Is seeking social change through law a hollow hope? This question has become a kind of crossroads, a potential meeting place for sociolegal scholars working in many traditions. Scholars of legal mobilization see law as a potential political resource and examine how activists use rights-based litigation as tactical leverage to force organizational change (Reference EppEpp 1998; Reference FrymerFrymer 2003; Reference Goldberg-HillerGoldberg-Hiller 2002; Reference McCannMcCann 1994; Reference PeltasonPeltason 1961; Reference ReedReed 2001; Reference RosenbergRosenberg 1991; Reference Feeley and RubinRubin & Feeley 1998; Reference SilversteinSilverstein 1996; Reference VoseVose 1957, Reference Vose1958), or how they use rights claims outside of formal legal settings (Reference Engel and MungerEngel & Munger 2003; Reference ScheingoldScheingold 1974). Neo-institutionalists explore how organizations internalize ambiguous legal requirements and translate them into templates for organizational responses, which are then legitimated and diffused (Reference AlbistonAlbiston 2005, Reference Albiston2010; Reference DobbinDobbin 2009; Reference Dobbin and SuttonDobbin & Sutton 1998; Reference Dobbin, Sutton, Meyer and ScottDobbin et al. 1998; Reference EdelmanEdelman 1990, Reference Edelman1992; Reference Edelman and Suchman1997a,Reference Edelman and Suchman1997b; Edelman et al. Reference Edelman, Uggen and ErlangerEdelman 1999; Reference EppEpp 2009; Reference GouldGould 2005; Reference Jenness and GrattetJenness & Grattet 2001; Reference Kalev and DobbinKalev & Dobbin 2006; Reference Kalev, Kelly and DobbinKalev et al. 2006; Reference Scheid, Suchman, Schutt and HartwellScheid & Suchman 2001; Reference Sutton, Dobbin, Meyer and ScottSutton et al. 1994). Scholars of regulation offer a parallel set of analyses of agencies, examining how enforcement practices affect organizational responses to law and exploring how attitudes toward the law within organizations affect organizational behavior (see generally Reference Bardach and KaganBardach & Kagan 1982; Reference Braithwaite and AyresBraithwaite & Ayres 1992; Reference CoglianeseCoglianese 2001; Reference Gunningham, Kagan and ThorntonGunningham et al. 2003; Reference Howard-Grenville, Nash and CoglianeseHoward-Grenville, Nash, & Coglianese 2008; Reference MayMay 2004, Reference May2005; Reference May and WinterMay & Winter 1999; Reference May and WoodMay & Wood 2003; Reference MuirMuir 1973; Reference Reiss, Hawkins and ThomasReiss 1984).
We are enthusiastic readers of each of these strands of research, though we are often struck by their lack of engagement with one another. Perhaps that should not be surprising: scholars in these diverse traditions embrace fundamentally different conceptions of law and so have very different ideas about how to study it. For example, scholars who work on regulation, especially those in the law-and-economics and rational-choice traditions, tend to conceive law as prescriptive rules that legislatures, agencies, and courts send out to various targets. From this perspective, the central questions revolve around “compliance”: Under what conditions do the targets of the law follow the law's requirements? When do they fall short? And why would rational organizations engage in costly “beyond-compliance” behavior? Many scholars working in the law and society tradition start from radically different premises, however. They embrace a much more fluid view of law, in which law is constantly constructed and reconstructed through human interaction. From this perspective, law is “all over” (Reference SaratSarat 1990), made not just by formal institutions—legislatures, courts, and agencies—but by every person and organization that encounters it.
We value this fluid, decentered view of law. Law is rarely unambiguous in application; it must be interpreted—and the process of interpretation will inevitably be highly variable across settings and time. But the decentered perspective on law deeply complicates the task of the researcher. From this perspective, cause and effect are difficult to specify, thus rendering a simple, parsimonious account of the effects of law impossible (Reference Burke and BarnesBurke & Barnes 2008; Reference McCannMcCann 1996). Perhaps for that reason, much research in this tradition focuses on consciousness and discourse, downplaying or even eschewing any attempt to assess improvements in concrete social conditions (Did more minorities get jobs? Was the water cleaner? Was the company safer?) that originally motivated the proponents of the law in question.
We worry about this tendency, because we believe the study of outcomes, however messy and labor-intensive, lies at the heart of understanding how and why law matters. Thus we would like to see research in this field combine some of the strengths of the varied literatures that contribute to it, retaining the fluid, decentered approach to law championed by law and society scholars and paying attention to the discourses that organizations and individuals use in interpreting law, while also observing the practices they engage in and the outcomes these practices create.
Neo-institutionalist scholarship on law has, to its credit, attempted to walk this line. It takes seriously the construction of law, and it has offered the most elaborate account of how legal rights are translated into organizational practices, as well as how they are diffused in professional and organizational fields. Yet neo-institutionalism has also struggled with some fundamental issues. The most prominent and influential studies of law in the neo-institutionalist literature are analyses of how large organizations respond to civil rights employment law. This research typically relies on surveys or self-reports to measure what goes on inside the organization. This reliance, together with the ambiguity of outcomes in this area of law—what exactly counts as successful implementation of civil rights law?—intensifies the generic problem all scholars face in this area: the measurement of outcomes. From the beginning of neo-institutionalism, scholars have questioned whether the typical responses of large organizations to legal commands—creating offices and policies—are merely symbolic window dressing, “rational myths” that legitimate the organization while protecting it from significant change (Reference Meyer and RowanMeyer & Rowan 1977). Neo-institutionalists continue to wrestle with this question, sometimes quite cleverly (Reference Kalev and DobbinKalev & Dobbin 2006; Reference Kalev, Kelly and DobbinKalev, Kelly, & Dobbin 2006), but largely in isolation from the other research traditions in this field of inquiry.
In this article, we seek to demonstrate how insights from the diverse literatures that examine law and social change can be brought together to better understand organizational response to law. We start from the law and society premise that law is constructed and reconstructed as it is diffused, but we believe that these highly contingent processes can give rise to stable practices that can in turn be correlated with specific outcomes related to the underlying goals of the laws. Our empirical focus, disability access rights, allows us to examine simultaneously how organizational leaders understand law alongside the tangible outcomes organizations produce. In probing those understandings and the practices that organizations engage in, we provide a set of variables that can be incorporated into future studies that we hope will cut across various divides in the literature on law and social change.Footnote 1
In this article we trace the responses of ten organizations to the right of wheelchair users to access to public facilities, which can be found in various provisions of the Americans with Disabilities Act (ADA) and related state laws. This right applies to public spaces, both governmental and nongovernmental, and requires organizations in some cases to rearrange their spaces so that wheelchair users can operate within them. Our analysis uses original qualitative and quantitative data that encompasses dozens of interviews and inspections of 161 facilities, which allow us to compare the self-reported responses of organizations that faced different levels of legal mobilization under the relevant access laws with the actual levels of accessibility at their facilities. Thus our data allow us to juxtapose what organizations say they do with the results they produce, which is critical to any account of how and under what conditions law matters.
We find that key personnel in organizations across our sample articulated similar understandings of wheelchair access rights, indicating that they believed they were under an obligation to make “reasonable” accommodations. Inconsistent with a rational-choice perspective on law, they demonstrated little understanding of the details of the access statutes, even potential defenses and safe harbor provisions that might have allowed them to avoid the law's mandates and thus might have reduced the costs of “compliance.” Given their common understanding of the law—and the low cost of many strategies for improving access that applied to all of our cases—one might expect the organizations’ practices and outcomes to converge. Yet the opposite was true: the self-reported organizational responses and results we directly observed varied from those that did next to nothing and offered virtually no access to wheelchair users, to others that had spent thousands of dollars on efforts that clearly went well beyond compliance and resulted in high levels of access.
As discussed more fully below, we explore this puzzle by focusing on the interaction between legal mobilization—which occurs when the organization has faced some type of enforcement action based on a legal threat—and three organizational response variables, drawn from the sociolegal literature, that have typically been studied in isolation.Footnote 2 These variables are commitment, the degree to which organizational personnel who are primarily responsible for interpreting and implementing the relevant law embrace its underlying social goals; professionalization, the degree to which the organization has written procedures and policies related to the law, creates a formal office with responsibility for access law, and interacts with outside groups and experts to learn about the law and best practices; and routinization, the degree to which the organization's consideration of the law's underlying goals and purposes permeates the daily practice of the organization, so that planning and management incorporate consideration of those goals. By measuring each of these variables separately and examining their interaction, our data reveal intriguing patterns, although our sample is admittedly small.
In the following pages, we first define our key variables in light of the relevant literature and describe how we operationalize them. We then set forth our research strategy, describe the range of organizational responses to law we observed, and detail how those responses relate to the accessibility of the organizations’ facilities. We end by discussing our findings in light of the existing literature and suggesting further avenues of inquiry. We argue that our findings illustrate the promise of combining insights from political science, sociology, and regulatory studies, and of simultaneously studying how organizations interpret law and the tangible outcomes they produce.
Legal Mobilization, Commitment, Professionalization, and Routinization
Why do organizations governed by the same laws act so differently? Perhaps the simplest answer is that formal rules are not self-executing; they must be mobilized. Accordingly, we would expect that organizations facing greater levels of legal mobilization would respond more vigorously to the law's demands.
As is often the case in this area, the hypothesis seems straightforward, but operationalizing the central concept involves tricky choices. Mobilization can take many forms, including the filing of lawsuits or complaints with regulatory agencies, public protests, or actions that are harder to detect, such as the sending of informal demand letters. A larger concern is whether to measure mobilization objectively or subjectively, as the perceived threat of legal action. Some scholars have focused on perceived threat due to their reasoning that the danger of future enforcement is the key factor in organizational response. But whose perception is relevant? The personnel charged with responding to the law? The head of the organization? Or should the researcher attempt to measure an organizational perception of threat?
In this project, we measured mobilization by scanning court filings and federal agency complaints and by conducting structured interviews with key personnel within the organization—people who are in charge of responding to facility access issues and who gave us a narrative of their organization's experiences with access law. In analyzing these data, we sought to ascertain the extent to which the organization has faced legal action. (See Table 1 for a summary.)
Table 1. Legal Mobilization Index
Sociolegal scholars of various stripes have long recognized that legal mobilization is only part of the story of why and how law matters. Organizations facing similar levels of legal mobilization can respond in starkly different ways. Some may adopt a bunker mentality and resist the law (Reference Bardach and KaganBardach & Kagan 1982), while others may embrace it and change core activities. Faced with this variation, regulatory scholars have generally focused on issues of compliance, starting with the assumption that firms will follow rules to the extent that the benefits of compliance outweigh the costs. From this vantage point, a central puzzle is why organizations, particularly profit-making entities, would engage in costly “beyond-compliance” behaviors. To answer this question, scholars tend to stress the role of organizational commitment to the law in shaping its impact (Reference Gunningham, Kagan and ThorntonGunningham, Kagan, & Thornton 2003; Reference Howard-Grenville, Nash and CoglianeseHoward-Grenville et al. 2008). Where there is commitment to the goals of some regulation (environmental, safety, civil rights), these scholars demonstrate, the organization may act in ways that go beyond what even the most rigorous plausible reading of the law requires. Where there is no such commitment, the organization may do the least possible to evade punishment, or may ignore the law entirely.
As with mobilization, this hypothesis has strong intuitive appeal, but operationalizing and measuring the key concept—commitment—involves some vexing choices. A central issue is whose commitment is relevant. Is it the CEO's? Is it that of the managers who actually supervise operations connected to the regulation? Or is commitment an organizational property—part of its culture—that transcends the attitudes of any single individual within the organization? And finally, if commitment is an organizational property, how can this property be reliably and efficiently measured? There are no simple answers to these questions.Footnote 3 In our study, we conceptualize commitment as the degree to which key personnel—those within the organization who are primarily responsible for interpreting and implementing the relevant law—embrace the law's underlying social goals. Again, we employed structured interviews to probe the commitment of the key actors within the organization. Our interviews combined open-ended questions, such as “Describe a typical access problem,” and more specific questions, such as “Do you think the disability access laws are fair?” We reserved the highest commitment score for personnel who embraced the social model of disability, the view that access is a right, and the view that failure to provide access is a form of discrimination against people with disabilities. As with legal mobilization, we used an index to code the various interview transcripts. (See Table 2.)
Table 2. Commitment Index
Compared to regulatory scholars, neo-institutionalists are less concerned with the attitudes of organizational members and more focused on the structures that organizations build in response to laws. Neo-institutionalism's origins are in organizational sociology, especially studies of the convergence of structures and practices within and across organizational fields (Reference DiMaggio and PowellDiMaggio & Powell 1983). A major impetus for the development and diffusion of structures is threats in the environment, such as economic competition. Neo-institutionalists examine how organizational structures develop in response to new laws, how these structures diffuse, and how they become legitimated. Faced with vague legal mandates whose commands defy simple notions of “compliance with law,” organizations typically respond through what we have labeled professionalization: the creation of offices and positions specifically charged with responding to regulations, internalizing legal expertise, and reaching out to other experts and professionals to stay abreast of the “best practices” (Reference DobbinDobbin 2009; Reference Dobbin and SuttonDobbin & Sutton 1998; Reference Dobbin, Sutton, Meyer and ScottDobbin et al. 1998; Reference EdelmanEdelman 1990, Reference Edelman1992; Reference Edelman, Uggen and ErlangerEdelman, Uggen, & Erlanger 1999). These practices, over time, become templates for other organizations and thus spread throughout an organizational field.
Building on the insights of neo-institutionalists, we measured professionalization by determining whether an organization has created a formal office with responsibility for the law, whether the organization has written procedures and policies related to the law, and the extent to which organizational staff interact with outside groups to learn about the law. (See Table 3.)
Table 3. Professionalization Index
A major concern in the neo-institutional literature is whether professionalization in its various forms is merely symbolic window dressing, a “rational myth” that legitimates the organization while protecting it from significant change (Reference Meyer and RowanMeyer & Rowan 1977). To assess this concern, we look at the degree to which concerns about rights are routinized—that is, integrated into daily planning routines and core activities, such as trainings and policies that seep into the everyday life of the organization. Organizational theorists call this process “coupling” (Reference WeickWeick 1976, Reference Weick, Goodman and Sproull1990) and have found that coupling significantly varies across organizations. Even ostensibly committed and professionalized organizations can produce rights practices that are not routinized and thus lost in the shuffle of competing policies, goals, and factions that characterizes complex organizations. So, for example, in their work on mining safety, Reference Gunningham and SinclairGunningham and Sinclair (2009) show that apparently sincere commitments to the law's goals at the top levels of management were disconnected from the firm's everyday mining practices in loosely structured, complex organizations, even when organizational leaders employed state-of-the-art management strategies for implementing their preferences.Footnote 4 Alternatively, organizational leaders may consciously “decouple,” which involves maintaining the symbols of professionalization (an office, a policy) but ensuring that rights practices do not permeate the core operations of the organization. The strategy of decoupling offers the organization the promise of both legitimacy and efficiency; it can be seen as responding to the law yet leaves the organization's central goals undisturbed (Reference Boxenbaum, Jonsson and GreenwoodBoxenbaum & Jonsson 2008).
In our study, we conceptualize routinization as the degree to which concerns related to the law permeate the daily practices of the organization. In coding our interviews, we examined the extent to which people with expertise and responsibility for access issues within the organization are involved in the core facility planning processes. In an organization in which access rights have become truly routinized, access concerns would be handled as a matter of course, built in to the facilities planning process rather than being considered episodically. We also determined whether the organization had a separate budget for access improvements. While in theory a routinized organization should seamlessly integrate access costs into overall budgets, in practice we found that a separate budget was a strong indicator of the incorporation of access concerns into the facility's processes. Finally, because an accessible facility can be rendered inaccessible by neglect—for example, supplies stacked where they block a path to the bathroom, electronic doors failing because of lack of maintenance or battery power—we also measured whether the organization trained its staff in access issues, so that access concerns filtered below the level of the key personnel. (See Table 4.)
Table 4. Routinization Index
Research Strategy
The Law
We focus on disability access law for several reasons. First, we are interested in the relationships among organizational understandings of law, organizational practices, and tangible outcomes. The measurement of outcomes has always been a troublesome issue in the mobilization, neo-institutionalist, and regulatory literatures, which often rely on organizational self-reporting. The outcomes in access law, though, are objectively measurable and easily observed. The access provisions we study apply only in areas open to the public, so we can independently inspect organizations’ facilities to see if their self-reported attitudes and practices produce outcomes that advance the goals of access law.
Second, like the neo-institutionalists, we are interested in studying organizational responses to broad social change legislation, and disability access law is remarkably ambitious. The most important access law is the ADA. The best-known provision of the ADA is Title I, which governs discrimination in employment. If ours were a study of ADA implementation, or of the effectiveness of the disability rights movement generally, we would likely focus on this segment of the ADA, which has received the bulk of attention from scholarsFootnote 5 (Reference Acemoglu and AngristAcemoglu & Angrist 2001; Reference BlanckBlanck 2000; Reference Burke and PietroBurke 1997; Reference DeLeireDeLeire 2000; Reference JollsJolls 2000; Reference Stapleton and BurkhauserStapleton & Burkhauser 2003). But for the purposes of this study Title I is less useful: the effects of civil rights employment law have been extensively researched, and as neo-institutionalist scholarship has demonstrated, measuring outcomes in this area is particularly difficult. Thus we focus on some of the accessibility provisions in the ADA and on related state access laws. Title II of the ADA covers access to governmental programs, requiring states and localities to make their programs and services equally available to disabled and nondisabled people. Title III regulates accessibility in places of public accommodation operated by nongovernmental entities. These titles generally require “readily achievable” removals of physical barriers (42 U.S.C., Sec. 12182(b)(2)(A)(iv)) and “reasonable modifications” to programs and services that would otherwise screen out people with disabilities (42 U.S.C., Sec 12182(b)(2)(A)(ii)). In addition, both titles create accessibility guidelines for the construction or remodeling of facilities, and these guidelines coexist with state laws.
For organizations, interpreting the requirements of these laws is a significant task. Both federal titles and corresponding state laws have an array of defenses and exceptions that potentially limit the reach of these requirements, but these are not safe harbor exceptions that would allow large categories of organizations to comfortably ignore the law. Instead, the defenses and exceptions tend to be open-ended, ambiguous, and subject to interpretation. So, for example, Title III states that program directors do not have to make “reasonable modifications” to their policies, practices, or procedures if that would “fundamentally alter the nature” of the goods or services they provide (42 U.S.C. sec. 12182 (b)(2)(A)(iii)).Footnote 6 This makes the disability “mandate” in these laws particularly open to interpretation and dispute. Nonetheless, access laws clearly require organizations to reconsider a diverse set of organizational practices. While some of those, such as adding a ramp or handrails to a path, are marginal to the organization's core functions, others are fundamental—for example, a store may have to transform the layout of its merchandise, or a program may have to change its eligibility requirements.
Yet disability access laws have an even more ambitious goal. From the perspective of the disability rights movement, which advocated access legislation, access law is a tool not simply for reconfiguring programs and facilities but for changing consciousness. The disability rights movement is premised on the “social model” of disability, in which disability is literally created by social attitudes and arrangements. “The general public does not associate the word ‘discrimination’ with the segregation and exclusion of disabled people,” writes Robert Funk, the first director of the Disability Rights Education and Defense Fund. “Historically the inferior economic and social status of disabled people has been viewed as the inevitable consequence of the physical and mental differences imposed by disability” (Reference Funk, Gartner and JoeFunk 1986: 7). But from the perspective of the social model, the arrangements of society—a bathroom too narrow for a wheelchair, a subway without an elevator, an elevator without Braille buttons—can be considered a form of discrimination because they fail to take into account the full range of human diversity. The view that disability is created by social arrangements rather than by an individual's impairment, and that organizations have a moral responsibility to make their programs and facilities accessible to all, requires a radical gestalt switch from the conventional understanding of disability. Buildings like the U.S. Supreme Court, with its entrance of inaccessible marble steps, would be seen in the same light as a “colored only” water fountain. Of course, it is likely that few, if any, members of Congress who voted for the ADA subscribed to this vision; the predominant view seems to have been that the ADA was simply a good thing to do for people with disabilities, particularly if it led some of them to employment (Reference Burke and PietroBurke 1997). From the perspective of many in the disability rights movement, however, a shift in consciousness is the ultimate goal of disability rights laws. Thus it is particularly important to measure both consciousness and concrete outcomes, changes both in discourse and in practice, when attempting to understand organizational response to access law.
Judged by the admittedly ambitious standard of consciousness changing, access laws have been largely unsuccessful, another aspect of this case that makes it particularly useful to study. While some organizations have responded vigorously to access statutes, others openly ignore them, yet a social stigma against inaccessible facilities has not yet developed. An organization that is seen to pump polluted water into a river, or to discriminate against women in employment, risks losing its “social license”—its legitimacy in the community (Reference Gunningham, Kagan and ThorntonGunningham, Kagan, & Thornton 2003). In our background research for this project, we did not find examples in which organizations that failed to comply with access law were similarly stigmatized; in fact, it was the enforcers of the law—lawyers and plaintiffs who brought access complaints—who were often criticized in media accounts of controversies over access. Unlike in the cases that Reference EppEpp (2009) studies, we do not see the development of a “legalized accountability” model, a shared consensus among a professional group about how organizations should handle access issues. In this respect, disability rights are a “vanguard” case in which a social movement is struggling to establish a new norm, much as in Reference SilversteinSilverstein's (1996) study of the animal rights movement or McCann's study of the comparable worth movement (Reference McCann1994). But unlike the animal rights and comparable worth movements, the disability rights movement has successfully enshrined many of its goals in legislation. Thus, legal change has in some respects preceded a change in consciousness. Disability access law thus provides both a useful contrast and a complement to many other areas of regulation that have received scholarly attention.
The Organizations
Given our central goal of exploring the effects of organizational responses to the law, it would be ideal to study organizational responses in as many contexts as possible. However, we could not cover the broad sweep of organizations that fall under the ADA's access provisions in a single study. We could, however, select diverse organizations that operate within divergent settings yet under similar laws and legal understandings. Accordingly, we selected sets of organizations within the private sector (restaurants), the public sector (city governments), and the nonprofit sector (colleges and universities), all located in a single state. The organizations vary in size, resources, relationships to the state, the degree to which they have faced enforcement actions, and the degree to which they are networked with other organizations. All these variables should affect their internal responses to law, yet the personnel charged with interpreting the law within these organizations articulated roughly similar understanding of the law's requirements, even though there are some technical differences in the formal rules that govern these organizations under the relevant ADA and state access provisions.Footnote 7
Within the private sector, we compared five independent, family-run restaurants,Footnote 8 which we collectively refer to as the “independent restaurants,” with Johnny's, a family-owned, medium-size restaurant chain with over 20 outlets, which had been sued under federal and state access laws. (See Table 5 for a summary.) In the public sector, we compared two prosperous, midsize cities with populations between 50,000 and 100,000): Sunny Valley, which had been the subject of a federal ADA action, and Shady Grove, which had not been the target of any formal enforcement action. In the nonprofit sector, we compared Sunny Valley University, a small (the student body is slightly more than 2,000) undergraduate and graduate university that has not faced any formal legal action, with Shady Grove University, a large (over 20,000) and well-funded research university that had been the target of a student mobilization that led to a federal agency complaint and, later, some lawsuits. Thus, while we do not have a representative sample of organizations under the ADA—far from it—we believe that our organizations’ variation within and across sectors provides a theoretically interesting sample for generating hypotheses and developing concepts for understanding organizational responses to the law and their relationships to outcomes related to the law's goals.
Table 5. Summary of Organizations
The Performance Measures
Assessing organizational response to law requires a strategy for measuring performance. Alas, as in many fields of regulation, including civil rights laws such as the ADA, it is not so clear what the yardstick should be—that is, what counts as successful implementation of the ADA access provisions. The access requirements under the ADA and its state counterparts are notoriously indeterminate. Regulations based on the ADA can be quite concrete and specific—for example, toilet rims are to be 17 to 19 inches from the ground—but the law itself has defenses and standards that are general and vague. For example, the ADA requires managers of facilities that are open to the public to make changes when they are “readily achievable,” a term whose parameters are not easily pinned down. As Reference KaganKagan (2001) has shown, an important aspect of life in adversarial legal societies is intense conflict over the meaning of legal texts. In disability law such conflict is rampant. It would be arbitrary and pointless to develop our own interpretations of the meanings of state and federal access laws and impose them on the data. Moreover, even if we could noncontroversially define compliance, we would not want to stop there, for one goal of social change law is to change consciousness and to stimulate organizational leaders to go “beyond compliance” (Reference Gunningham, Kagan and ThorntonGunningham, Kagan, & Thornton 2003).
Accordingly, our dependent variable is not compliance but accessibility, which is a central stated goal of both the ADA and disability rights activists. In practice, accessibility is a complex concept. Because there are many kinds of disabilities, there are many kinds of accessibility. To make our study more tractable, we decided to limit ourselves to wheelchair accessibility, but even here there is considerable complexity. There are roughly 1.7 million wheelchair and scooter users who live outside of institutions in the United States—a little more than 0.5 percent of the population—and the group ranges widely, from world-class athletes who use speedy manual chairs to people with quadriplegia who drive mechanically sophisticated vehicles operated by mouth (Reference Kaye, Kang and LaplanteKaye et al. 2002). Wheelchair accessibility involves hundreds of items that might seem trivial to nonusers, everything from the shape of door handles and the placement of bathroom mirrors to the insulating of heating pipes and the height of signs. Accordingly, we sought to identify a few key matters that could be measured relatively easily and unobtrusively, would appear in diverse settings, and would tend to be relatively inexpensive to address and thus likely to be “readily achievable.”
To do so, we started with the U.S. Department of Justice's checklist of readily achievable barrier removal, which was developed by the Adaptive Environments Center, Inc., and Barrier Free Environments, Inc. We then refined this checklist through a series of pilot studies in the field and focus groups with wheelchair users at the Berkeley and Boston Centers for Independent Living. From this process we created an accessibility index and scored each facility from 0 to 100, where 0 means wholly inaccessible across categories and 100 indicates complete equality with walkers across categories. (See Appendix A for further details on the construction of our accessibility measure, Appendix B for a discussion of our reliability checks, and Appendix C for a description of the general methodology for our organizational case studies.)
Findings
Organizational Response
Our organizations varied widely in terms of their responses to the ADA's and the state's access provisions, despite their common understanding of these requirements. On each dimension—commitment, professionalization, and routinization—there were examples that ran the gamut from high to low. (See Table 6.)
Table 6. Comparison of Organizational Responses: Relative Levels of Commitment, Professionalization, and Routinization
Before looking at how these variables related to the facilities’ actual levels of access, it is useful to take a closer look at each organizational response to the law to provide a more concrete sense of the richness of our sample.
Restaurants
As noted above, we compared five family-run, independent restaurants with Johnny's, a family-owned, medium-size restaurant chain with more than 20 outlets. The managers of the independent restaurants—typically the owners—evinced little knowledge of access law. They reported no designated staff to deal with access issues, no training about these issues, and no formal procedures. They reported doing their best to help individual customers with disabilities, but they had only the barest sense of access law, so they had no idea about how far their obligations extended. They also were the least networked of our organizations, so they had received little input on access law from lawyers, other businesses, or community organizations. In short, largely because of their ignorance of the law and their isolation, these organizations scored low across the board in terms of commitment, professionalization, and routinization.
Johnny's provides an interesting contrast. Like many chain restaurants in the state, it had been sued under state and federal access law. According to management, the suit was brought in bad faith; they considered it a strike suit aimed at wringing fees from the business under the applicable state law's fee-shifting provisions. As such, the managers expressed little commitment to the law; indeed, they were somewhat hostile to it.
Unwilling to face a protracted and expensive legal battle, Johnny's agreed to settle the suit, in part because it was in the process of updating its facilities anyway. Under the resulting settlement, Johnny's spent hundreds of thousands of dollars on upgrading its existing facilities to make them more accessible. Beyond this, Johnny's changed its design template for future sites. Thus accessibility guidelines were literally written into Johnny's standard plans for prospective restaurants during a time when the chain was expanding its operations. In one important respect, then, Johnny's operations had been transformed by its encounter with litigation. In all other ways, Johnny's response was limited and grudgingly adopted.
Johnny's response to the law was not professionalized. It neither created a staff to deal with new disability issues on an ongoing basis nor developed procedures or a written policy to address the issue. Instead, organizational leaders worked with their lawyers to comply with the minimum terms of the settlement. This meant that aside from its templates for new facilities, Johnny's response was not routinized. No understanding, much less concern, for accessibility issues filtered down through the organization's employees. This had significant consequences. For example, as part of the settlement, the company installed an intercom system at one of its older locations, which was particularly inaccessible. At the time of our inspection, the intercom had fallen into disrepair. Coincidentally, a company employee responsible for checking the area's facilities visited the site during our inspection. Although he had an extensive checklist of safety items to inspect, he had no items related to access on his checklist and did not bother to press the button on the intercom to see if it worked.
Johnny's key personnel were not committed to the law's social goals and were hostile to the lawyers who had brought the access suit; they described the lawyers as shakedown artists. (One of the lawyers was later declared a vexatious litigant by a federal judge.) Accordingly, they sought to meet the minimum requirements of the settlement as they understood it. Yet even so, concerns about access had clearly permeated Johnny's planning processes. In the shadow of litigation, management scanned the horizon for win-win strategies, particularly ways in which it could align the goals of improving access with the ongoing redesign of its facilities. Thus, though Johnny's response lacked commitment and professionalization, it had one important element of routinization: it had routinized facility planning by incorporating access features into its design template.
Cities
Overall, the cities’ responses were similar, scoring moderately in terms of commitment, professionalization, and routinization. The professionalization score is not surprising given the requirements of the ADA: cities are specifically required to designate an official responsible for access issues. Cities are not, however, required to promulgate official access policies or to train their employees in access issues, and neither city in our sample did these things, thus yielding a medium score for professionalization. Both cities designated longtime building and facilities officers to be their disability coordinators. These officials, befitting their background, took a pragmatic, service-oriented approach to their disability duties. They were not steeped in the philosophy that animates access law—the social model of disability—and so did not use the language of civil rights or discrimination in describing their approach to disability. Even so, they were respectful of the law's requirements. When asked to describe episodes in which access complaints had been made against the city, the officials evinced sympathy and respect toward people with disabilities and recounted cases in which they went beyond what they considered the minimum requirements of access law. Thus, the cities demonstrated a medium level of commitment along with a medium level of professionalization.
The ADA also requires cities to create a review of their facilities, thus setting the stage for a routinized, proactive plan for improving accessibility. In both cases these plans seemed long forgotten, and we found them collecting dust on bookshelves. That said, while there seemed to be no overall plan for improving accessibility in either city, there was some evidence that an informal system had developed: managers of city facilities appeared to understand that they should refer questions about access issues to the city's designated officer. In Shady Grove, city officials who managed facilities such as libraries and community centers understood that access issues were important in a general way and that if they had questions, they should ask the designated compliance officer, whom they knew by name.
When we interviewed the designated disability coordinator at Sunny Valley, the city was in the midst of implementing a settlement agreement that resolved an access lawsuit brought by the U.S. Justice Department against the city. The coordinator was (perhaps understandably) reluctant to reveal many details about the organization's historical response to access law. The settlement agreement, however, clearly represented a break with business as usual in the city. It mandated improvements in more than a dozen city facilities and set aside money specifically for those improvements. Perhaps more important, the settlement had stimulated the city to hire an outside consultant to help them assess issues; any time a significant issue arose, it was now sent to the consultant for review. The access lawsuit had changed the city's approach to access issues so that the city had become more proactive in spotting problems. As the disability coordinator noted, “Now [the consideration of disability issues is] part of our process, and that's a good thing.” The city had, in fact, set aside part of its budget to make the mandated improvements, though at the time of our inspections not all of these improvements had been made. In short, it appeared that, stimulated by the lawsuit, Sunny Valley was becoming more professionalized and routinized in its handling of access issues than was its counterpart, Shady Grove, though its efforts still fell well short of the most energetic response in our sample: Shady Grove University.
Universities
Shady Grove University scored the highest on all of our measures. In response to a high-profile student protest and a federal agency complaint, it had created a specialized office dedicated to disability issues. This office was well funded, with dedicated staff that routinely toured the campus with students with disabilities and took the initiative to address problems, even if it meant closing popular but inaccessible facilities and programs, such as a mobile dental clinic for students and staff. Unlike Johnny's key personnel, who viewed disability lawyers and lawsuits with suspicion, the Shady Grove University staff saw litigation as a potential lever for change within the organization and had even cooperated with a lawsuit against the university. These professionals, many of whom had disabilities, fully embraced the social change goals of the law, explaining that they wanted everyone on campus to see barriers to access in the same light as “colored” drinking fountains in the segregated South. In sum, this was a highly committed, professionalized office.
The Shady Grove University disability offices, moreover, were influential within the organization. The disability office was represented at meetings about proposals to build new facilities, deliberations over the rehabilitation of older buildings, and even plans to acquire temporary space. Moreover, when conflicts about access issues emerged, the disability office was able to appeal to the highest levels of university administration and reported success in doing so. Thus, Shady Grove University's response featured all three positive attributes: its key personnel were committed, the organization had become professionalized, and concerns about access had become routinized in the planning activities on campus.
In important respects, Sunny Valley University was similar. It had a disability office, and, in the two years prior to our inspections, it had hired a full-time disability specialist. This official had an extensive background in disability issues and embraced the social model of disability; like her colleagues at Shady Grove University, she framed inaccessibility as a civil rights issue. She was a member of multiple professional networks dedicated to access issues, including a regional group that met in person each month. But the disability official was marginalized within her organization. She was not involved in day-to-day decisions about the design of new buildings, nor was she given a budget specifically for access improvements. She had no regular input into the facilities planning process. Thus, while Sunny Valley University was professionalized, and its key personnel were committed, there was no routinization. Sunny Valley University represents a classic example of a decoupled response, in which the organization creates the symbols of response to law, including an office, but disconnects the office from its core operations.
Taking a step back from these details, several patterns emerge in the organizational responses in our sample. First, there was no necessary relationship between the commitment of key personnel and the degree to which access issues became routinized within the organization. At Johnny's the key personnel had no commitment at all, but accessibility had been routinized in the planning of new facilities; at Sunny Valley University there was a much stronger commitment to disability access than at Johnny's, but it seem confined entirely to the disability office.
Second, legal mobilization did not seem to lead organizations to become professionalized or committed in its responses to the law. In the case of Johnny's, being sued did not create a “conversion experience” that made the managers more committed to the law; in fact, it reinforced their negative attitudes toward the ADA. Legal mobilization did, however, seem associated with routinization. Two of the three organizations in our sample that had been the subject of a legal attack—Shady Grove University and Johnny's—were the most routinized in their approaches to access, and there were indications that the third, the City of Sunny Valley, was on its way to becoming more routinized as a result of the lawsuit it had settled.
Finally, size mattered. Only the more resourced organizations developed professionalized responses, and the only officials who expressed any commitment to the ADA's goal of accessibility were those designated as compliance officers. The smallest organizations, the restaurants, expressed little commitment to the goals of the ADA; their responses were thoroughly nonprofessional, as was Johnny's.
The larger organizations—the universities and the cities—exhibit what neo-institutionalists call isomorphism: a tendency for organizations in the same field to converge on the same professionalized responses, which are easily diffused by networks in the field. The smaller organizations, often overlooked in studies of response to law, look quite different. For smaller organizations there is no internal capacity to respond to law, so templates for responding to regulation tend to be diffused through specialists, lawyers, and consultants, and through business networks (Reference Barnes and BurkeBarnes & Burke 2006). This suggests that organizational response to law is fundamentally different for large and small organizations, and that social movements may need to think of smaller organizations differently when crafting legal mobilization strategies (Reference Thornton, Kagan and GunninghamThornton et al. 2009). It also suggests that studies of organizational response to law are strongly affected by the sizes of the organizations they sample.
Organizational Performance
Our data on accessibility are admittedly limited, in terms of both numbers and controls, but they suggest complex relationships between the patterns of organizational response and performance. Table 7 summarizes the results. Based on simple descriptive statistics, the independent restaurants performed the worst, with a mean access score of 37.66 out of 100. Shady Grove University performed the best by far, with a score of 76.11. The cities fell roughly in the middle, with scores of 44.37 and 53.46 for all of their facilities (including parks), and 51.72 and 66.92 when we looked only at their buildings.
Table 7. Comparison of Commitment, Professionalization, Routinization, and Access Scores
* p < .10, **p < .05, ***p < .01, ****p < .0005.
p-values are for differences within sector and, in the case of the cities, across facility type, using ANOVA and excluding facilities where accommodation is not readily achievable.
There were notable differences within each sector as well, and, in this sample at least, it corresponded with mobilization. Johnny's outperformed the other restaurants; the City of Sunny Valley outperformed the City of Shady Grove overall and with respect to its buildings; and Shady Grove University significantly outstripped Sunny Valley University. Thus, in all three pairings the organization that had been the subject of a legal mobilization produced the most access, although the number of observations and lack of controls require these findings to be viewed with caution.
The most surprising findings were Johnny's and Sunny Valley University. Johnny's lacked commitment and professionalization, yet its partially routinized response yielded relatively good results—a score of 59.75, which was roughly on par with the cities’ buildings. Sunny Valley University was highly committed and professionalized, yet because its disability officer was marginalized its response was not routinized; it received a mean access score of 38.26, or about the same as the independent restaurants and about half as high as its counterpart, Shady Grove University.
Discussion
“Talk is cheap,” goes the common dictum. And indeed, there was no necessary relation between the way key personnel in our organizations talked about disability access and the actual accessibility of their facilities. While the disability officer at Sunny Valley University was a disability rights believer, the managers at Johnny's were scornful of access law. Yet a wheelchair user would have an easier time navigating Johnny's than he or she would navigating the university—and would notice recent, costly improvements that make it easier to get around. This finding reminds us that measures that are grudgingly adapted can produce meaningful results, or, more simply, that, as the rational-choice approach to regulation emphasizes, the threat of punishment works.Footnote 9
But that is not to say that talk—or, to put it more precisely, the ways in which key personnel responded to disability access law, which we label commitment—did not matter. In organizations in which these personnel were in a position to routinize their interpretations, and in which their view of the proper approach to access permeated the organizations’ everyday practices, the backgrounds of these interpreters—steeped in the disability rights movement, or, as in the case of the city managers, only dimly aware of it—had a significant effect. At Shady Grove University, the disability office had achieved significant power within the organization, and the true believers in the organization had driven the university to become the most accessible location in our sample, scoring 76 on our index, well above the cities. In fact, this number only partly captures the quality of Shady Grove University's response. It was unique in this study because it went well beyond what any fair-minded observer would consider its obligation under the law—it was “beyond compliance.” For example, most of the facility entrances, and even some bathrooms, had electric doors, a feature we rarely saw in other sites. Similarly, the university took the extraordinary step of shaving down a stairway to widen access to a bathroom in a historic building. Finally, when compared to other organizations, including the cities and Johnny's, the university campus was remarkably free from careless practices that can render accommodations useless, such as boxes left in the middle of access ramps or trashcans partially blocking doorways or sinks. This suggests a high level of vigilance on the part of university staff. The case of Shady Grove University suggests that commitment, professionalization, and routinization are at their most powerful when they come together.
This finding has important implications for litigation strategies. As part of our study, we interviewed both private and state disability rights attorneys, who emphasized the importance of getting organizations to routinize their practices. In describing how to bring this about, one lawyer argued that the key is finding the “right” person inside the organization, meaning someone who would internalize the disability rights perspective and serve as an advocate within the organization. Our cases suggest that having an internal advocate, though valuable, is not sufficient for producing the best results on the ground. Internal advocates may be marginalized, as in the case of Sunny Valley University; only when the disability office is integrated into the everyday workings of the organization are the best results achieved.
Mobilization, similarly, was no magic bullet. For each type of organization, restaurant, city, and university, the organization that had faced some kind of mobilization was most accessible. But mobilization did not create convergence; the access scores of the organizations that had faced a legal complaint varied significantly, from a low of 59.75 to a high of 76. A wheelchair user who rolled through Shady Grove University would have a significant easier time than he or she would as a customer at Johnny's (though the user would also be grateful not to be stuck at one of the independent restaurants).
Why do we emphasize all the disjunctions in our sample? We began by suggesting the wide range of literatures that analyze how and under what conditions law can create social change, including scholarship on the mobilization of rights, rational choice–based regulatory research, and neo-institutional accounts of organizational response. We also noted a lack of these literatures’ connections with one another, despite their common aspiration to understand social change through law. In part this is driven by these literatures’ different conceptions of law, but we wonder if there is more to it than that. We perceive a tendency in all these literatures to emphasize some patterns and mechanisms at the expense, sometimes the exclusion, of others. This may be driven, or at least reinforced, by the cases on which these different traditions focus. If, as with neo-institutionalism, the focus is on large organizations and civil rights employment law, the question of symbolic response—of highly visible actions performed in order to retain legitimacy (but with ambiguous effects)—becomes paramount. If, on the other hand, the focus is on controversial “vanguard” rights movements, then changes in discourse and consciousness may be most consequential. And if one is studying how small businesses respond to environmental or safety laws, an emphasis on deterrence and punishment rather than on discourse or organizational structure is likely appropriate.
In our small sample, we see examples of all these phenomena at work. The designation of disability officers was a way for the cities to legitimate their responses to access law, but the effectiveness of these responses in fostering accessibility was somewhat ambiguous. The new consciousness about disability generated by the disability rights movement seems to have passed by the cities and restaurants, but it had a powerful effect on the personnel at the universities. And the brute fact of punishment clearly had a tangible effect on Johnny's. So each of these patterns is a part of the story of how access law results, or fails to result, in greater accessibility for people with disabilities.
Perhaps it is understandable that each of the literatures on social change through law tends to focus on a subset of this story. The downside of this specialization, though, is a more piecemeal, less holistic understanding of how laws are translated into organizational practices, discourses, and concrete outcomes. Further, the interactions among these elements—the relationships between discourse and practice, practice and outcome—are neglected. Our empirical focus on disability access rights and our methodology, which involves talking to key personnel, tracing organizational patterns, and measuring concrete outcomes, have allowed us to probe these interactions, albeit in a small sample, and to develop concepts—commitment, professionalization, and routinization—that can be useful building blocks for a more holistic study of organizational response to law. Even in our small sample, this approach generated interesting hypotheses to be tested as data accumulate. These include hypotheses about (1) the relationships among these factors (e.g., commitment and professionalization are highly correlated, while routinization varies independently); (2) the relationship between external pressures and these factors (e.g., legal mobilization engenders routinization but not commitment); and (3) the relationships between the organizational response variables and the outcomes (e.g., routinization is a necessary condition for tangible results, but high levels of commitment, professionalization, and routinization together have the greatest effect).
Of course, some might fairly argue that our findings, in addition to being limited by their small sample size and lack of statistical controls, are conditioned on the specific issue area. Disability access issues feature a whole host of distinctive attributes, including the public nature of organizational responses; the fact that some issues can be resolved with relatively cheap, episodic actions (such as installing a ramp); and the fact that, unlike other areas, there is no widely accepted professional template for organizational responses similar to “legalized accountability” (Reference EppEpp 2009). Recognizing these distinctive features, however, underscores the broader point of our argument. Given the protean and context-specific nature of rights, it is critical to develop an approach to studying them that is flexible enough to account for a wide range of interactions among key variables, such as mobilization, commitment, professionalization, and routinization.
Conclusion
Research agendas proceed at different paces depending on the state of existing theory, methods, and data. Some research agendas are mature, featuring well-developed data sets and well-specified theories that lend themselves to rigorous hypothesis testing. Others are less mature, as data prove limited and hard to gather and concepts continue to evolve. In our view, the study of how organizations translate law into practices that yield social outcomes remains relatively immature, though these issues have percolated though various literatures for quite a long time. One major stumbling block has been the disconnects among these literatures.
In this article, we have attempted to illustrate an alternative approach that brings together the insights of these literatures by developing a core set of concepts—mobilization, commitment, professionalization, and routinization—and applying them to a variety of organizational responses to access law. Our preliminary analysis found that these concepts could adequately describe the wide range of organizational responses in our sample and reveal intriguing patterns that warrant further investigation. Our fondest wish is that this approach helps establish a foundation for empirical studies of organizational response to law that systematically build on one another—something that would surely advance our understanding of a central concern of sociolegal studies: how and when law changes society.
Appendix A
Measuring Accessibility
For reasons described in the text, we decided to focus on wheelchair accessibility, not compliance, as a key dependent variable. Accordingly, we sought to identify a few key matters that could be measured relatively easily and unobtrusively, would appear in diverse settings, and would tend to be relatively inexpensive to address and thus likely to be “readily achievable.”
We started with a document on the U.S. Department of Justice Web site (http://www.ada.gov/checkweb.htm), entitled “Checklist for Readily Achievable Barrier Removal,” which was developed by the Adaptive Environments Center, Inc., and Barrier Free Environments, Inc. This checklist was useful, but it provided only a starting point because it is simply an interpretation of what the ADA requires of facilities built before the law came into effect. Given that our dependent variable is accessibility, we needed to understand what an ideally accessible facility—one in which walkers and wheelchair users would be equally mobile—would look like. This ideal of equal accessibility clearly goes beyond the requirements of state and federal law, even for new facilities. On the other end of the spectrum, we needed to understand what elements of a facility could make it completely inaccessible.
To learn what was ideal and what was most problematic, we turned to the best experts on wheelchair accessibility: wheelchair users. We conducted focus groups and surveys of a half dozen wheelchair users at each of two sites, the Berkeley Center for Independent Living and the Boston Center for Independent Living. We asked the wheelchair users to rate the significance of various features and then conducted an open-ended discussion with the participants.
We found again that accessibility is not a simple concept. The wheelchair users we surveyed varied widely in their impairments, and this affected their views on what was ideal and what was most important. In some cases, features that make a facility more accessible to people with one type of impairment may make the facility less accessible to people with another type. We tried to find areas of agreement or at least areas where there was a broad majority, but we cannot pretend that any index of accessibility will be accurate for all wheelchair users, much less for all people with disabilities.
Based on the U.S. Justice Department's checklist, our focus groups, and several rounds of field testing, we developed an inspection checklist that includes 51 items relating to outside access, parking, rest rooms, drinking fountains, elevators and lifts, and service counters. (A copy of the inspection checklist is available from the authors upon request.) Each initial measure was scaled on a −2 to 2 scale, with −2 indicating complete inaccessibility and 2 indicating complete accessibility. We also recorded spare comments about each facility. All inspections were conducted by the authors, sometimes together, usually separately, in 2007 and 2008. We had extensive discussions about coding decisions in the initial stage of coding, which occurred in a four-month period in early 2007.
During our discussions, we realized that the large number of parameters for different aspects of each facility (outside access, entrance, bathroom, parking, drinking fountain, and counters) could be combined into a single score for each aspect. We developed a much simpler coding scheme for these six aspects and, using our original sheets and the new coding scheme, recoded the scores. In the new coding, four aspects of the facility—general access, entrance, bathroom, and parking—were coded from −3 to 3, with −3 representing complete inaccessibility, 3 representing complete equality with people on foot, and 0 representing conditions that could be said to roughly fall in line with the standards indicated in the U.S. Department of Justice checklist. Drinking fountains, a much less important feature, were coded from −1 to 1, and counters, a small—but, according to our wheelchair-using respondents, important—aspect, were rated on a scale of −2 to 2. This weighting generally reflected the judgment of the wheelchair users in our focus groups. (A copy of the coding rules and a coding sheet for the composite coding based on our initial inspection sheets are also available upon request.)
To create an overall facility accessibility score, we added the individual aspect scores, and then added this number to the maximum positive score, divided by the range of possible scores for each facility, and multiplied by 100. For example, a facility that had all six features, giving it a maximum positive score of 15, and scored a 0 on each feature would receive a score of 50: [(15 + 0)/30] ∗ 100. If the facility scored −1 across the board, it would receive a score of 30: {[15 + (−6)]/30} ∗ 100. The resulting index of accessibility is a scale from 0 to 100. In the aggregate, inspections produced a distribution of composite accessibility scores with a range of 0 to 100, a mean of 54.48, a median of 57.14, a standard deviation of 25.11, skewness of −0.24, and kurtosis of 2.29.
Appendix B
Reliability
Reliability is a central concern in this type of research. To be at all useful, coding must reflect actual variation in the cases as opposed to the idiosyncrasies of the coders. To test reliability, we took a 20 percent random sample of sites inspected by one of the authors. The other author then returned to the site and independently reinspected it at a later date. This process had the added benefit of assessing whether rights practices changed significantly over time in our sample. They did not.
An analysis of intercoder agreement using Kappa suggests that had the cases been coded randomly (but with the probabilities equal to the overall proportion of cases), we would have expected agreement in about 65.75 percent of the cases. In fact, there was agreement in 94.98 percent of the cases, which is significantly above that which would be expected by chance (p > 0.00005). It should be added that, because we coded on a series of scales, we weighted the observations to account for the degree of disagreement. Thus, perfect agreement on an item would be weighted as 1. So, for our initial coding, in which scores ranged from 2 to −2, total disagreement (2 versus −2) would be weighted as 0, indicating no agreement. Partial disagreement, such as 1 versus 2, would be weighted as 0.8, suggesting four-fifths agreement. Differences of more than one point on our scale were very rare. (For more on this issue, see STATA Reference Manual, Version 7.0, Volume 2, page 151.) We also ran the Kappa test without weighting the results, so that any disagreement would be coded as 0 or total disagreement, and the results still indicated that the levels of intercoder reliability were still significantly greater than would be expected by chance.
Finally, we tested the reliability of the recoding of the facilities data into composite scores. So, as before, we took a 20 percent random sample of cases by one author, and the other author independently coded them. Again, the Kappa tests suggest that our coding was reliable, as we agreed in 93.55 percent of the cases, which was significantly greater than if we had coded the cases randomly according to the underlying distributions in the sample (p > .00005).
Appendix C
Case Method
In conducting our case studies of organizations, we engaged in the following steps.
First, using local phone books and various online sources, we developed a list of possible organizations and contacts.
Second, we made initial contacts, asking if they would participate in our study and, if so, who was in charge of access issues for people with disabilities. In every interview, we asked if there were others within the organization whom we should interview. Through this type of snowball sampling, we were able to identify and interview the network of personnel who may have served as interpreters of the law in the organizations.
Third, after obtaining appropriate consents, we conducted interviews with our subjects. The interviews were structured by a common set of questions and lasted from 20 minutes to over an hour, with most interviews lasting about 30 to 40 minutes.
Fourth, whenever possible, we examined the organizational written policies and media accounts of any access issues that may have arisen in the past. Thus, the case studies reflect a mix of data sources, including interviews and content analyses of documents.
