Passed in 1990, the Americans with Disabilities Act (ADA) is one of the most comprehensive disability rights laws in the world. The passage of the ADA is significant because it addresses the exclusion that people with disabilities face in the workforce due to their physical and mental differences. Specifically, Title I of the ADA protects the rights of people with disabilities in the workplace. The law states that employers cannot discriminate against qualified people with disabilities in employment decisions regarding hiring or firing. In addition, the law requires employers to provide "reasonable accommodations" to qualified employees with disabilities, if they need them to perform their jobs. In 2001, Congress extended the hostile work environments doctrine of Title VII of the 1964 Civil Rights Act to protect individuals covered under the ADA (Robert and Harlan 2006).

In spite of the ADA protections, discrimination against people with disabilities still persists in workplaces. Attitudes about disability also can become embedded in organizations, as Schur, Kruse and Blanck (2005) have shown, corporate culture can create attitudinal, behavioral, and physical barriers for workers and job applicants with disabilities. In another study, Schur (2002) found that employed men and women with disabilities were significantly less likely than their non-disabled coworkers to have given a presentation or speech at work or participated in decision making. Workplaces are often physically inaccessible for people with physical disabilities, and people with disabilities are socially marginalized in relation to their colleagues. The construction of workplaces that do not account for the bodies and minds of workers with disabilities tends to naturalize the idea that people with disabilities do not belong in certain jobs, which contributes to their disadvantage relative to the "able-bodied." This article examines how when workplaces and the duties of jobs are constructed in such exclusionary ways, the ADA is not enough to ensure the full inclusion of workers with disabilities.

Operationalized by job type, class has been shown to have a particular impact on the experiences of workers with disabilities. Harlan and Robert (1998) found that people in higher-grade jobs, who knew of the ADA, were more likely to ask for and receive accommodations. Employers denied women's requests for accommodations more often than men's requests. Moreover, employers were more likely to deny the requests of people in lower grade jobs (Harlan and Robert 1998). Generally women's jobs were in lower grades than men's jobs, and the type of impairments women had was more incompatible with the work they did. For example, they found that the men often had mobility impairments, which did not seem to interfere with their ability to perform "white-collar" work. By contrast the women tended to have chronic illnesses (diabetes, arthritis) which made it difficult for them to "conform to the rigorous regimentation of lower level clerical jobs" (Harlan and Robert 1998:414). Therefore, they found that women were more likely than men to request accommodations.

These gender differences illustrated by Harlan and Robert may also reflect a class issue. Joan Acker views organization controls as class controls in that they maintain the power of managers over workers (Acker, 2006). The process of asking one's boss for accommodations and his/her power to either grant or deny these requests is an example of these controls. In addition, Schur (2003) found that people with disabilities are almost twice as likely to work in part-time contingent jobs. While, as Schur points out, this can be a benefit to people whose disabilities require them to work more flexible schedules, such contingent jobs usually pay very little and incumbents often have very little power over their schedules.

Schur's (2003) findings mirror research that has been done on women in contingent work. Women occupy a much higher percentage of contingent jobs than men. This has been understood as a result of women's need for more flexible schedules in order to balance their family demands with their paid work. Spalter-Roth and Hartmann (1998), however, offer an alternative explanation: employers make decisions not to invest in female workers and steer them into part-time contingent work. In fact, contingent workers almost never have any control over their schedules and the majority of the women in contingent jobs do not have the "breadwinning" spouse assumed to make contingent work a "choice" for women (Spalter-Roth and Hartmann 1998). In fact, this group of women workers was more likely to make ends meet through welfare benefits.

Similar to the female contingent workers, the same pattern also possibly happens to people with disabilities, particularly women with disabilities. Robert and Harlan (2006) found that more than half of the women interviewed reported being steered into entry-level positions like clerk and other traditionally female jobs. At the same time, more than three-quarters of men with comparable qualifications reported being placed in jobs with better career ladders. These studies illustrate how women with disabilities find themselves in lower grade jobs, which may make them more vulnerable to discrimination based on their disabilities.

In a study of reasonable accommodations in organizations Harlan and Robert (1998) drew upon Acker's organizational theory to explore the issue of workplace accommodations. They interviewed employees with disabilities about their experiences asking for accommodations. They found that employers were more likely to approve requests for changes to the physical space than requests for changes in work schedules or job functions. This could also be interpreted as a class issue. As discussed above, people in lower level jobs have less control over their schedules, and these results may reflect the finding that people in lower grade jobs are less likely to have their requests for accommodations granted. Schartz et al (2006) point out that employers tend to perceive accommodations for disabled employees as costly. They also find that workplace accommodations, however, are effective and inexpensive.

Harlan and Robert's (1998) interviews also revealed a number of social barriers that made it difficult for employees with disabilities to ask for or receive accommodations. The researchers noted that people who ask for accommodations face a double-edged sword. Specifically, they have to prove their disability to receive accommodations, only to subsequently experience stigma attached to their disability. This is especially problematic for people with nonvisible or non-apparent disabilities.

In a study of workers with disabilities in government organizations, Robert and Harlan (2006) found that many of their respondents reported harassment. Often the workers with disabilities found that their supervisors would not take their claims of harassment seriously and had the attitude that the disabled worker should just put up with it. Another form of harassment occurred when coworkers felt they could delve into the personal lives of the worker with disabilities, asking them deeply personal questions and judging their personal decisions (Robert and Harlan 2006). Finally, the workers were harassed when their requests for reasonable accommodations were seen as "special treatment," particularly when they had to participate in the delivery of these accommodations (Robert and Harlan 2006). In an ADA discrimination law suit, Robel v. Roundup Corp., coworkers harassed a woman because she had filed a worker's compensation claim for a back injury. and the employer placed her on light duty work. The disability request for accommodations motivated the harassment, but the harassment also was gendered in nature as her coworkers called her derogatory words, typically used against women, such as "cunt" and "bitch."

Theoretical Foundations- Gendered Organizations

Joan Acker (1990) has theorized that jobs are created within organizations with assumptions of a disembodied and universal worker. But in fact, Acker argues, this worker is implicitly male, and therefore men's bodies, sexualities and relationship to procreation and paid work are assumed in the image of the worker. Acker's insights can apply to disability in the workplace as well. The disembodied worker is not only comprised of a male body, but also an able-body. Cockburn (1993) briefly explores the issue of disability in the workplace. She found that employers tend to disadvantage disabled workers by failing to remove barriers for them and by underestimating them (Cockburn 1993: 209). In furthering Acker's concept of the disembodied worker by including disability, she states, "we measure ourselves against a model of physical perfection in contrast to which impairment is seen as a curse, an omen, a source of shame. The model [worker] is male and women's bodies are seen as less strong, less effective. The model is white. The model is physical fitness" (Cockburn 1993). She asserts that people with disabilities must claim their bodies and make them visible. It is important to challenge the ideology that disability comes from 'acts of God' and show that handicaps are created and perpetuated by organizational structures, which act as barriers to disabled people's full participation in society and especially in the workplace. The presence and visibility of disabled people in workplaces is important for challenging organizational bias but this cannot happen without fair hiring practices and access to reasonable accommodations.

Iris Marion Young's (1990) work dovetails with gendered organization theorists in that she argues that inequality is embedded in institutions, which have been built to exclude certain people. Young takes issue with ideas that people can achieve equality on the basis of everyone having an equal opportunity to compete for the same jobs. For Young, injustice is defined as oppression and domination and therefore cannot be solved without at least some restructuring of institutions and organizations. In her discussion of affirmative action, Young firmly believes that differential treatment is essential to achieving equality in the workplace. Differential treatment in some cases means providing reasonable accommodations. The purpose of these accommodations is to restructure work environments and work demands to make them more accessible. Young seems to raise the stakes from the organizational theorists in that she is concerned with the inequality embedded not only in organizations themselves, but also in the way we conceive of work in the first place. For Young (1990:200), workplaces are organized in a "natural" hierarchy of intellect and skill. Underlying this is an assumption that the most competent and hardest working people will get ahead. This affects the image of disabled people in the workplace because the perception of people working with an accommodation is that they are not truly 'competing' and that they are naturally incompetent (Wilkerson and Frieden 2000).

The combined works of Acker, Cockburn and Young suggest that the experiences of both women and people with disabilities are embodied in problematic ways in workplaces built on the assumption of an able-bodied male worker. In addition, these differences can intersect with class (operationalized as job type), having an impact on the experiences of people with disabilities in the work place. If work places are built for an able-bodied male, then how can disabled women gain access to these workplaces? How might class intersect with disability in ways that make certain workplaces less accessible than others? Women with disabilities have been shown to experience an extra disadvantage, one that appears heightened based on class. Research has shown that people with disabilities in lower grade jobs are more vulnerable to discrimination and less likely to receive requested accommodations. I examined the differences in types of discrimination cited in ADA law suits between job types (of higher and lower class) especially how they intersect with gender, as women with disabilities are even more likely to hold lower class jobs than men with disabilities. While researchers tend to look at gender and disability as an additive effect of gender discrimination PLUS disability discrimination, they have not explored how gender and disability interact to create a unique experience of discrimination for disabled women. By intersecting gender with disability type in my analysis, I intend to find whether women with disabilities have a unique experience of discrimination that isn't accounted for in additive models.

Methods

I use court cases as data to study workplace discrimination because they document instances where discrimination has allegedly occurred. Using the Westlaw database, I can code a sample for quantitative analysis as well as conduct a content analysis of the information contained in the summaries. The Westlaw database is one of the largest databases of primary law, which lawyers and academics use. Each case in the Westlaw database has summaries that tell the story of the discrimination event and gives the perspective of the employee and the employer. Some cases are more descriptive than others and seem to tell the "story" of the plaintiff's experience. These stories can read almost like interview data because they outline the events that occurred to make the plaintiff believe he or she was being discriminated against on the basis of disability. I use the Americans with Disabilities (awd-cs) data set that includes all federal cases, which have been determined to be related to Americans with disabilities.

I chose a random sample of 200 cases from the results yielded from a search using broad search terms such as: Americans with Disabilities Act, Americans with Disabilities, ADA, Title VII, discrimination, work, employment, and job. I limit my search to cases that occurred between 2012 to July of 2013 in order to read my results as an indicator of the contemporary situation. My initial search produced 586 results. I divided that number by 200 (my desired sample size), which returned a result of 2.93. This means that in order to get a sample of 200 cases from this sample I will have to select every 3^rd case. I used a random number generator to select my starting point, which was 7. I started from the seventh case in the results. If a case was not relevant to my study, I skipped it and went on to the next case.

I manually coded each case into an Excel spreadsheet. By reading through the synopsis and digest of each case, I coded the information I needed from the text of the documents. Some cases were arguing procedural matters, so I skipped those in favor of cases that were presented as civil rights cases. Scholars have used the Westlaw database to examine the outcomes of court cases. This project is unique because I am not coding for legal matters but rather conducting a content analysis in order examine discriminatory processes in the workplace. The variables in this study are gender, job type, discrimination type and disability type.

Once the sample of 200 cases had been coded, I analyzed the data using SPSS (version 21), a statistical software package. I used multiple regression models to test which types of discrimination people with different types of disabilities, jobs or genders are likely to experience. This is important because it will help us understand which populations of people with disabilities are more vulnerable to certain types of discrimination. Discrimination type was the dependent variable and was coded for discrimination regarding firing, accommodations, hiring, harassment, multiple forms of discrimination. I used "fired" as the reference category and regressed type of discrimination with the independent variables for job type, gender and disability type. Job type was coded as working class, semi professional and professional; gender was coded as a dummy variable with female as 1 and male as 0; disability type was coded as physical impairments, nonvisible mental disabilities, nonvisible pain disabilities, chronic illness, and other.1 I also created interactions terms for gender and job type.

Results

The first regression model tested the effects of gender, job type and disability on the type of discrimination that was claimed. The expected relationship between job type and discrimination was supported by this model. First, people in semiprofessional jobs were about 64% less likely than people in working class jobs to report being denied an accommodation rather than being fired. People in semiprofessional jobs were also 85% less likely than people in working class jobs to report being harassed compared to being fired. This situates people in working-class jobs as more likely to be denied an accommodation and be harassed in the workplace rather than being fired. The women were also more likely to report being denied an accommodation than being fired.

In the next model I tested the interaction term for women in working class jobs. The interaction between gender and job type was confirmed in this model as I found a significant interaction for women in semiprofessional jobs under the category "denied accommodation." The interaction term for women in semiprofessional jobs was significant in the "denied an accommodation" category, showing that they were less likely than women in working class jobs to report being denied an accommodation compared to being fired. Women in semiprofessional jobs, then, may be more vulnerable to being fired than being denied an accommodation. Conversely, women in working class jobs then, are significantly more likely than women in semiprofessional jobs to report being denied an accommodation. This is not an additive effect of women plus working class, but rather it is a unique intersectional effect, which affects women with disabilities in working class jobs.

People in working class jobs tend to have very little autonomy or control over their schedules. Women in working class jobs tend to work in clerical positions compared to men in working class jobs who may work in jobs, which tend to be unionized. This is important in the case of accommodations where an employee has to request an accommodation from their supervisors who have to the power to either grant or deny the request. For people with disabilities, it is more difficult to have accommodation requests granted in jobs with very little autonomy. Unions can offer some protection of the rights of the workers but these protections are not available to women in many clerical positions.

In addition, people in working class jobs were more likely than the other job types to report harassment as the type of discrimination they experienced rather than firing. Further, people with nonvisible mental disabilities were more likely to experience harassment than firing compared to people without nonvisible mental disabilities. Mental disabilities are a highly stigmatized form of disability, which may be why people with those kinds of disabilities are more vulnerable to harassment in the workplace.

Qualitative Analysis

For the qualitative data there is a sample of 34 cases that were taken from the sample of 200 used for quantitative data. I coded each case using the same variables I used in the quantitative portion of the study. I examined the cases for themes that emerged in the plaintiff's experiences of discrimination. The quantitative data illustrated some of the trends in discrimination when viewing discrimination through the broad categories of firing, harassment, denied accommodations. However the qualitative analysis was effective in uncovering the multiple processes involved in these broad categories of discrimination.

In this sample, the majority of respondents report being fired, though this is even more frequent for the men (14 out of 16 men and 10 of 16 women). What the qualitative analysis shows are two common processes by which employers fired people with disabilities in this sample: through a restructuring of jobs in the company, and by being deemed unqualified for the job due to disability. Restructuring of jobs refers to the way that jobs may be rearranged in a company so that some positions are eliminated. Regardless of job type, people with disabilities in my sample found themselves lost in this shuffle and without their jobs as a result. Such reshuffling allows employers to let go of workers with disabilities whom they may perceive as less valuable employees for what appears to be nondiscriminatory reasons.

For instance, a tractor operator was working "swing-shifts." After he was diagnosed with fibromyalgia, his doctor recommended he switch to a straight shift because the swing-shift would exacerbate his symptoms. He was able to get a straight shift job until the company restructured; this change meant that one straight shift tractor operator had to become a rotating shift worker. Regardless of his disability restrictions, this man was transferred to the rotating shift job because he was the least senior of the tractor operators. He was laid off when, after two weeks of working the rotating shift, he submitted another doctor's note requesting a straight shift. There were no more straight shifts.

In a similar case, doctors gave strict lifting restrictions to a man who was an aircraft mechanic with a shoulder injury. He was accommodated by being reassigned to the parts-turn-in job, which he could perform with his restrictions. Later, however, the company restructured their scheduling, and this position was no longer available on his new schedule. He was offered the position of aircraft monitor, which paid less than his previous job. Ultimately, then, his choice was between losing pay or losing his job. His injury became worse and his restrictions kept him from being able to do even that job and the only job he was able to do, parts-turn-in, was still not available on his schedule.

While the previous examples are of men in working-class jobs, similar narratives also emerged for women in semiprofessional and professional jobs. In one case, a woman was a vice president at a bank. After taking some time off to deal with breast cancer, she came back to find that the bank was streamlining and looking for people to layoff. She was chosen for a layoff, no one was hired to replace her, and her job duties were distributed among the remaining employees. In another case, a bookkeeper with multiple sclerosis found that after taking time off to get treatment for a flare-up of her disease, the company was restructuring and eliminating her position. The company outsourced her duties to an outside bookkeeper. She claimed a coworker said he overheard her supervisors discussing whether the cost of her payroll was "worth it," given her condition. Indeed, she was one of the highest paid employees in that office.

Another important process by which people with disabilities may be fired is seen in many of the cases, especially in physical jobs, in which the person is fired because the restriction of their disability makes them "not otherwise qualified for the position." Use of this language is geared toward the stipulation in the ADA that, to qualify for protection under the ADA, someone has to be qualified for the job and able to do it with accommodations. Employers in my sample have used this as grounds to fire people with disabilities, if their disability restrictions mean they aren't able to do the duties of the job. This can be tricky when the duties of the job are purely physical and the disability restrictions also are physical. An employer only has to accommodate someone who is qualified for the job, and in such a situation, the employer can refuse to accommodate, because the employee is not qualified for the job. If their medical restrictions seem to go against the duties of the job, then they are deemed not qualified for the job and the employers also could argue that they are not qualified under the ADA.

For instance, a woman working in a factory who had developed carpal tunnel syndrome was restricted from doing repetitive motions with her hands. According to her employer "over two-thirds of the position involved using hands 'to finger, handle, and feel' and reaching with hands and arms (Ivey v. First Quality Retail Serv.)." When she requested an accommodation to switch tasks every 15 minutes, the employer denied her request and fired her, because she could not perform her job. The employer further stated that they had no obligation to either accommodate or employ her on the basis that she was no longer qualified for the job. In this instance, the person's disability, which she acquired at work, not only disqualified her from her job physically, but also disqualified her from ADA protections, which only protects "qualified" laborers. Therefore, accommodations requests potentially risk not only the job, but also coverage of the ADA.

Similarly, a man who was a locomotive machinist was ordered to see an eye doctor, when his coworkers complained that he seemed to have trouble seeing. Doctors diagnosed him with a degenerative vision disability. He received several restrictions including, "no walking on uneven surfaces; no more than occasional bending or stooping; no operating vehicles or machinery; no climbing ladders or scaffolds; no working on unprotected heights; no more than occasional lifting of greater than 20 pounds; and no job that requires more than 15 degrees visual field (Hohn v. BNSF Ry. Co.)." Although the company supervisor was willing to find him another job within the company, the company doctor ruled that all the available jobs clashed with the medical restrictions. Therefore, he lost his job. Here, while his restrictions went against the basic duties of the various jobs available, the employer/company never discussed how to potentially modify available jobs to accommodate his restrictions. This kind of disqualification seems particularly salient in working class, physically demanding jobs. In this class of jobs, it is all too easy for an employer to dismiss someone with a physical disability. In both themes, restructuring and disqualification due to disability, the employer is able terminate the employment of a person with a disability, while at the same time claiming they did not fire the person.

One important finding in my analysis is that more men than women claim being fired as the sole form of discrimination they face. As I noted above, the majority of women say they were fired; but in addition, women are more likely to cite multiple forms of discrimination. It may be that while men with disabilities are more at risk for losing their jobs, women with disabilities may suffer more negative treatment during their employment.

Half the women (8 of 16) in the sample reported discrimination that related to being denied a reasonable accommodation, compared to only one of the men. This sample, however, was too small to find the class effects for women that were found in the quantitative analysis.

Women claimed harassment or retaliation by more than half (10 of 16) in the sample, compared to 5 of 16 for men. This means twice as many women than men claimed to have been harassed in the workplace. Of the 10 women who claimed harassment, 6 claimed to have been harassed based on their disability. In addition, three claimed they were sexually harassed, 2 women claimed they were retaliated against, and one woman alleged both sexual and disability harassment.

There were only 3 women in the sample from working-class jobs, and all three of them claim to have been harassed on the basis of their disabilities. For instance, one woman who was a postal employee and who hurt her back on the job describes an incident, where her supervisor bullied her into signing an agreement to work five hours a day instead of the four hours she had been working as an accommodation (workers' compensation supplemented her pay). The supervisor kept her in his office and would not let her leave, until she signed. Another woman working in a factory reported that her co-workers laughed at her and called her "Robocop" because of the brace she wore on her wrist. This supports the quantitative finding that people in working class jobs are more likely than the other job categories to report being harassed than being fired.

Such instances were not confined to women in working-class jobs, however. Two women in professional jobs reported harassment by their supervisors because of their chronic illnesses. In addition, a woman with a mental disability, who worked in a semi-professional job, reported that private details of her medical condition were released to her co-workers. Those co-workers then used this information to intensify the harassment in which they were already engaged.

Of the three women with disabilities who were sexually harassed, two were in semi-professional jobs and one was in a professional job. In one case, a woman with multiple sclerosis worked as an employee concerns coordinator at a power plant. She reported that her co-workers sexually harassed her with anonymous reports to her that contained vulgar comments about her body and about her husband, who also worked at the plant. Her co-workers also sent prank phone calls and posted personal documents to humiliate her. Another woman with physical and mental disabilities, who worked as a lab assistant, reported that her coworkers referred to her in a sexually derogatory manner and physically harassed her by intentionally bumping into her. In another case, a woman in a professional job, who was an executive position, claimed sexual harassment and sex discrimination. Although the brief did not elaborate on her sexual harassment claims, it does outline the incident, which led the employer to fire her. The employer fired her after an argument erupted with her supervisor, where the supervisor said, "sit down little girl," and when she refused, the supervisor told her to "get out."

Of the five men who claimed harassment, 3 were harassed based on their disability. Of those three, two were men with intellectual disabilities. Both of these reported being called names like, "dummy" and "retard" by their bosses. Their bosses also encouraged the coworkers to participate in the name-calling. The employers treated the men very badly. One, a church janitor, worked twenty-five hours a week until the other janitor left. He then took over all the janitorial duties, working 40 hours, but his employer offered to only pay him for 25 hours' worth of work. The other man worked construction, and his boss would repeatedly physically harass him with "titty twisters" and "Charlie horses" even after he asked his boss to stop. Finally, the boss shoved him, and when the disabled man swung back at him he knocked off the supervisor's respirator mask, which he was wearing for workplace safety. The company fired the disabled man for this action, while the supervisor, who instigated the incident, was merely put on suspension. These cases also illustrate the effect of class on harassment in the workplace found in the quantitative models. The exclusion of people with disabilities in working class jobs is made to seem natural due to the physical demands of the jobs. For this reason, people with disabilities likely are more vulnerable to disability-based harassment in these settings where they appear not to belong.

Qualitative Summary

I framed this project by feminist theories of gendered organizations (Acker 1990, Cockburn 1993), which argue that jobs in organizations are created for what is assumed to be a universal worker but is actually an able-bodied man. Further, Iris Young (1990) argues that society cannot overcome these biases through equal treatment. Rather, organizations need to restructure and oppressed groups, like people with disabilities, require special treatment to achieve equality. I expected to find that women with disabilities would have a unique experience of discrimination because of both the gendered and able-bodied norms embedded in organizations. My quantitative and qualitative analysis has supported this expectation in part and has also revealed intersections between disability, gender and class.

The discrimination category, "fired" was broken into two sub categories when I examined it through qualitative analysis. This revealed two mechanisms of discrimination that one cannot see when analyzing the category of firing more generally. In both of these mechanisms, restructuring and disqualified due to disability, the employer is able to terminate the disabled employee for seemingly nondiscriminatory reasons. In addition, the class of job becomes significant in the category of not qualified due to disability. Here, employers in working class jobs that are more physical in nature are able to fire people whose disabilities cause them to have physical restrictions. In other words, an employer has to accommodate someone with a disability who is qualified for the job. In more professional jobs, qualifications could take the form of certification or education. However, physical, working class jobs where the qualifications are physical leave people with disabilities more vulnerable to losing their jobs. This distinction also makes people with disabilities more vulnerable during periods of widespread unemployment when most of the few jobs available are these kinds of physical jobs that tend to be unaccommodating to people with disabilities. While it seems like a simple proposition that physically impaired people cannot do physical work, there is clearly organizational biases embedded in these decisions about how the work is supposed to be done and what kind of worker is supposed to do it. For example, the woman with carpal tunnel syndrome, who worked in the factory, had proposed an accommodation that she would switch tasks every 15 minutes to avoid repetitive motions that would exacerbate her condition. Not only did the employer fire this woman because her disability rendered her unqualified for the job but they also refused the suggested accommodation in part because no one has ever done the job that way before. Organizational biases make it almost impossible to imagine alternative ways of doing jobs that make these jobs available for people who have disabilities which otherwise appear to disqualify them for the job.

When the disqualification of people with disabilities is made to seem natural through organizational bias, people with disabilities who do work in these places potentially face hostility from their coworkers and supervisors. The quantitative data show that people with nonvisible mental disabilities are more likely than people without nonvisible mental disabilities to report being harassed as the type of discrimination they experienced. People in working class jobs were more likely to report harassment rather than being fired. The qualitative data also showed that there is an effect of class on harassment. Although harassment occurred across job types and took different forms such as, sexual or racial harassment, harassment based on disability concentrated in the working class jobs for both men and women. The women that claimed disability harassment all worked in working class jobs and had physical or pain disabilities, which meant that they either had to work with accommodations or appeared unqualified for the job. Of the men who had working class jobs, two were harassed for mental disabilities and one was harassed for his illness. It appears that there is a class effect in harassment, which indicates that people with disabilities are potentially more likely experience harassment for their disabilities, when they are in working-class jobs for which, due to organizational biases, they may appear unqualified. The literature reviewed examines harassment in the workplace against people with disabilities, but the literature did not specify the differences in harassment based on class that this study illuminated.

The literature had suggested a class difference where women in lower level clerical jobs had disabilities that were less compatible with the duties of their jobs than men in "white-collar" jobs whose mobility impairments did not interfere with their work (Harlan and Robert 1998). By including clerical work in the category of working class jobs, my data supported this interaction. Harlan and Robert were right to point out that certain disabilities potentially are more or less amenable to certain kinds of jobs. In their study, these differences were observable based on gender and class of jobs. In my study, I found that physical, blue collar jobs were less accessible to both men and women who had pain or physical disabilities, who were harassed or terminated. My study demonstrates that what is poised at this intersection is an able-bodied bias that informs the way that we conceive of work in the first place. When disability restrictions disqualify someone from being able to perform the essential duties of the job, then perhaps the essential duties need some restructuring. The protections of the ADA may not be enough until we restructure the way that we perceive work and what appropriate workers should be like.

The qualitative data support a gender and disability intersectionality. Twice as many women reported being harassed in the workplace than men. The harassment experienced by women with disabilities sometimes took the form of sexual harassment. This could indicate that women with disabilities may be more vulnerable than men with disabilities to harassment in the workplace. In addition, some women claimed both disability and sex discrimination in their law suits. This supports the theoretical expectation that women with disabilities may be made to appear disqualified due to organizational biases based on both gender and disability. The literature reviewed showed that harassment was a common type of discrimination experienced by people with disabilities (Robert and Harlan 2006). However, the literature did not specify differences in harassment between men and women with disabilities as this study has shown.

This study is different from other studies on disability discrimination because it uses an intersectional approach to examine the experiences of disabled women. Other studies have used court data to study disability discrimination but none have done so with a focus on gender. This study has found important intersections between disability and gender as well as disability and class. This is important because it can contribute to disability policy which affects the lives of disabled men and women in working class jobs and women with disabilities more generally. In addition, the qualitative findings of the ways in which employers terminate people with disabilities were not found in the literature. This is due in part to the unique nature of this study. Some studies interviewed people while they were employed and thus would not examined termination. Other studies, used similar data to this study, which includes termination, but examined those data quantitatively and therefore missed the different processes through which employers fire people with disabilities. These processes of termination should continue to be studied in more detail.

Further, this work can contribute to work in women's studies, which deals with intersectional or interlocking matrices of oppression. Intersectional approaches have been used to show how race and gender combine to create distinctive opportunities for all groups (Brown and Misra 2003). While intersections of race and gender have shed light on inequalities in the labor market, disabled women's experiences also can shed light on organizational barriers as well as what counts as an appropriate body in the workplace.

Limitations

This project has some limitations. First, sampling from federal court cases does not give a rounded picture of the experiences of people with disabilities in the workplace more generally. This does not represent a sample of people with disabilities more generally, rather it is a sample of those who sue for discrimination. In addition, many people who feel discriminated against do not bring their cases to court, therefore this sample represents only a particular group of people. These people may be more litigious or more likely to have the resources needed to bring a law suit. The sample was too small to be able to examine various combinations of discrimination such as fired and denied an accommodation therefore a more generalized "multiple forms of discrimination" category was used. Also, with a bigger sample, it may be possible to make more precise disability categories rather than merging the categories as I did here. In addition, starting the project from quantitative analysis may have primed me to view my qualitative data less inductively, because I was already exposed to broader trends in the data. Given that limitation, I did find themes in the qualitative, which were grounded in those data.