Congress enacted the Americans with Disabilities Act (ADA, 1990) to combat barriers faced by those with disabilities (Blanck, 1996). Title I of the ADA states that employers cannot deny a position to a current employee or applicant because of a disability, as long as the person is able to perform the essential job functions with reasonable accommodation(s). To enable a qualified person with a disability to perform job duties, employers are required to provide reasonable accommodation(s) upon request. The employer does not have to provide an accommodation that would present an “undue hardship” to the company. The law requires a case-by-case analysis of whether an accommodation creates a hardship but is vague about the factors that should be considered in the analysis, providing ample opportunity for extralegal factors to influence the decision-making process.

EXTRALEGAL INFLUENCE ON ACCOMMODATION DECISIONS

The Equal Employment Opportunity Commission (EEOC, 2002) specified that a decision to refuse accommodation because of undue hardship is to be based solely on the financial resources of the company, not the individual's status, position, or salary. However, people have their own perceptions of commonsense justice, which may lead them to use the individual's characteristics in their assessments of what is an undue hardship (Finkel, 2001). Several individual characteristics may contribute to an assessment of whether to grant a worker with a disability reasonable accommodations, such as attributions of responsibility for disability origin or work history. Although there is no case law that addresses the legal appropriateness of the EEOC guidelines, a careful reading of these guidelines suggests that, even though work history is a legal consideration for hiring decisions, neither work history1 nor disability origin is a legally permissible factor in determining reasonable accommodations. Legal representatives of the EEOC also report that these characteristics are legally impermissible considerations in this context (R. Shonfield, personal communication, April 25, 2003).

Attribution of Responsibility for Disability

If people view an event as caused by the person, they attribute the event to something internal to the person; if they believe the situation or environment caused an event, they will make an external attribution for the event (Heider, 1958; Weiner, 1979). Attributions about the cause of a person's need influence willingness to provide help, with people who need help due to external causes being more likely to receive the requested aid than those who are thought to have contributed to the events that created the need (i.e., Barnes, Ickes, & Kidd, 1979; Meyer & Mulherin, 1980). Causal attributions about the cause of a disability also affect the employment options of people with disabilities. Participants were more likely to say they would hire individuals who provided an external cause for their disabilities than individuals who provided an internal cause for their disabilities or who did not provide a cause (Bordieri & Drehmer, 1986, 1988).

Most past research regarding employment-related interactions with individuals with disabilities has focused on hiring decisions. Moreover, although recent research has begun to investigate the accommodation decisions in the corporate culture (see Blanck & Schartz, 2005), much of the existing research on accommodation decisions has used a global approach, as opposed to the case-by-case assessment required by the ADA. One study had employers check off accommodations from a list, without considering if the accommodation was reasonable for a particular disability or a particular individual (Roessler & Sumner, 1997). The failure to examine accommodation decisions in an individualized context is a major oversight in the existing research on reasonable accommodations. Attribution theory suggests that two individuals with identical disabilities and job descriptions working for the same company (or companies with the same financial standing) may receive different accommodations, if one of them has an externally caused disability whereas the other's disability is internally caused.

Work History

Work history influences employment decisions about individuals with and without disabilities. Not surprisingly, a good work history leads to more favorable hiring recommendations than a poor work history (Pradez de Faria & Yoder, 1997; Struthers, Colwill, & Perry, 1992). Stone and Sawatzki (1980) found this was the case even if the individual was depicted with a disability. Graduate level business students given applications with a positive work history “hired” those individuals more often than those given applications with a poor work history, independent of disability. Thus, any individual who has an excellent, or at least good, prior work history is more likely to be hired than an individual who has a poor prior work history. The question remains whether an employee's work history influences an employer's willingness to grant accommodations to a current employee with a disability. As long as an employee's work history is satisfactory, EEOC representatives state that it should be irrelevant to accommodation decisions according to their guidelines.

Study One

We examined whether an internal versus external attribution of responsibility for a disability and an individual's work history influences participants’ willingness to make accommodations. Within a summary of the individual's request for accommodation, we manipulated attribution of responsibility by depicting the cause of the disability as either an accident caused by someone else or an accident caused by the individual. We manipulated work history by depicting the individual as either an outstanding employee or an average, yet satisfactory, employee.

We expected that this study would reveal that impermissible factors influence decisions about reasonable accommodations, with participants granting more, as well as costlier, accommodations to those individuals having an externally caused disability than to those having an internally caused disability. Similarly, participants should grant more and costlier accommodations to employees with an outstanding work history than to those with an average, yet satisfactory, work history. Finally, we expected that disability origin and work history would interact to influence accommodation decisions such that disability origin would be less influential when the work history was excellent than when the work history was average. Employers might be motivated to expend more resources to keep an excellent employee in the company than an average one, regardless of the cause of the disability. In contrast, it is also possible that disability origin will alter the perception of an employee's work history. If a person is believed to be the cause of his or her own disability, for example, the employer may view that person as irresponsible, negating any influence of prior work history. In contrast, an external cause of the disability suggests nothing about the integrity or character of the individual and, consequently, the influence of prior work history would not be muted.

METHOD

Design

The study had a 2 (disability origin: internal vs. external cause) × 2 (excellent work history vs. average work history) factorial design.

Participants

Participants (N=80) were recruited to maximize the possibility that they were in the position to make employment decisions. We contacted a southeastern university's Executive MBA programs and human resource consultants to help identify managers who might make accommodation decisions. For the EMBA student sample, we gained permission to make an announcement during EMBA classes to solicit participation in the study. Surveys were left in the classes and interested students completed the surveys and returned them at a later date. The consultants who agreed to contact the management of the organizations with which they worked primarily dealt with organizations in North Carolina, South Carolina, and Florida. The organizations involved included government agencies, manufacturing firms, financial institutions, and retail firms. Again, surveys were left with key personnel within these organizations and those interested completed and returned the questionnaires.

Recruitment of a management sample was successful. The majority of the participants (91%) currently worked full time and 56% made hiring decisions for their organizations, although only 6% had handled claims for accommodation. The sample included human resources managers and directors of sales, marketing, and accounting departments. The mean number of years as part of the full-time work force was 10 years, with an average of 6 years spent at their current organization. The participants were predominantly female (65%), with a mean age of 32 years.

Materials

Participants read a two-page single-spaced scenario that described an individual currently employed by a mid-size manufacturing firm in the Human Resources Department. To ensure that the greatest number of employers would understand and be familiar with the position depicted, it consisted mainly of clerical duties. The scenarios provided information about the manufacturing firm, the employee's job duties, the employee's performance evaluations, and the disability.

The job description for a mid-level human resources position was also included in the materials. The job description was one used by a mid-sized manufacturing company. The position in question was within the Human Resources department, dealing strictly with the administration of employee benefits in a mid-sized manufacturing company. The position required that the individual be knowledgeable in benefits plans, company policies and benefits administration. The physical demands of the job were also included in the job description.

To ensure that all participants, regardless of their previous experience with making ADA-relevant decisions, were knowledgeable about the requirements of the ADA, we provided participants with the details of the ADA that were relevant to accommodation decisions in the scenarios and the job description. The scenario reminded the respondents that they were not required to make accommodations that were considered to be an undue hardship to the organization in terms of financial responsibilities, organizational structure or the productivity of other employees. Respondents were also reminded of the fact that the employee must request any accommodation necessary to complete the job functions. The job description included a statement that reasonable accommodations may be made to assist a qualified individual in performing job duties. Participants read this information before making their accommodation decisions. In addition, Section 105 of Title I of the ADA requires that the Act be posted in an accessible area, thus respondents should have been exposed to this material in their own workplace.

Disability Origin Manipulation

We manipulated disability origin by stating that the disability was caused by an action of the employee or by the action of another. In the internal attribution conditions, the employee ran a red light when he was driving home from a social event, hitting a sport utility vehicle with his compact car. Tests revealed he had a blood alcohol level slightly over the legal limit. Although the driver of the SUV had only minor injuries, the employee suffered spinal cord damage and is now confined to a wheel chair. In the external locus conditions, the employee was driving through the intersection when the driver of the SUV ran the red light and hit the employee's car. It was later revealed that the driver of the SUV had a blood alcohol level over the legal limit. The injuries to the drivers were the same.

Work History Manipulation

Orthogonal to the manipulation of the disability origin, we manipulated the employee's work history within the company. In the average work history condition, the employee was described as meeting his deadlines competently and his performance evaluations had always been satisfactory. Although the employee was deemed competent, he had a tendency to refer questions to coworkers when he did not immediately know the answer. He also had occasional unexplained tardiness or conveniently scheduled sickness. Because he had consistently been rated as competently performing his job during his performance evaluations, the employee clearly met the ADA's definition of a qualified person with a disability (although we did not explicitly state this in the scenario). In the excellent work history condition, the employee was described as going above and beyond his job duties by helping other departments and by rearranging his own vacation schedule to help the organization meet deadlines. For the last 5 years, he received an “excellent” on his performance evaluations.

Dependent Measures

In the scenario, the employee requested nine workplace accommodations to enable him to perform essential job functions. We created a list of nine reasonable accommodations for an individual with paraplegia by taking accommodations from case studies provided by the Job Accommodation Network (JAN), as well as ADA related books (JAN, 2000; Jones, 1993). An approximate cost of implementation was provided along with each accommodation. We made an effort to provide an equal number of accommodations that could be considered high, moderate, or low in cost. See Table 1 for the cost associated with each of the accommodations. The survey required respondents to indicate which accommodation(s), if any, they would be willing to grant as a package. It also asked them to indicate any accommodations that they would be willing to provide that were not listed.

Table 1. Participants’ Accommodation Choices by Condition (N=80)
Full size table

A questionnaire assessed the effectiveness of the manipulations of attribution of responsibility and work history. Participants indicated their agreement with the following questions: “David Johnson is responsible for the onset of his disability,” “David Johnson's disability is “just one of those things” that sometimes happen to people, outside of their control,” and “David Johnson should have expected that something bad would happen as a result of his actions.” Participants indicated their agreement with each statement on a 7-point scale (with 1 being strongly disagree and 7 being strongly agree) and the questions were coded so that higher scores reflected an internal attribution. We created an attribution of responsibility scale, by averaging scores across these statements, with higher scores reflecting an internal origin of the disability (Cronbach's α = .81). Participants also indicated their agreement with the following questions: “David Johnson does not go out of his way to contribute to the success of Bothwell Industries’ day-to-day functioning,” “David Johnson is an asset to Bothwell Industries,” and “David Johnson is a good employee.” Participants indicated their agreement with each statement on a 7-point scale (with 1 being strongly disagree and 7 being strongly agree). We created a work history scale by averaging scores across these statements (Cronbach's α =.69), with higher scores reflecting a better work history.

Procedure

Participants read the employee's request for accommodations, the employee's job description, and the ADA guidelines for accommodation decisions. They then completed a packet of materials assessing whether they would grant any of the requested accommodations, demographic information, and the effectiveness of the manipulations. Participants completed the packet of materials individually and returned them by mail or in person.

RESULTS

Manipulation Checks

We conducted an analysis of variance (ANOVA) with disability origin and work history as independent variables and the attribution of responsibility scale as the dependent variable. Those participants who read the internal origin conditions rated the employee as being more responsible for his disability (M=4.81) than did those who responded to the external origin conditions (M=2.16), F(1, 76)=140.14, p < .01, η2=.65. Work history did not influence attributions of responsibility nor was there an interaction between disability origin and work history.

We conducted a second ANOVA with disability origin and work history conditions as independent variables and the work history scale as the dependent variable. Those participants who read the excellent work history conditions judged the employee to have a better work history (M=5.68) than did those who read the average work history conditions (M=4.43), F(1, 76)=33.23, p < .01, η2=.30. Participants rated the employee to be a satisfactory worker (i.e., well above the mid-point of the scale) in both conditions; thus, there would be no legitimate reason for denying the employee accommodation on the basis of their prior performance, as participants viewed the previous performance to be at least adequate. Disability origin did not influence perceptions of work history nor was there any interaction between work history and disability origin.

Accommodation Number and Cost

Because the number of accommodations granted and the total cost of accommodations granted were highly correlated, r(80)=.67, p < .01, we conducted multivariate ANOVAs (MANOVAs) for subsequent analyses. There was a significant main effect of disability origin in the multivariate analysis, F(1, 76)=4.82, p=.01, η2=.11, driven by a univariate main effect of attribution of responsibility on the number of accommodations granted, F(1, 76)=9.61, p < .01, η2=.11. Participants granted more accommodations to the employee depicted as having an external cause for his disability (M=5.85) than to the employee who was depicted as having an internal cause for his disability (M=4.80). There were no main or interactive effects of work history on these measures.

To rule out the possibility that participant decisions differed as a function of whether they had made hiring decisions (indicating that they had made decisions governed by the ADA in the past), we conducted additional analyses with hiring experience as a between-participants variable. These analyses revealed no significant main effects or interactions based on whether participants made hiring decisions, all Fs < 2.16, p > .12.

Relevance of Job Accommodations

Perhaps participants granted all relevant accommodations to the employee and awarded extra, unneeded accommodations to the employee with an externally caused disability because of sympathy for the employee. To test this hypothesis, two subject-matter experts (SMEs) who were responsible for implementing the ADA in their workplace read the job description in the study and rated the relevance of the nine accommodations for performing the job duties on a 7-point Likert scale, with higher numbers representing higher relevance. All but one of the accommodations was rated at or above the scale mid-point, indicating that these SMEs deemed the accommodations relevant, with a mean rating of 5.67. Thus, accommodation relevancy cannot explain the results obtained.

DISCUSSION

The hypotheses for this study were partially confirmed with the finding that disability origins influenced the granting of reasonable accommodations. Participants granted more accommodations to the individual who had an external explanation for his disability than to an individual who caused his own disability. The effect of disability origin on accommodations cost, despite mean differences in the predicted direction, was not significant. In contrast, work history did not influence accommodation decisions, perhaps because of the smaller effect size for the work history manipulation (η2=.30) in comparison to that of the disability origin manipulation (η2=.65).

Study Two

This second study conceptually replicated the first study using a different medical condition. Study One involved a manipulation in which the individual was either breaking the law by driving under the influence, providing an internal attribution, or not breaking the law, providing an external attribution. It is possible that lawfulness, rather than the manipulation of causality, is responsible for the findings. This possibility was addressed by manipulating causality in a different context, removing issues of lawfulness. Our hypotheses for Study Two remained the same as those for Study One.

Method

Participants

Eighty new participants were recruited using the same method as in Study One. The majority of the participants (75%) worked full time. Many (41%) made hiring decisions for their organizations, and 6% handled ADA claims. Participants included directors in human resources departments and upper level management of sales and finance departments. The mean number of years working full-time work was 13 years, with an average of 5 years spent at their current organization. The majority of participants were female (59%), with a mean age of 34 years.

Design, Materials, and Procedure

The experimental design, materials, and procedure used in Study Two were identical to those used in Study One, with two exceptions. In Study Two, the scenarios described an individual with low vision rather than paraplegia. With this disability, work performance would be severely disrupted unless accommodation is provided, as a significant proportion of the work time is spent working on a computer. An additional thirty percent is spent in phone communication, which could also be disrupted by low vision (e.g., dialing numbers, prioritizing calls based on number provided by LCD display).

Because of the change in disability, the disability origin manipulation was also changed. In both conditions, the employee had started experiencing light-headedness, dizziness, and fatigue 18 months previously and had visited the doctor at that time. In the internal attribution of responsibility condition, the employee was diagnosed with diabetes but repeatedly ignored his doctor's orders to change his diet, to exercise regularly, and to check his blood sugar levels daily. After 18 months of ignoring these medical instructions, the employee suffered a partial loss of sight that leaves him almost legally blind. In the external responsibility condition, the treating doctor failed to diagnose the employee's diabetes when the employee initially complained of his symptoms. Because of this failure, the employee's diabetes remained untreated, resulting in the employee's vision loss.

Dependent Measures

We created the reasonable accommodation questionnaire using the same sources of information as in Study One (JAN, 2000; Jones, 1993). See Table 2 for the costs associated with each accommodation. Included in the questionnaire were manipulation checks for attribution of responsibility (Cronbach's α=.88) and work history (Cronbach's α=.94), consisting of the same questions used in Study One.

Table 2. Participants’ Accommodation Choices by Condition (N=80)
Full size table

RESULTS

Manipulation Checks

We conducted an ANOVA with disability origin and work history as independent variables and the attribution of responsibility scale as the dependent variable. Those participants who responded to the internal origin conditions rated the employee as being more responsible for his disability (M = 5.05) than did those who responded to the external origin conditions (M=2.64), F(1, 76)=117.21, p < .01, η2=.61. Work history did not influence perceptions of attribution of responsibility, nor was there any interaction between disability origin and work history.

We conducted a second ANOVA with the disability origin and work history conditions as independent variables and the work history scale as the dependent variable. Those participants who responded to the excellent work history conditions judged the employee to have a better work history (M=5.91) than did those who responded to the average work history conditions (M=3.93), F(1, 76)=104.63, p < .01, η2=.58. Participants’ ratings of the employee with an average work history again indicated that they believed his performance to be adequate. Disability origin did not influence perceptions of work history, either independently or interactively.

Accommodation Number and Cost

Because the number of accommodations granted and the total cost of accommodations granted were highly correlated, r(80)=.59, p < .01, we conducted MANOVAs for these analyses. There was a trend for the main effect of disability origin in the multivariate analysis, F(3, 79)=2.48, p < .10, η2=.06. This effect was driven by a significant univariate main effect of attribution of responsibility on number of accommodations granted, F(1, 76)=4.68, p < .05, η2=.06. Participants recommended more accommodations for the employee with an external cause for his disability (M=5.28) than to the employee with an internal cause for his disability (M=4.65). There was a significant main effect for work history in the multivariate analysis, F(3, 79)=5.57, p < .01, η2=.14. This effect was driven by a significant univariate main effect for work history on both the number of accommodations granted, F(1, 76)=8.88, p < .01, η2=.11, and the cost of the granted accommodations, F(1, 76)=8.32, p < .01, η2=.11. Participants granted more accommodations to the employee with an excellent work history (M=5.40) than to the employee with an average work history (M=4.53). Participants also granted more expensive accommodation packages to the employee who had an excellent work history (M=$20,418) than to the employee who had an average work history (M=$7,769).

To determine whether participant decisions differed as a function of whether they had made hiring decisions, we conducted additional analyses with hiring experience as a between-subjects variable. These analyses revealed no significant main effects or interactions as a function of decision-making status, all Fs < 1.29, p > .28.

Accommodation relevancy cannot explain the results obtained. We collected SME's ratings of the relevance of the requested accommodations on 7-point Likert scales, with higher numbers representing higher relevance. All but one of the accommodations was rated at or above the scale mid-point, indicating that the accommodations were relevant, with a mean relevancy rating of 5.5.

DISCUSSION

The results of this second study supported the majority of our hypotheses. Although the effect of disability origin in the multivariate analysis was only marginally significant, it influenced the granting of accommodations in the univariate analysis. This effect is consistent with the results of Study One, in which the univariate analyses showed that disability origin influenced the number of accommodations granted but did not influence the cost of the granted accommodations. One possible reason that the effect is only marginally significant at the multivariate level is the fact that some disabilities appear to be given to more internal attributions, regardless of the actual cause of the disability. Research suggests that some disabilities, such as drug dependencies, are perceived as more internally caused, even under external situations (Bordieri & Drehmer, 1986). It is possible that the smaller effect for disability origin in Study Two is caused by a perception that adult-onset diabetes is more likely to be internally caused than paraplegia. Support for this argument can be found in the mean responses on the attribution manipulation check measure, which are higher in Study Two than in Study One, indicating that participants were more likely to see the diabetes-related low vision as internally caused. Additionally, although the effect sizes for the disability origin manipulations were approximately equal (η2=.65 and η2=.61, respectively), the power for the manipulation across the two studies was quite different, with the observed power being .78 for Study One but only .48 for Study Two. Thus, the marginal significance of the effect in Study Two could be the result of lower power.

Work history influenced both the number and the cost of granted accommodations. Work history was influential in Study Two and not Study One even though the work history manipulations were identical in both studies. However, the effect size of the work history manipulation in Study Two was η2=.58, producing an observed power of .88, which is almost twice the effect size of the work history manipulation in Study One. It is most likely this difference in power that explains the differences between the results of the two studies.

GENERAL DISCUSSION

Disability origin consistently influenced the granting of accommodations across both studies. Participants granted more accommodations for the employee who provided an external explanation for his disability. These findings are consistent with the existing social psychological literature, which shows that, in a variety of contexts, external attributions of responsibility for an event (such as a disability or a need for class notes) lead to a greater willingness to provide aid (e.g., Meyer & Mulherin, 1980). These results provide evidence that disability origin influences an additional phase of the employment process.

Work history influenced both the number of accommodations granted and the cost of those accommodations, although inconsistently. It is not surprising that work history had an independent influence on these decisions, considering the previous research. Existing research shows that prior work history influences the willingness to hire an individual with a disability, regardless of the type of the disability (Bordieri & Drehmer, 1986; Stone & Sawatzki, 1980). What is interesting, however, is the fact that work history did not have a consistent influence across studies.

The most likely reason for these discrepant findings is the fact that the work history manipulation was much more effective in Study Two. It is possible that this discrepancy is due in part to the fact that the symptoms of the disability in Study Two were more likely to influence job performance directly whereas the disability in Study One was more likely to influence access to the workplace. Previous research suggests that employers may be differentially influenced by information under different contexts (Smith, Edwards, Heinemann, & Geist, 1985). Considering that other factors, such as disability origin, have been consistently found to influence employment decisions, it may be possible that employers’ decisions are guided by the most salient piece of information. In Study Two, the disability symptoms were directly relevant to the ability to perform the job duties. For example, the ability to perform computer work is drastically reduced by low vision. In Study One, however, the disability symptoms were not directly relevant to the ability to perform the job duties and were more closely related to workplace access. It is possible that the low vision manipulation made job performance more salient than the paraplegia manipulation, due to the greater relevance of the low vision symptoms to the actual job duties performed, producing the greater effectiveness of the work history manipulation in Study Two.

Practical Implications

Employees with disabilities are profoundly affected by accommodations that they receive. The ability to work has been touted as one of the most important aspects of American life (Blanck, 1996). Already facing potential discrimination due to their disability, individuals with disabilities expect the ADA to help them combat discrimination from potential or current employers. Instead, particularly if there is any way to attribute their disability onset to an action they have chosen to take, they may find themselves further disadvantaged when they are unable to receive an accommodation from their employer.

The use of extralegal factors such as disability origin and work history in accommodation decision-making has serious implications for the efficacy of Title I of the ADA, as well as the ADA as a whole. Congress intended for the ADA to “balance the scales,” allowing people with disabilities equal access to services and equal opportunity to employment. Particularly with employment, there should be some consistency in the application of the ADA across similar jobs and disabilities. For example, two individuals who hold the same position and have the same disability should be accommodated in a similar manner, to the extent that the two individuals work at companies with similar financial standing. Instead, according to the results of this study, employers may rely on impermissible factors, such as disability origin and work history, to guide their decisions.

Although it would appear easy to fix the disability origin problem by not informing the employer of the origin of the disability, there is evidence that, when employers do not have causal information about the disability, they will act as if the disability has an internal cause (Bordieri & Drehmer, 1986, 1988). If this is the case, then it may prove especially difficult to alleviate the influence of attribution of responsibility on accommodation decisions. It should, however, also be possible to alleviate the influence of work history in the same manner, by assigning the duty of making accommodations decisions to one department that lacks access to employee evaluation judgments. Smaller companies, however, may not have the personnel to assign one person, blind to the work performance of other employees, the task of granting reasonable accommodations (although an organization must have at least 15 employees before it is liable under the ADA). These smaller companies, though, could join together in coalitions, or use the services of an outside consultant to accomplish the same goal. These potential remedies are purely speculative, however, and should be the subject of future research.

It is also important to consider which, if any, variables influence judges and juries faced with these types of decisions. Although the majority of accommodation claims may be resolved between the employer, employee, and (potentially) the EEOC, the courts have been dealing with questions raised by the ADA, including the type and number of accommodations appropriate in a given case. Businesspeople, who are presumably trained to apply the ADA, are influenced by extralegal factors. It seems likely, therefore, that juries also may be influenced by these factors. This empirical question has not yet been addressed.

Some methodological issues should be raised when considering these results. Although participants in this study were predominately in a position to make decisions governed by the ADA, very few of them had ever handled a request for accommodation for their organizations. It is possible that individuals who routinely make these decisions would be differentially influenced by the manipulations. The number of individuals who had handled ADA claims in these studies was too small to make such comparisons with any degree of reliability. Exploratory analyses, however, suggest that participants who handled ADA claims reacted in much the same way as participants who had not handled these claims. They were more likely to give a greater number of accommodations to those with an external attribution (M=6.00) than an internal attribution (M=4.00).

The limitation of the sample seems minor when one realizes that the ADA governs hiring decisions as well as accommodation decisions for current employees. The majority of participants made hiring decisions for their organizations, making it likely that they were familiar with the requirements of the ADA. A subsequent analysis comparing the decisions of those participants who made hiring decisions to those who did not revealed no main effects or interactions for decision-making status in either study. Thus, our manipulations did not influence participants who make hiring decisions differently than those who do not make hiring decisions. It is also important to note that information regarding the ADA was included in the information participants received. An analysis of the accommodations that participants suggested (in addition to the ones provided) were modifications of the accommodations suggested by the employee, an action that is implied by the ADA, providing further support for the idea that many of the participants were familiar with the ADA. Additionally, it is important to remember that many small organizations deal with accommodation claims on a very rare basis and may not have any one member of the organization dedicated to handling ADA claims. There is no reason to believe that the decisions made by a person in a small organization who finds him or herself making accommodation decisions, particularly for the first time, differs from the decisions made by those who are responsible for making hiring decisions.

Also, it should be noted that participants were reacting to a written scenario, describing a hypothetical employee. This reliance on written materials is a shortcoming of the study, in that actual ADA claims are very “interactive” between the employee and the employer (Jones, 1993). There is also a concern that the use of written materials may cause respondents to take the task less seriously than they would if the materials were more realistic. This concern seems unnecessary given that many of the respondents took notes on the provided materials, with approximately 20% providing additional accommodations that they would provide. These accommodations provide further proof that the participants were carefully considering the materials, as many of them were modifications of the original accommodations listed. The additional accommodations predominantly addressed access to the facility (entrance doors and restroom modifications), workspace (rearrangements of the desk area), and additional help (hiring an assistant, assigning employee to help navigate through facility).

Even with the methodological limitations discussed, the findings show that attribution of responsibility and work history influence accommodation decisions. Although these findings need to be replicated, they suggest that employees with disabilities may not be treated within the spirit of the ADA. The ADA was enacted to help individuals with disabilities overcome the obstacles to fully participating in American life (e.g., work, access to public transportation). Instead, these results suggest that individuals with disabilities are being judged by other, legally impermissible factors. Willingness to fulfill the spirit of the ADA seems to be in part based on whether the individual with the disability is “good” (external cause for disability or excellent work history) or “not good” (internal cause for disability or average work history). This runs counter to the goals of the ADA, which include equitable treatment of individuals with disabilities, not the further classification of people into those who are worthy or not worthy of accommodation.

In addition to reducing the efficacy of the ADA, the use of impermissible factors in accommodation decisions may prevent the successful integration of individuals with disabilities in the work force. With the importance of work in American society and for the socioeconomic well-being of people with disabilities, it is vital that this portion of the population be fully integrated into the work force. This integration can only be accomplished by eliminating the influence of impermissible factors on accommodation decisions.

Footnote 1