Karlene H. Roberts
The field of high reliability organizations (HRO) research is now over thirty years old. This chapter discusses original reasons for delineating this area of research and the nature of the early research. I go on to indicate the reasons this book is needed at this time and conclude with a brief description of each chapter.
In the Beginning
HRO research began in the mid-1980s at the University of California, Berkeley. Berkeley researchers Todd La Porte, Karlene Roberts, and Gene Rochlin were soon joined by Karl Weick, then at the University of Texas. We were interested in the growing number of catastrophes that appeared to us to have been partially or wholly caused by organizational processes, including the 1981 Hyatt Regency walkway collapse; the 1994 South Canyon Fire, about which Weick wrote (1995, 1996); and the 1984 Union Carbide disaster in Bhopal, India. Unfortunately, the list goes on and on. Some time ago we were asked to write reflections of the early work. The first section of this chapter draws heavily on those reflections.
As Weick (2012) points out:
Prominent ideas were available to analyze evocative events such as the preceding [Hyatt Regency walkway collapse, etc.]. [Charles] Perrow (1984) had proposed that increasingly tightly complex systems fostered accidents as normal occurrence, a proposal that encouraged counterpoint. Barry Turner (1978) had sketched the outlines of organizational culture, the incubation of small failures (later to be conceptualized as “normalization”) and the organizational blind spots. “As a caricature it could be said that organizations achieve a minimal level of coordination by persuading their decision makers to agree they will all neglect the same kinds of considerations when they make decisions (p. 166).” Not long thereafter, Linda Smircich (1983) in a definitive [Administrative Science Quarterly] article gave legitimacy to the notion of organizational culture. Trial and error learning was a basic assumption which meant that, the possibility that groups in which the first error was the last trial provoked interest and a search for explanations. . . . My point is that these ideas, and others not mentioned, were available to make sense of an emerging set of organizations that were complex systems, under time pressure, conducting risky operations, with different authority structures for normal, high tempo, and emergency times, and yet in the best cases were nearly error free. (p. 2)
At the beginning of our research, we were introduced to three technologically sophisticated, complex subunits of organizations that were charged with doing their tasks without major incident: the US Navy’s Nimitz-class aircraft carriers, the US Federal Aviation Administration’s [FAA] air traffic control [ATC] system, and Pacific Gas and Electric Company’s [PG&E] Diablo Canyon nuclear power plant. To us these organizations appeared to engage in different processes, or seemed to bundle the same processes differently, than the average organization studied in organizational research. We kicked off the research with a one-day workshop held on the aircraft carrier USS Carl Vinson, which was also attended by members from the other two HRO organizations we planned to study. One outcome of the workshop was that managers in all three organizations felt they faced the same sophisticated challenges.
Important Characteristics of the Initial Work
THE HRO project did not start by looking at failures but rather at the manner in which organizations with a disposition to fail had not. It became readily apparent that HROs do not maintain reliability purely by mechanistic control or by redundancy or by “gee whiz” technology. They work into the fabric of these mechanistic concerns a mindset and culture that makes everyone mindful of their surroundings, how they are unfolding, and what they may be missing. High reliability organizing deploys limited conceptions to unlimited interdependencies. These organizations are set apart from other organizations because they handle complexity with self-consciousness, humility, explicitness, and an awareness that simplification inherently produces misrepresentations.
The initial high reliability project focused on current functioning because we researchers knew little of past practices and operations. In all cases it took many years to reach the level of performance observed by the researchers. For example, the Air Commerce Act was signed into law in the United States in 1926. It contained provisions for certifying aircraft, licensing pilots, and so on. By the mid-1930s there was a growing awareness that something needed to be done to improve air travel safety. At the same time, the federal government encouraged airlines to embed control centers in five US airports. Maps, blackboards, and boat-shaped weights were used to track air traffic. Ground personnel had no direct radio link with aircraft, and ATC centers contacted each other by phone.
Technological changes have vastly altered what high reliability functioning looks like today. Not long after the emergence of ATC centers, semi-automated systems were developed based on the marriage of radar and computer technology. In 2004 the US Department of Transportation announced plans to develop a “next gen” plan to guide air traffic from 2025 and thereafter. This plan will take advantage of the growing number of onboard technologies for precision guidance (Federal Aviation Administration, 2015).
Because the researchers had no experience with the histories of these organizations, many existing organizational processes that may no longer serve a purpose were probably not uncovered. An apocryphal story is told about the US Army on the eve of World War II. A senior officer was reviewing an artillery crew in training. The officer noticed that each time the gun fired, one of the firing team members stood off to the side with his arm extended straight out and his fist clenched. The inspecting officer asked the purpose of this procedure, but no one seemed to know. Sometime later a World War I veteran reviewed the gun drill and said, “Well, of course, he’s holding the horses.” An obsolete part of the drill was still in use (Brunvand, 2000), as is probably true in HROs.
The units under study in the original research were subunits of larger organizations and not necessarily representative of the organization as a whole. Flight operations, for example, are essential to the missions of an aircraft carrier but not its entire menu of complex tasks, which include navigation, weapons handling, supply, housing and feeding six thousand people, and so on (Rochlin, 2012). The carrier is central to the task force and is an important part of the navy, which is part of the US Department of Defense. It was beyond the scope of the project to determine the contribution to the nested series by safer or, rather, more reliable operation of the suborganizations. By studying subunits the team may have created an error of the third kind (Mitroff & Featheringham, 1974)—that is, solving the wrong problem precisely by only examining part of the problem. Paul Schulman and Emory Roe attempt to rectify this problem in Chapter 9.
More research is needed to explore how organizations or units of organizations that can fail disastrously are linked to other organizations or units of organizations. Specifically, more attention needs to be given to organizations that help other organizations fail or fail alongside them (Roberts, 2012). The failure of BP and its semisubmersible deepwater drilling rig Deepwater Horizon is a good example. As reported, on April 20, 2010, the Macondo well blew up; this accident cost the lives of eleven men and began an environmental catastrophe when the drilling rig sank and over four million barrels of crude oil spilled into the Gulf of Mexico (National Commission on the Deepwater Horizon Oil Spill and Offshore Drilling, 2011, back cover).
There are a number of reasons for the paucity of research on interlinked and interdependent organizations. Such research is costly and resource demanding. Moreover, most organizations in which some units need to avoid catastrophe are complex in a manner that would require large, multidisciplinary research teams. The Diablo Canyon nuclear power plant is an example of an interdependent, hence complex organization requiring excessive resources. It is enmeshed in the problems, politics, and legalities of PG&E and its regulator, the California Public Utilities Commission (CPUC), to say nothing of local community politics. Building a multidisciplinary (or interdisciplinary) team is not easy. Scholars are not used to talking with other scholars who speak different languages and are enmeshed in different constructs.
The observations at the root of the original HRO conceptual and process findings were intense case studies of the three organizations. Early on, we realized that formal interviews and questionnaires were of little value in organizations in which researchers didn’t have an available literature on which to build. Both these research methodologies assume researchers know some basics about what is going on in the organization.
A Conceptual Problem that Doesn't Go Away
A frequent criticism of HRO research is the lack of agreement on a definition by the (now) many authors contributing to the work (e.g., Hopkins, 2007). The chapters in this book reflect this lack of consensus. Despite this repeated criticism, early in the work Rochlin (1993, p. 16) provided a list of defining criteria that seem to provide fairly clear boundaries for organizations to be labeled as high reliability or reliability seeking:
The labeling problem is further compounded by the fact that most high reliability research still selects on the dependent variable by first identifying organizations that researchers think are or should be high reliability or reliability seeking. But, does reliability mean the same thing to all employees in a single organization or across organizations? Definitional problems, too, may have led to the fact that the research project went by several names before “high reliability” stuck as the work matured. It is disconcerting that the acronym HRO has become a marketing label: “When it is treated as a catchword this is unfortunate because it makes thinking seem unnecessary and even worse, impossible. The implication is that once you have achieved the honor of being an HRO, you can move on to other things” (Weick, 2012).
Early Research Findings
According to Chrysanthi Lekka (2011), the original research identified several characteristics and processes that enabled the three organizations to achieve and maintain their excellent safety records (e.g., Roberts & Rousseau, 1989; Roberts, 1990b, 1993a; La Porte & Consolini, 1991; Roberts & Bea, 2001). These include:
Deference to expertise. In emergencies and high-tempo operations, decision making migrates to people with expertise regardless of their hierarchical position in the organization. During routine operations decision making is hierarchical.
Management by exception. Managers focus on the “bigger picture” (strategy) and let operational decisions be made closer to the decision implementation site. Managers monitor these decisions but only intervene when they see something that is about to go wrong (Roberts, 1993a; 1993b).
Continuous training. The organization engages in continuous training to enhance and maintain operator knowledge of the complex operations within the organization and improve technical competence. Such training also enables people to recognize hazards and respond to “unexpected” problems appropriately and is a means to build interpersonal trust and credibility among coworkers.
Safety-critical information communicated using a number of channels. Using a variety of communication channels ensures that workers can receive and act on information in a timely way, especially during high-tempo or emergency operations. For example, at the time of the research, nuclear-powered aircraft carriers used twenty different communication devices ranging from radios to sound-powered phones (Roberts, 1990b). Currently, the Boeing 777 aircraft uses eight communication devices.
Built-in redundancy. The provision of backup systems in case of failure is one redundancy mechanism. Other such mechanisms include internal crosschecks of safety-critical decisions and continuous monitoring of safety-critical activities (e.g., Roberts, 1990b; Hofmann, Jacobs, & Landy, 1995). Nuclear-powered aircraft carriers operate a “buddy system” whereby activities carried out by one individual are observed by a second crew member (Roberts, 1990b).
Organizing for Reliability: A Guide for Research and Practice
The central purpose of Organizing for Reliability is to showcase the different perspectives about high reliability organizing that have emerged over the past thirty years. We feel that all too often reliability is embedded in studies of accidents (e.g., Caird & Kline, 2004), safety, (e.g., Naveh & Katz-Navon, 2015), errors (e.g., Jones & Atchley, 2002), and disasters (e.g., Guiberson, 2010) rather than being given attention in its own right. The basic question in HRO research is, What accounts for the exceptional ability of some organizations to continuously maintain high levels of operational reliability under demanding conditions? Its importance is underscored by distant accidents (e.g., the Challenger and Columbia space shuttles) and more recent mishaps (e.g., the Volkswagen emissions debacle, PG&E’s San Bruno, California, pipeline explosion) in which organizations failed to operate reliably.
Contributors to this book are richly diverse in terms of career stages (from junior to very senior authors), academic disciplines (organizational studies, industrial/organizational psychology, social psychology, sociology, communications, public health, and public policy), and familiarity with many organizational contexts. One consensus of these authors is clear: it is time to provide a relatively full panoply of HRO research in one place so that researchers can identify what they think are next steps, and practitioners can pick up strategies to help them in making their organizations more reliable. Organizing for Reliability is organized in three parts. Part I, “Setting the Stage,” offers some important background to high reliability research. Chapter 1 provides a short history of the work and an introduction to the remaining chapters.
Chapter 2 discusses the various meanings of reliability in organizational research and the need to develop context-specific models of reliability. Rangaraj Ramanujam notes that reliability is an increasingly multifaceted construct that currently covers multiple notions (levels of analysis, organizational capabilities, and assessment criteria). As a result, the overlap among various disciplinary approaches is growing. This overlap presents untapped opportunities for research on reliability, and Ramanujam identifies many for us. Finally, Ramanujam discusses implications of multiple notions and models of reliability for future research. In all this he broadens out the relatively narrow purview of the history of the research.
John S. Carroll opens Part II, “Important Aspects of Reliable Organizing,” by examining reliability in the broader context of interrelated theoretical concepts and perspectives. Framing reliability as “an intersection of effectiveness, safety, and resilience,” he proposes that organizing for reliability can be more meaningfully understood by viewing organizations through three distinct lenses: strategic design, political, and cultural. Carroll cautions against relying exclusively on any one perspective noting that multiple perspectives must be considered together to recognize “the complex interdependencies of the organization, the difficulties of implementing change, and the heterogeneity among individuals and groups.” He concludes Chapter 3 by identifying the dominant perspective underlying each of the subsequent chapters.
In Chapter 4 Kathleen M. Sutcliffe investigates mindful organizing, a dynamic process comprising ongoing patterns of action that fuel capabilities to more quickly sense and manage complex, ill-structured contingencies. Mindful organizing enables collective mindfulness through actions that promote “preoccupation with failure, reluctance to simplify interpretations, sensitivity to operations, commitment to resilience, and flexible decision structures that migrate problems to pockets of expertise,” all of which operate at multiple organizational levels. Sutcliffe traces the conceptual underpinnings of mindful organizing and the growth in theory and empirical research. She concludes that although mindful organizing was initially studied in the context of “prototypical” HROs, it is increasingly important for all organizations and for nonreliability outcomes such as innovation. She calls for more research to better understand the meaning of reliability as a property of relationships that can be developed and enhanced in mundane settings.
Seth A. Kaplan and Mary J. Waller focus on team resilience in Chapter 5. Team composition in high-risk work settings is typically subject to ongoing change that accelerates during crisis situations. These authors discuss several strategies that teams can implement to enhance resilience. First, they discuss the roles of internal team dynamics such as changes in team composition and emergent states. Second, they point to the importance of team boundary dynamics such as adapting team composition to events, rapid onboarding of new members, and managing internal and external communication with stakeholders. Third, they suggest that these capabilities can be the effects of specific actions developed and enhanced through simulation-based training. Simulations can create a learning context that allows teams to experience the interplay of dynamic behaviors and states and develop resilience-enhancing behaviors and emergent states.
Peter M. Madsen and Vinit Desai examine goal conflict in the context of organizational reliability in Chapter 6. They note that the tension between reliability goals and efficiency goals is particularly magnified and problematic in high-risk work settings where multiple goals are high in both their performance relatedness (e.g., decisions that affect reliability are also likely to affect other performance-linked goals) as well as their causal ambiguity (e.g., uncertainty regarding specific actions on various performance dimensions). One implication is that typical approaches to resolving goal conflict such as identifying and pursuing a single overriding goal or addressing multiple goals concurrently or sequentially may be inadequate or even infeasible. A collective dynamic capability to continually assess and resolve goal conflicts is required. These authors identify and elaborate on four processes that enable HROs to manage goal conflicts—continuous bargaining over goal priority, balancing incentives to reward both efficiency and safety, incremental decision making that builds on the analysis of even very small errors and failures, and commitment to resilience that enables the organization to allocate operational resources to areas of operational weakness whenever problems are anticipated.
In Chapter 7 Madsen assesses the role of organizational learning in enhancing reliability. He points out that sustained reliability in high-risk/high-hazard settings requires organizations to engage in different forms of learning—experiential learning, vicarious learning, learning from near misses, and simulation learning. He reviews the HRO literature as well as the broader organizational literature and offers an integration using Turner’s (1978) disaster incubation model. The four learning strategies together facilitate the early identification and correction of latent errors, appropriate timely response to contain the severity of a disaster as it unfolds, and effective cultural readjustment in the aftermath of a disaster.
Jody L. S. Jahn, Karen K. Myers, and Linda L. Putnam discuss three mechanisms of reliability enhancement and state that they depend on communication to make it possible to negotiate meanings and actions in organizations. These authors show that studies of HROs often conceptualize communication as accessory to action or as operating at the periphery of organizations. Chapter 8 illuminates these perspectives by comparing communication-relevant studies of HROs with different views of what communication is and how it functions. The authors adopt five metaphorical lenses to review the role of communication in the HRO literature, which they find has relied on some lenses rather than others. Finally, they demonstrate how HRO research could be strengthened by adopting more complex understandings about the relationship between communication and highly reliable performance.
Paul R. Schulman and Emery Roe propose a reformulation of the scope and time frame of reliability research in Chapter 9, arguing that reliability is increasingly the property of an interorganizational network, not just a single organization as typically assumed. They identify public dread of hazard and regulation as critical yet understudied factors that shape the practice of reliability management. They examine reliability using different standards and performance states. Most studies focus on a single standard—precluded events (i.e., avoiding outcomes that simply cannot happen such as a radioactive release into the environment). However, several other standards must also be considered—for example, inevitable events (i.e., cannot be avoided, but the focus is on speedy recovery such as in a power outage) and avoided events (i.e., should be avoided such as foreseeable errors). Similarly, the assessment of reliability must take into account an expanded set of performance states (e.g., normal operations, disruption, restoration, failure, recovery, and establishment of a “new normal”) and adopt a much longer-term (“generations”) orientation than is currently the case.
Many readers may want to turn immediately to Part III, “Implementation,” as it contains numerous ideas about how to improve reliability and provides some cautions about what, perhaps, not to try. In Chapter 10 Peter F. Martelli focuses on improving reliability in the health-care sector. He begins by noting that uniform reliable safety has not been achieved in these organizations and continues by discussing assumptions and challenges to high reliability in the quest for quality and safety in health care. Health care’s first forays into the HRO world were in anesthesia, which borrowed principles from aviation. High reliability was rarely mentioned in the health-care literature in the 1990s. Instead, formal language and perspectives came from human factors engineering. By the mid-2000s, reliability was being regularly discussed in the literature, and two books that greatly influenced the future of high reliability research and application were published. In the last decade, acceptance of high reliability as an approach to providing greater safety in health care has been steadily growing. Still, there are significant gaps in understanding HRO as a theory, especially with respect to scope conditions and transferability and preconditions to sustained implementation. Martelli describes one of the larger implementation efforts. After reviewing what HRO is and what problem it is trying to solve in health care, he ends with a call to “preserve our pioneer spirit, drawing lessons across disciplinary boundaries in order to explore the character of work, the character of error, and the technical, economic, and institutional forces that promote reliability seeking in complex, interdependent organizations.”
Louise K. Comfort examines organizing for reliability at the community level in Chapter 11. While the majority of HRO research has dealt with single organizations, she notes that organizations are nested in various levels of other organizations and systems of organizations, which create complex adaptive systems of systems (CASoS). The issue is how each system can scale up and down to achieve coherent performance. To illustrate the breakdown in CASoS, Comfort examines responses to the 2004 Indonesian earthquake and the 2011 Japanese earthquake, tsunami, and nuclear breach. She reviews how planning processes differed in each case and how various factors reduced or increased risk. Finally, she provides steps essential to building global resilience.
In Chapter 12 W. Earl Carnes addresses the issue of just how difficult it is to implement high reliability processes in organizations. He comes from the world of implementation and provides a unique perspective in this book. Preferring the term “reliability seeking organization” (RSO), Carnes first discusses what a high reliability world looks like by answering the following questions: (1) How do RSOs get to be this way? (2) Why is it so hard to see what people in RSOs do? and (3) What’s important to know if you want to be an RSO? To address the issue of implementation, Carnes invited thirty-one professional colleagues to share their answers to the question, “What are the most important lessons you have learned about applying high reliability principles and procedures?” He categorizes their responses into five high-level lessons learned. Carnes then turns to some of the important behaviors mentioned by his respondents. He concludes by asking how to implement these behaviors. In implementation we must ask three questions: What does “good” look like, how are you doing, and how do you know?
In the epilogue, we point out that a common definition of reliability remains elusive and that the expansive definitional set we have suggests the need for integrating research from different overlapping fields of inquiry. Since different authors discuss different levels of analysis, we also need work on how reliability at one level influences reliability at another level. While authors discuss different theoretical perspectives, most focus on only one or, at best, two perspectives; the political perspective, as discussed by Carroll, is not often discussed in the wider literature. Finally, one is still left with the question of whether HRO findings are generalizable across organizations
1. In December 2016 Worldwatch Institute reported that natural disasters are on the increase. Even earlier, in 2014, the secretary general of the International Civil Defense Organization stated, “Today, natural and man-made disasters are increasing in frequency and destructive capability” (Kuvshinov, 2014, p. 1).
2. The original research was done on Nimitz-class carriers. The US Navy is currently building a new carrier class, the Gerald R. Ford class, and the first carrier in the series has been delivered for sea trials. This class is characterized by new technologies as well as other design features intended to improve efficiency, reduce operating costs, reduce crew size, and provide greater comfort. (See http://www.thefordclass.com; Bacon, 2014).
Bacon, L. M. (2014, October 13). Crew’s ship: Sailors’ comfort a centerpiece of new supercarrier Ford. Navy Times. Retrieved from https://www.navytimes.com/story/military/tech/2014/10/13/crews-ship-sailors-comfort-a-centerpiece-of-new-supercarrier-ford/17180325/
Brunvand, J. H. (2000). The truth never stands in the way of a good story. Urbana: University of Illinois Press.
Caird, J. K., & Kline, T. J. (2004). The relationships between organizational and individual variables to on-the-job driver accidents and accident-free kilometres. Ergonomics, 47, 1598–1613.
Federal Aviation Administration. (2015). A brief history of the FAA. Retrieved from https://www.faa.gov/about/history/brief_history/
Guiberson, B. Z. (2010). Disasters: Natural and man-made catastrophes through the centuries. New York: Holt.
Hofmann, D., Jacobs, R., & Landy, F. (1995). High reliability process industries: Individual, micro, and macro organizational influences on safety performance. Journal of Safety Research, 26(3), 131–149.
Hopkins, A. (2007). The problem of defining high reliability organisations (Working paper 51). Canberra: Australian National University.
Jones, T. C., & Atchley, P. (2002). Conjunction error rates on a continuous recognition memory task: Little evidence for recollection. Journal of Experimental Psychology: Learning, Memory, and Cognition, 28, 374–379.
Kuvshinov, V. (2014, March 1). Message of Dr. Vladimir Kuvshinov, secretary general of the International Civil Defence Organisation, on the occasion of the 2014 World Civil Defense Day. Retrieved from http://www.icdo.org/files/9113/9357/9173/WCDD_Message_EN.pdf
La Porte, T. R., & Consolini, P. (1991). Working in practice but not in theory: Theoretical challenges of “high-reliability organizations.” Journal of Public Administration Research and Theory, 1(1), 19–47.
Lekka, C. (2011). High reliability organisations: A review of the literature (Research Report 899). London: Health and Safety Executive.
Mitroff, I., & Featheringham, T. (1974). On systemic problem solving and the error of the third kind. Behavioral Science, 19(6), 383–393.
National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling. (2011). Deep water: The Gulf oil disaster and the future of offshore drilling. Report to the president. Washington, DC: US Government Printing Office.
Naveh, E., & Katz-Navon, T. (2015). A longitudinal study of an intervention to improve road safety climate: Climate as an organizational boundary spanner. Journal of Applied Psychology, 100(1), 216–226.
Perrow, C. (1984). Normal accidents: Living with high-risk technologies. New York: Basic Books.
Roberts, K. H. (1990a). Managing high reliability organizations. California Management Review, 32, 101–113.
Roberts, K. H. (1990b). Some characteristics of one type of high reliability organization. Organization Science, 1(2), 160–176.
Roberts, K. H. (Ed.). (1993a). New challenges to understanding organizations. New York: Macmillan.
Roberts, K. H. (1993b). Some aspects of organizational culture and strategies to manage them in reliability enhancing organizations. Journal of Managerial Issues, 5, 165–181.
Roberts, K. H. (2012). Reflections on high reliability research (Paper prepared for the Vanderbilt workshop). Nashville, TN: Vanderbilt University.
Roberts, K. H., & Bea, R. G. (2001). When systems fail. Organizational Dynamics, 29, 179–191.
Roberts, K. H., & Rousseau, D. M. (1989). Research in nearly failure-free, high-reliability organizations: “Having the bubble.” IEEE Transactions on Engineering Management, 36, 132–139.
Rochlin, G. I. (1993). Defining high reliability organizations in practice: A taxonomic prolegomenon. In K. H. Roberts (Ed.), New challenges to understanding organizations (pp. 11–32). New York: Macmillan.
Rochlin, G. I. (2012). Bullets over Vanderbilt (Paper prepared for the Vanderbilt workshop). Nashville, TN: Vanderbilt University.
Rochlin, G. I., La Porte, T. R., & Roberts, K. H. (1987). The self-designing high-reliability organization: Aircraft carrier flight operations at sea. Naval War College Review, 40, 76–90.
Schulman, P. (1993). The analysis of high reliability organizations: A comparative framework. In K. H. Roberts (Ed.), New challenges to understanding organizations (pp. 33–54). New York: Macmillan.
Smircich, L. (1983). Concepts of culture and organizational analysis. Administrative Science Quarterly, 28(3), 339–358.
Turner, B. A. (1978). Man-made disasters. London: Wykeham Science.
Weick, K. E. (1995). South Canyon revisited: Lessons from high reliability organizations. Wildfire, 4, 54–68.
Weick, K. E. (1996). Drop your tools: An allegory for organizational studies. Administrative Science Quarterly, 41(2), 301–313.
Weick, K. E. (2012). Continuities in HRO thought (Paper prepared for the Vanderbilt workshop). Nashville, TN: Vanderbilt University.