particular exception to the fact pattern; for example, if the company policy prohibiting reassignment was put in place in bad faith. The judges in Dalton v. Subaru (1998) noted this exception in their decision:

An employer cannot, of course, convert its responsibility to look to a "broad range" of jobs into a "narrow band" simply by adopting a "no transfer" policy. Any such policy would remain subject to challenge both for any disparate impact it might impose on disabled employees, and for any unreasonable inflexibility in the face of a demand for reasonable adjustments to accommodate a disabled candidate for transfer. See 29 C.F.R. Pt. 1630, App. § 1630.10. See also Matthews, 128 F.3d at 1195-96 (Dalton v. Subaru (1998))

The text of the opinion shows that, in the case of a company with a CBA in place, the judges weigh the rights of the employees with disabilities who are covered under the ADA with the rights of all company employees, disabled or not, who are covered under the CBA. The results of this study showed a marked reluctance on the part of the judges to interfere in labor relations. For example, in Boersig v. Union Electric:

Our Benson decision recognizes the importance of protecting these rights from unnecessary interference arising from the perceived need to accommodate a disabled employee under the ADA. We believe that these rights are entitled to protection regardless of whether a plaintiff seeks a reasonable accommodation [or claims the seniority system has a disparate impact on disabled employees. In either case, the plaintiff invites the court to disrupt a carefully negotiated agreement between union and employer at the expense of other union employees who hold legitimate expectations of advancement based on the governing CBA. This sort of judicial intrusion into labor relations is unwarranted unless an employee can show that a seniority system was designed to discriminate against the disabled. Because we find the promotional system in the instant case is a bona fide seniority system, we reject Boersig's disparate impact claim. (Boersig v. Union Electric Co., 219 F.3d 816 (2000)).

Collective Bargaining Agreements (CBA)

An interesting and unexpected result of this study is the fact that the courts' decisions in cases where a CBA is in place appeared to show a distinct chronological trend that followed the relative decline of the power of labor unions throughout the 1990s and into the first decade of the 21^st century (Geoghegan, 2004). Throughout the 1990s, when the ADA was less than a decade old, the judicial opinions in the cases in the study sample demonstrated reluctance on the part of the judges to disrupt existing CBAs. However, in 2012, as the analysis for this study was nearing completion, the U.S. Appellate Court in the 7^th Circuit reversed its own precedent in two EEOC reasonable accommodation cases, EEOC vs. United Airlines and EEOC vs. Humiston-Keeling (2000), after a rehearing of the EEOC vs. United Airways case, citing the U.S. Supreme Court decision in U.S. Airways, Inc., v. Barnett (2002) as the reason for its reversal. Although in Barnett the U.S. Supreme Court found that U.S. Airways was not required to violate its seniority system under its CBA in order to provide reassignment as an accommodation under the ADA, the court stated:

[T]he simple fact that an accommodation would provide a 'preference'—in the sense that it would permit the worker with a disability to violate a rule that others must obey—cannot in and of itself, automatically show that the accommodation is not 'reasonable' (U.S. Airways, Inc., v. Barnett (2002)).

The Barnett decision therefore opens the door to allowing reassignment as a reasonable accommodation under the ADA in cases where a seniority agreement under a CBA is in place. Whether the decision in Barnett represents the high court's willingness to disrupt existing labor relations contracts due to the decline in the power of organized labor (Geoghegan, 2004) or if it simply represents the impact of the growing body of case law in ADA Title I cases is beyond the scope of this study. However, what is important for disability scholars and advocates is that the decisions in the U.S. Airways, Humiston-Keeling, and Barnett cases indicates a willingness on the part of the courts to consider reassignment as a reasonable accommodation in situations where a CBA is in place, which represents a major shift in the courts' views on accommodations under CBAs. The reversal of these decisions demonstrate that courts have interpreted the facts of a particular case in a way that results in a decision that goes against precedent (Maltz, 1988). In addition, employees with disabilities may indeed benefit from the declining importance of unions over the past several decades; however, that gain may come at a cost to all employees' — disabled or not — working conditions.

Circuit and Judge Effects

The other similarity the Eckles, Matthews, and Dalton cases shared was the fact that they were litigated in the 7^th Circuit. The review of the text of the judicial opinions revealed a consistent approach on the part of the 7^th Circuit to cases that dealt with the issue of a reasonable accommodations request under a CBA. Specifically, the 7^th Circuit expressed three reasons for not allowing the plaintiffs to prevail in these cases: (a) Requiring that an employee with a disability be given priority over more senior or more qualified non-disabled employees is affirmative action and the ADA is not an affirmative action statute; (b) allowing accommodations that would violate the training and seniority requirements in CBAs puts the rights of the person with a disability above the rights of non-disabled workers whose rights are protected under longstanding, bona fide CBAs; and (c) requiring a company to pass over the best-qualified applicant for a position in order to accommodate a minimally qualified person with a disability is affirmative action, and contrary to the best economic interests of the employer who has the right to hire the best applicant for the job.

The researcher interpreted the courts' reasoning and decisions in these early (1996 — 2000) cases to mean that the courts were determining how to incorporate the requirements of a new law within the confines of existing collective bargaining agreements that are governed by another federal agency, the National Labor Relations Board. The ADA provided new remedies — accommodations — that the courts were incorporating as well. The decisions in the three cases (Eckles, Matthews, and Dalton) contained discussions about the relative rights of employers, non-disabled employees, and people with disabilities under the ADA. These discussions provided examples of the court's role as a protector of the rights of employees and businesses versus its role as a mechanism for social change.

The text of the EEOC v. Humiston-Keeling (2000) case, also a 7^th Circuit case, is provided below, both as an example of the court's reliance on the Dalton and Matthews decisions in later decisions:

The Tenth Circuit cases are not distinguishable from the present case, but they are inconsistent with decisions of this court that hold that the Americans with Disabilities Act is not a mandatory preference act. In Dalton v. Subaru-Isuzu Automotive, Inc., supra, 141 F.3d at 679, we held that an employer is not required "to reassign a disabled employee to a position when such a transfer would violate a legitimate, nondiscriminatory policy of the employer. The contrary rule would convert a nondiscrimination statute into a mandatory preference statute, a result which would be both inconsistent with the nondiscriminatory aims of the ADA and an unreasonable imposition on the employers and coworkers of disabled employees." A policy of giving the job to the best applicant is legitimate and nondiscriminatory. Decisions on the merits are not discriminatory. See also Malabarba v. Chicago Tribune Co., 149 F.3d 690, 699-700 (7^th Cir.1998), where we said that "the ADA does not mandate a policy of 'affirmative action in favor of individuals with disabilities, in the sense of requiring that disabled person be given priority in hiring or reassignment over those who are not disabled,'" and Matthews v. Commonwealth Edison Co., 128 F.3d 1194, 1196 (7^th Cir.1997), where we said that "the Americans with Disabilities Act does not command affirmative action in hiring or firing." It is true that antidiscrimination statutes impose costs on employers. That is obvious in disparate-impact cases, when the employer is told to change a policy that may not have been adopted for discriminatory reasons (though that is its effect) and so presumably is efficient. The duty of accommodation operates in a similar way. It requires the employer to incur (if it need be) an expense rather than just to desist from invidious discrimination. The requirement is implicit in the ADA's creating an "undue hardship" safe harbor for employers; the safe harbor would be otiose if the employer's only duty were to stop doing something. But there is a difference, one of principle and not merely of cost, between requiring employers to clear away obstacles to hiring the best applicant for a job, who might be a disabled person or a member of some other statutorily protected group, and requiring employers to hire inferior (albeit minimally qualified) applicants merely because they are members of such a group. That is affirmative action with a vengeance. That is giving a job to someone solely on the basis of his status as a member of a statutorily protected group. It goes well beyond enabling the disabled applicant to compete in the workplace, or requiring the employer to rectify a situation (such as lack of wheelchair access) that is of his own doing. Cf. City of Richmond v. J.A. Croson Co., 488 U.S. 469, 505, 109 S.Ct. 706, 102 L.Ed.2d 854 (1989)." …The EEOC does not deny that in every case the applicant chosen for the job was better than Houser in the sense of likely to be more productive. Nor does it deny that the company had a bona fide policy, consistently implemented, of giving a vacant job to the best applicant rather than to the first qualified one. Nor does it suggest that Houser's disability played any role in the decisions favoring her competitors. None of the jobs involved a degree of lifting that her disability would have interfered with her performing, and it is not suggested that the defendant harbors any animus toward disabled workers. Rather the Commission interprets the "reassignment" form of reasonable accommodation to require that the disabled person be advanced over a more qualified nondisabled person, provided only that the disabled person is at least minimally qualified to do the job, unless the employer can show "undue hardship," a safe harbor under the statute. §12112(b)(5)(A); Vande Zande v. Wisconsin Dept. of Administration, 44 F.3d 538, 542 (7^th Cir.1995). The fact that the disability isn't what makes the disabled person unable to perform the job as well as the person who got it is, in the Commission's view, irrelevant. (EEOC v. Humiston-Keeling).

As mentioned in the discussion under Trend 2, the 7^th Circuit reversed itself in the EEOC vs. Humiston-Keeling and EEOC vs. U.S. Airways cases, demonstrating a shift in the court's reasoning in cases that involve reassignment as an accommodation under Title I of the ADA in situations where a CBA is in place that would otherwise prohibit such reassignment. The text of the opinion in the Humiston-Keeling case highlights the difference between the legal definition of accommodations and the sociopolitical model of disability definition of accommodation. In the court's view, accommodations are a legal requirement with a ceiling and a floor; in other words, accommodations are allowed within a narrow band of situations. In Humiston-Keeling, the court expressed its view that in this case, the reassignment accommodation requested would go beyond what was required by the ADA. In contrast, the sociopolitical model views accommodations a method to increase the participation of people with disabilities in society, including the workplace; therefore, accommodation under the sociopolitical model of disability is viewed as a human right as well as a legal right (Hahn, 1984; Oliver, 1990; Shakespeare & Watson, 2006). Under the sociopolitical model of disability, accommodation has no ceiling or floor. The differing views of accommodation may cause confusion and misunderstanding among various stakeholder groups.

The 7^th Circuit's reversal in the EEOC vs. Humiston-Keeling and EEOC vs. U.S. Airways cases demonstrates the evolution of the law in ADA Title I cases. Although the majority of outcome research on Title I of the ADA has shown that plaintiffs (employees) lose in court in nearly 90 percent of the cases (Colker, 2009), an examination of individual cases demonstrates that courts can and do shift their reasoning over time in a given body of law. The Dalton, EEOC vs. Humiston-Keeling and EEOC vs. U.S. Airways cases provide examples of patterns and trends in ADA Title I litigation that favor plaintiffs.

EEOC Successes

Disability scholars and activists have criticized the EEOC (Krieger, 2003; Mezey, 2005) for not bringing more cases on the behalf of plaintiffs under Title I of the ADA, and for its dearth of successes in the cases it does litigate. Some scholars have noted the ongoing lack of funding for the agency, which results in understaffing and backlogs, as the source of the problem (Krieger, 2003). Other scholars have suggested that the EEOC has not yet been presented with the type of watershed case that Griggs vs. Duke Power was in Title VII cases (Bagenstos, 2009). This study contained two cases in which the EEOC was the charging party: EEOC v. Kronos, and EEOC v. Randstad. The two EEOC cases in this study were administrative subpoena cases and involved large-scale investigations of assessments used in hiring, in the Kronos case, and literacy requirements for hiring, in the Randstad case. The appellate courts in both cases reversed the district courts' decisions to deny EEOC enforcement of its subpoenas. As mentioned, in addition, in 2012, the 7^th Circuit reversed its decision in the EEOC vs. Humiston-Keeling case as a result of precedent set by the U.S. Supreme Court in a similar case, U.S. Airways v. Barnett (U.S. EEOC, 2012).

The EEOC vs. Humiston-Keeling case was reviewed but eliminated from the study because it was brought and litigated as a reasonable accommodations case. Stein and Waterstone (2006) have noted that many ADA reasonable accommodations cases are similar to disparate impact cases. The legal reasoning underlying Stein and Waterstone's assertions is beyond the scope of this study; however, their discussion about the similarities in ADA reasonable accommodation and disparate impact cases provides the rationale for including the EEOC vs. Humiston-Keeling case in the discussion here.

As of the time of this writing, all three of the EEOC cases are ongoing; therefore, it is too soon to know how the EEOC will fare in district court on these issues. However, the favorable rulings for plaintiffs have broad implications for employees with disabilities, as cases charged under the Americans with Disabilities Act Amendment Act of 2008 (ADA AA, 2008) begin to make their way through the courts. The ADA AA 2008 is expected to make it easier for plaintiffs to make their prima facie case (qualified individual with a disability); the expectation is that more reasonable accommodations and disparate impact cases will make it past the summary judgment stage (Wax, 2012). EEOC wins in these two cases have broad implications for plaintiffs in reasonable accommodations and hiring and promotions cases. Hiring practices, policies, assessment tools, and reasonable accommodations are all areas in which disability scholars, advocates, and rehabilitation professionals have knowledge and skills. There is a role for disability scholars, practitioners, and advocates in education and training for businesses, human resource professionals, and clients about the reasonable accommodations process under the ADA, along with information about the types of workplace accommodations available to employees and employers.

In addition, disability scholars may provide an important link between the theoretical perspective posited by the sociopolitical perspective of disability that accommodations foster the goals of access and inclusions, which has the goal of maximizing individual potential and the legal requirement of "reasonable accommodations" under the ADA, which is not concerned with maximizing potential, but with leveling the playing field. Under the legal requirement of "reasonable accommodations, the boundaries of accommodations are limited by case law in ADA Title I cases. The Concept Map (Figure 1) depicts the tension between the two stakeholder perspectives regarding the legal requirement of reasonable accommodations and the sociopolitical perspective, or disability rights view, of accommodations. The courts weigh and balance accommodations as a policy solution or legal requirement with accommodations as a social good that promotes the employment of people with disabilities in making decisions in disparate impact cases that involve reasonable accommodations, such as the Eckles, Matthews, Dalton, and the EEOC v. Humiston-Keeling cases discussed in the previous section.

Sutton Effect

In 1999, the U.S. Supreme Court decided an important case under ADA Title I: Sutton v. U.S. Airlines. The Sutton case was viewed as a huge defeat by disability rights advocates in that it demonstrated a further narrowing of the Court's views on who is a qualified individual with a disability under the ADA. The case was viewed as a way for courts to proceed directly to summary judgment on behalf of the employer, without considering the merits or facts of the case (Stein & Waterstone, 2006). The U.S. Supreme Court in Sutton focused on mitigating measures and individualized inquiry:

In Sutton v. United Airlines, ___ U.S. ___, 119 S.Ct. 2139, 144 L.Ed.2d 450 (1999), twin sisters brought an ADA claim based on their denial of employment as commercial airline pilots. The Supreme Court upheld a dismissal of plaintiffs' claim under Federal Rule of Civil Procedure 12(b)(6) for failure to state a claim on which relief could be granted. Both sisters suffered severe myopia and, uncorrected, their visual acuity was 20/200 or worse in the right eye, and 20/400 or worse in the left eye. However, with the use of corrective lenses, both women had 20/20 vision. The Court determined that neither woman was disabled under the ADA as "the determination of whether an individual is disabled should be made with reference to measures that mitigate the individual's impairment… ." Sutton, 119 S.Ct. at 2143. With glasses serving as a mitigating measure, both women had perfect eyesight; thus, the Supreme Court held they could not claim protection under the ADA. In so holding, the Court rejected the opinions of the Equal Employment Opportunity Commission (EEOC) and the Department of Justice (DOJ), as espoused in their ADA Guidelines, that a disability should be determined without regard to mitigating measures. 29 CFR pt. 1630, App., § 1630.2(j) (1998); 28 CFR pt. 35, App. A, § 35.104 (1998); 28 CFR pt. 36, App. B, § 36.104 (1998). The Court, noting that the existence of a disability is an individualized inquiry, felt that the guidelines approach "would often require courts and employers to speculate about a person's condition and would, in many cases, force them to make a disability determination based on general information about how an uncorrected impairment usually affects individuals, rather than on the individual's actual condition." Sutton, 119 S.Ct. at 2147. (Belk v. Southwestern Bell (1999)).

This study contained four cases in which the court cited the Sutton case in its opinion: Belk v. Southwestern Bell, Lawson v. CSX Transp., Inc., Schneiker v. Fortis, and Gonzales v. City of New Braunfels. In two of the cases, Schneiker and Gonzales, the court affirmed for the defendant on the basis that the plaintiff was not a qualified individual with a disability. However, in the other two cases, Belk and Lawson, the courts provided a detailed description of the individualized inquiry they used to come to the conclusion that the plaintiff was a qualified individual with a disability, and decided both cases in favor of the plaintiff. The court in Lawson made note that under Sutton's individualized inquiry, the court could take the severity of the disability into effect in its decision:

In explaining why mitigating measures should be taken into account in defining an ADA disability, Sutton indicated that "[a] diabetic whose illness does not impair his or her daily activities," after utilizing medical remedies such as insulin, should not be considered disabled. Id. at 483, 119 S.Ct. 2139. This statement does not mean, however, that no diabetic can ever be considered disabled under the ADA's meaning. Such an approach would contradict the Court's view that whether a person is disabled under the ADA is an individualized inquiry based on the particular circumstances of each case. See id. Moreover, as we have explained, the particular nature of Mr. Lawson's diabetes, even after treatment, could be said to significantly impair his daily activities, unlike the situation in Sutton (Lawson v. CSX Transp., Inc., 245 F.3d 916 (2001))

Future Education and Research

There are implications for research and education in these four cases. In cases where there are no damages awarded, such as in disparate impact cases, more research needs to be done to determine if courts are more willing to use the "individualized inquiry" in Sutton to plaintiffs' advantage than they are in cases where monetary damages may be awarded. More research needs to be done on cases in which the plaintiff prevailed. Although they are far few in number than ADA Title I disparate treatment cases , the text of the judicial opinions in the individual cases may contain useful information about the workplace policies at issue, reasonable accommodations, and the nature and severity of the disability.

Disability scholars (e.g., Hahn, 1994; Schriner, 1995) have long called for people with disabilities to become more involved in the policymaking process to gain more control over the development of policies that affect their quality of life. Understanding the process of litigation and how judges make decisions in Title I ADA cases is one way that people with disabilities can become more involved in the development of workplace policies that do not unnecessarily exclude them from participation in the workplace. Becoming more aware of their rights under the ADA is another way people with disabilities can advocate for change. People with disabilities who desire to become more informed about their rights under Title I of the ADA may benefit from using a structured approach, such as the Advocate Model developed by Brophy, Chan, and Mar (1974), which was designed to assist a person to "exert maximum control over his/her own life, while developing a positive self-concept and confidence that this control will be continuous and replicable" (p. 41), the Advocate Model is a three-phase process: (a) education, (b) facilitation, and (c) implementation to move a client from awareness to action (Hartley, Johnston, & Tarvydas, in press).

There may be important differences among ADA Title I, II, and III disparate impact cases. Therefore, this study is being extended to Title II and III cases (see Table 1). Data have been collected and the same mixed methods analysis used in the instant study is being used to examine the ADA Title II and III cases. Further research is necessary to examine what type of viewpoints of disability may be reflected in the legal record of decisions on the ADA. The sociopolitical model of disability posited that negative societal views of disability may perpetuate discrimination against people with disabilities (Hahn 1984; Percy), yet it remains to be evaluated the extent to which this model may play a role in legal decisions, or if some other pattern of constructs may emerge.

[a] Duplicate cases counted as one case
[b] Sandison v. Michigan High School was brought under both Titles II and II and was counted as two cases.

[a] Two appeals in the same case. The most recent case was analyzed.
[b] Two appeals in the same case. The 2012 case dealt with procedural issues (cost-sharing and confidentiality orders) so only the 2010 case was analyzed.
[c] Became U.S. Supreme Court Raytheon v. Hernandez, 540 U.S. 44 (2003).
#Two appeals in the same case; the 2004 case (on remand from Supreme Court) was analyzed
##ADA Reasonable accommodation claim
###ADA Disparate treatment claim
*RA: Reasonable Accommodations; DT: Disparate Treatment; ADEA: Age Discrimination in Employment Act; CA FEHA: California Fair Employment and Housing Act

Indeed, disability scholars have begun to question whether the sociopolitical model of disability is "inadequate to the task of affording people with disabilities the conditions to turn legal rights into realities" (Asch, 2001, p. 391). To that end, Selmi (2008) and other scholars (Bagenstos, 2009; Colker, 2009; Stein & Stein, 2007; Percy, 2000) have argued that civil rights legislation, such as the ADA, cannot and should not be the only means used to bring about equality for people with disabilities.

Other scholars (see, for example, Asch, 2001; Bagenstos, 2001; Scotch & Shriner, 1997; Shakespeare & Watson, 2002; Stein & Stein, 2007) have suggested that civil rights statutes have a limited role to play in advancing the rights of people with disabilities in the workplace. Indeed, Bagenstos suggested that although intentional discrimination remains a barrier to the employment of people with disabilities, other societal barriers, such as lack access to transportation, assistive technology and health insurance are often greater barriers to employment. Bagenstos and others (Asch, Shakespeare & Watson, Scotch & Schriner; Stein & Stein) asserted that disability be viewed as a normal part of the human experience, and called for a renewed focus on the removal of environmental barriers to increase the inclusion of people with disabilities into all segments of society, including the workplace.

The disability literature on the sociopolitical model of disability posited that anti-discrimination laws are an important strategy in addressing societal discrimination, including workplace discrimination, against people with disabilities (Hahn, 1984, 1985). However, the implementation of the law has been inconsistent with the original intent of the legislation (Colker, 2009; Hahn, 2003; Krieger, 2003). It remains to be seen if the ADA Amendments Act of 2008 will produce more favorable outcomes for employees with disabilities who bring lawsuits against employers under Title I of the amended act. Perhaps the time has come to rely less on litigation and more on proactive responses, such as identifying and changing employer practices that may unintentionally screen out employees with disabilities, to meet the needs of a global economy and a diverse workforce.

Limitations

This research examined judicial opinions decided under Title I of the ADA and charged under disparate impact theory. The study and the results of the study have several limitations. For example, this study employed qualitative design and methodology to examine judicial opinions. Qualitative studies do not incorporate the quantitative concepts of random sampling, reliability, and validity in their analysis; therefore, unlike quantitative research, the results of a qualitative study cannot be generalized to the larger population (Patton, 2002). Thus, the results of this research cannot be generalized to judicial opinions not included in this study (Denzin & Lincoln, 2005; Patton, 2002).

This research examined a subset of judicial opinions in ADA Title I cases, specifically disparate impact cases, from the Act's July 26, 1992, effective date through July 31, 2012. "The ADA Amendments Act of 2008 (ADAAA) was enacted on September 25, 2008, and became effective on January 1, 2009" (U. S. EEOC, Questions and answers on the final rule implementing the ADA Amendments Act of 2008, n.d.). This study did not include cases filed under the 2008 Amendments Act. Title I disparate impact cases filed under the ADA AA may be substantially different from cases filed under the ADA of 1990. For example, the ADA AA:

made a number of significant changes to the definition of 'disability.' It also directed the U.S. Equal Employment Opportunity Commission (EEOC) to amend its ADA regulations to reflect the changes made by the ADAAA. The final regulations were published in the Federal Register on March 25, 2011. (EEOC, Questions and answers on the final rule implementing the ADA Amendments Act of 2008, n.d.)

The focus of this study was disparate impact cases, which are a very small percentage of the cases decided by U.S. Appellate Courts under the ADA. In addition, this study limited its analysis to ADA disparate impact cases, which may be materially different from ADEA or Title VII or Rehabilitation Act disparate impact cases. The researcher chose to examine disparate impact cases because they focus on policies and practices that may discriminate against people with disabilities as a group rather than as individuals (U. S. EEOC, 2008). The policy-based or minority group approach is more closely aligned with the minority group approach within the sociopolitical perspective of disability (Gill et al., 2003; Hahn, 1984). However, because disparate impact cases are few in number, they may be materially different from other types of ADA cases, such as disparate treatment cases. Part of this difference may be due to the type of relief available to parties filing disparate impact cases: "Disparate impact cases are not litigated often in ADA settings because there are no damages available for employment claims and because most cases involve individualized assessments" (L. Sandler, personal communication, October 6, 2008). Damages are "the money that a person may recover at law when he has been harmed" (Mellinkoff, 1992, p. 147).

This study used computer-assisted legal research (CALR) to locate the cases to be analyzed in this study. The researcher chose Westlaw, a subscription-based or commercial electronic database, to identify the cases that comprised the sample in this study. Westlaw, along with LexisNexis, is one of the most widely used commercial electronic databases for legal research in both U.S. law schools and legal practice. One advantage of using electronic legal databases such as Westlaw is their ability to compile large amounts of court records in a searchable format to save legal researchers time and improve accuracy (Hall & Wright, 2008). However, there is a lack of standardization across databases; for example, Westlaw uses Keycite and LexisNexis uses Shepards, respectively, as citation systems. In addition, the cases contained in both Westlaw and LexisNexis databases are drawn from published print volumes, known as reporters. The reporters are organized by either jurisdiction or topic. ThomsonWest, the publishing arm of Westlaw, exercises editorial control over the content of its reporters; therefore, a limitation of using commercial reporters is that not every case is included in the reporters. In addition, some cases are not released for publication by the court. These "unpublished" cases are not precedential and cannot be cited as such. ThomsonWest includes unpublished cases in its online database and compiles them in its Federal Appendix reporter. Approximately 80% of federal court of appeals cases are unpublished (Gerken, 2004). This study examined both published and unpublished ADA disparate impact cases; however, it is important to note that there may be material differences between published and unpublished cases.

Another consideration is the type of judicial opinion. For example, some opinions are per curiam (Latin, "through the court"). In these cases, the authors of the majority opinion are not revealed; however, the name of the author (or authors) of the minority or dissenting opinion is revealed in per curiam cases.

Finally, although the researcher has 13 years' experience as a court reporter, and is therefore familiar with legal terminology and the process of litigation, it is important to note that the researcher has no formal training in the law and is not an attorney. Thus, the researcher views the law and cases discussed within this study from the perspective of a disability and rehabilitation scholar and educator rather than as a legal scholar.

Conclusion

The findings in this study may inform disability scholars, rehabilitation and legal professionals, people with disabilities and advocates about ADA disparate impact cases, a body of law about which little is known. The patterns and trends identified in ADA Title I disparate impact cases may provide information about the types of policies and practices that are most frequently litigated. The results of this research may be used to develop education and outreach strategies for employers on best practices for hiring, accommodating, and promoting employees with disabilities. Because the workplace policies and procedures identified in the cases that comprise this study are neutral on their face rather than intentionally discriminatory, employers may benefit from information that assists them in evaluating their policies and procedures proactively, which may avoid costly and time-consuming litigation. Workplace policies and procedures that fairly represent the essential functions of the job and are applied uniformly to workers with and without disabilities will contribute to a more diverse workforce. Finally, the data generated from this study may provide disability scholars, people with disabilities, and disability advocates with more detailed information about the nature of disability-related characteristics contained in judicial opinions; may expand the disability community's understanding of the words judges use when writing about disability in ADA Title I cases charged under the theory of disparate impact; and may provide information that will assist the disability community in becoming more conversant with the complex relationship between civil rights legislation and its implementation through the court system.