Ethics Review and the Minority Ethnographer

The IRB: A Brief History

The early 1970s to 1980 marked an era of rapid growth of regulatory bureaucracies in the United States (Dawson & Seater, 2013). As an extension of federal policy reform in the regulation of the use of human subjects in research, the IRB was defined and established in this era. Several infamous catalysts involving deliberate deception and harmful health consequences included persons of color and other vulnerable populations as human subjects, reflecting the racial features of many of these controversial studies (Beecher, 1966; Moreno, 2001; Reverby, 2011).

Since its official inception on July 27, 1981, the Department of Health and Human Services (DHHS) Policy for the Protection of Human Subjects and its 1970s precursors were met with robust criticism by social scientists (see, for instance, de Sola Pool, 1979). On the basis of the National Research Act of 1974, the scope of regulating research involving human subjects was extended from biomedical research to most types of social science. This expansion of scope and the implementation of review processes have prompted outcries of mission creep and overreach over the past few decades (Berrett, 2011; de Sola Pool, 1979; Gunsalus et al., 2007; Meyer & Rowan, 1977; Schwab, 2010; Zywicki, 2007), along with recognition of the extensive variability that exists in the operations of IRBs (Klitzman, 2015). Further, the research and regulatory communities recently engaged in formal public discussions on revising the federal code with regard to the efficiency and usefulness of particular forms of oversight (Federal Register, 2011). This Advanced Notice of Proposed Rulemaking (ANPRM) was titled “Enhancing Protections for Research Subjects and Reducing Burden, Delay, and Ambiguity for Investigators” and invited comments from researchers and other stakeholders on some 74 proposed changes across nine major areas of human subjects regulation. ¹

One of the most extensive formal commentaries in the recent human subjects rulemaking process came from the American Anthropological Association (AAA), which addressed a majority of the proposed changes in its 23-page report, making two overarching recommendations. The first one calls for limiting the scope of human subjects review to studies risking physical harm (i.e., biomedical trials and procedures) and those involving other types of human experimentation that control or limit study information (generally with placebos, deception, or misinformation). As it now stands, nearly all forms of research involving human participants are subject to review by an IRB (Code of Federal Regulations [CFR], 2009, 45 Part 46, Section 101). The second major recommendation of the AAA is the establishment of a separate IRB devoted to humanities and social science research at each institution (Dobrin & Lederman, 2011), which is already occurring at some universities.

The Modern Landscape of Human Subjects Protection

Several recent events in the regulatory arena were pivotal in the adoption of the current compliance-driven orientation of IRBs. In 1996, the General Accounting Office (GAO) restated the need for continued vigilance over human subject research, reinforcing that failure to observe 45 CFR Part 46, Subpart A (i.e., the Common Rule), was grounds for institutional ineligibility for most types of federal funding (GAO/Health and Human Services [HHS], 1996; Thomson et al., 2013). Then, in 1999, several high-profile injuries to subjects in biomedical research at Ivy League and other prestigious universities produced general awareness about human subject vulnerabilities and the need for increased oversight (American Association of University Professors [AAUP], 2000; Katz, 2006). This led to yet more reliance on the regulatory language of the standard clinical trial by the university IRBs, even for reviewing social science research (Bosk & De Vries, 2004).

As a result of these developments, the IRB is itself now prone to committing what one might view as Type I errors in its analyses of risks that protocols pose to research subjects (Zywicki, 2007). In prior eras, liberal reviews may have resulted in Type II errors in accepting the null hypothesis when it is false. That is, a case where the IRB believes that subjects are not likely to endure harm due to participation in the study when in fact they are at some significant level of risk of harm. Based on these mistakes, and much to the detriment of innovative qualitative field work, the contemporary IRB tends to err on the side of caution. It is now, therefore, more prone to committing Type I errors by concluding that subjects are likely to endure harm, when they are not at much risk of harm. The result is typically a set of revision “recommendations” to the study design that if not carried out will not allow for IRB approval.

Theoretical Framework

Weber (1930) identified four types of rationality in society and argued that the inherent dynamic of formal–procedural rationality or Western instrumentalism was to expand until humans were mere “cogs in a machine.” Embodied in our bureaucracies, formal–procedural rationality, through its expanding rules and regulations, emphasizes efficiency at the expense of substantive rationality, or the pursuit of ultimate values (in this case, the protection of human subjects). Weber’s metaphor for what we call today “red tape” was a “shell as hard as steel” in which humans became subject to or limited by the very bureaucratic organizations they have formed to enhance efficiency and effectiveness (Baehr, 2001).

We argue here that various problems inherent in IRB functions and processes are expressions of the incessant internal expansion of bureaucracy and its emphasis on control. Once rules and regulations are invoked to govern spheres of human behavior, the tendency is to recognize areas not covered by rules and policies and to bring more and more aspects of human activities within the scope of the regulatory paradigm (McLaughlin & Greene, 2014; Meyer & Rowan, 1977). Moreover, the siting of university IRBs in a relatively autonomous space that is not subject to administrative oversight creates the opportunity for power to become concentrated among very few administrative staff members and for abuses of power to occur.

This result, in some cases, combines with the hierarchical features of bureaucracy and its propensity for employing punitive measures against employees. Yet, for punitive measures to stem from this structural configuration raises concerns about power inequalities, namely, those germane to racial bias. In this context, IRB members and administrators, who are overwhelmingly White, become the local interpreters of relatively ambiguous federal codes regulating research with human subjects (Bankert & Amdur, 2006; Schwab, 2010).

A related framework that helps us analyze our case comes from the recent sociological literature on bureaucratic hybridization (e.g. Halpern 2008). Examining the conditions under which hybrid subentities emerge among university and privatized research communities, Murray (2010) evaluated whether they are best understood as “hostile,” “blended,” or “coexistence” models. Her case study (Harvard, DuPont Labs) concluded that such organizational hybrids are produced by differentiation, “maintained in productive tension rather than through easy coexistence” (p. 346). This perspective informs our study, first, where the ANPRM process and pending outcomes are concerned and, second, with regard to the manifestation of microinequities and invalidations.

First, given the numerous areas of IRB review under evaluation in the ANPRM, revisions have the potential to create a significant shift in this area of regulation. Ironically, if the recommendation of the AAA to create a separate IRB for social science research was implemented, it would mean still more bureaucratic differentiation and specialization within the IRB system. While some scholars might interpret this as the blended or coexistence model, it would have come from decades of increasing tension between qualitative/ethnographic researchers and the IRB (AAUP, 2000; Dobrin & Lederman, 2011; Katz, 2006; Shea, 2000).

Second, with regard to issues of modern forms of institutionalized racism, Sue et al. (2007) described microinequities as the process of having one’s contributions “overlooked, under-respected and undervalued” (p. 273) based on one’s race or gender. Chesler, Amanda, and James (2005) hold that within higher-education institutions, these issues are manifested in struggles to (de)legitimate and (in)validate the world views, cultural norms, and modes of investigation of minority scholars. In our case study, a minority ethnographer uniquely positioned to conduct innovative research due to his ethnic and social-class background, employed methods that did not neatly fit the template of the usual IRB proposal. When the approach was outside the typical protocol for the all-White, middle-class, and conservative IRB, and the investigator was not allowed to go forward, we contend that it was a case of racial invalidation. That is, the process by which the views and perspectives of racial minorities are rejected by those of the dominant group. When such a pattern emerges, it is likely to lead to some tension and pushback against the bureaucratic system in an effort to reform it.

The Case Study

The current case study chronicles the dealings of a pretenured ethnic minority researcher with his university’s IRB in seeking approval for federally funded fieldwork with immigrant gang members in the Midwest. The work would document and analyze the early stages of Latino gang formation at the research site using competing theoretical perspectives to address how and why these gangs emerged. Prior to receiving the funding award, the investigator, a second-year assistant professor at the time, was granted IRB permission to conduct exploratory fieldwork at the site. On the initial IRB proposal, this PI demonstrated his familiarity of the terrain and the expertise needed to protect the research subjects with whom he had previously built rapport and trust as a graduate student. When this study received federal funding and, presumably, the stakes were raised (see Moreno, 2001; Thomson et al., 2013; and Zywicki, 2007, for comments on this subtopic), the IRB placed the project under intense scrutiny. It called for drastic changes to most aspects of the research protocol, essentially limiting the traditional fieldwork methods that allowed for a successful pilot project.

The PI, a former street–gang intervention social worker repeatedly assessed the potential risks of subject participation in the study for re-review by the IRB. However, the board would not accept his characterization of the risks, believing that they were underestimated and incomplete. The PI knew these concerns were misplaced and grossly exaggerated by a board that lacked familiarity with the subject population and the fieldwork methods being employed. By this point, news of the project had spread among the gangs and broader Latino community at the research site, and it had built up much momentum. News of the grant award also created much excitement and anticipation among the volunteer field staff who now planned to assume paid positions within the project.

The PI was called upon by the board to educate its members on a host of issues relating to the project’s theoretical framework on gang migration, its research questions on the causes of Latino gang emergence, and logistical operations for data gathering. This occurred both in proposal revision correspondence and in person, and was extended, at the board’s request, to include information on Latino gang cultural norms and the causes of violent episodes between them. IRB members were concerned about the potential for police seizure of sensitive information on subjects although the PI had obtained an NIH-issued certificate of confidentiality to protect participants from such action.

The PI struggled to convince the board that local police lacked interest in seizing project data due to its highly theoretical orientation and lack of subject identifiers. There was no semblance of an expert in these matters on the board who could provide an informed review of the proposal. While federal regulations allow for the IRB, at its discretion, to invite individuals with competence in special areas to assist with a review (45 CFR 46.107), this particular IRB at a large, public university never indicated its willingness to exercise that option. This was peculiar, given that at least half of IRB chairs in the United States report using outside consultants on specific projects (Bell et al., 1998).

While there was sure to be some gulf between the ethical concerns of a group of middle class, mostly White IRB members and the actual threats to the well-being of Latino gang subjects, the position of the PI (the first author) was that the objections had risen to a level of absurdity. Similarly, in an NIH-sponsored study on the status of human subject protections, a PI commented that “[The IRB should] reduce the fanatical obsession with including every risk known to man…” (in conducting project protocol reviews; Bell et al., 1998, p. 62). In our case study, with each revision and review, the board’s “concerns” became so divorced from the plausible risks in the field that it became apparent to the PI that the board was simply intent on blocking the research effort, perhaps to avoid the liability of approving a funded study it unduly perceived as too risky (see AAUP, 2000; Heimer & Petty, 2010; and Schwab, 2010, for commentary on IRB aversion to legal risks). For example, there were concerns about rival gang members inadvertently crossing paths in the course of conducting private, one-on-one interviews with the PI, resulting in a violent confrontation. Any effort to describe the implausibility of this occurrence, coupled with the placing of careful protocols to avoid the possibility, was essentially ignored by the IRB. There were also concerns about the PI becoming the victim of aggravated robbery by subjects. To a qualitative gang researcher with ample experience dealing with this population (as both an insider and a professional), such concerns simply reflect the unfamiliarity with the street gang subculture of most middle-class Whites and the exaggerated fears that come along with it. Whatever its reason(s) for blocking the research, the review board’s staunch opposition to the PI’s work was an unfortunate irony for his funded research agenda and his impending tenure review.

IRB Reform and Career Damage

In 2013, the university’s IRB underwent an external review by a private firm specializing in this area of research regulation. Of the many findings and policy change recommendations relevant to the current case study, we highlight those that created the most problems for the PI prior to policy reforms. One was not asking investigators to reduce risks to themselves, only to subjects. The investigator was unable to convince the IRB that the gang subjects had no motive or interest in victimizing project staff. They would not accept the claim that subjects respected the PI and his staff, that participants hungered for a chance to tell their stories, and that their violent episodes occurred solely within the context of gang rivalry over turf, drug profits, and so on. As a result, safeguards dictated by the IRB were put in place that forced highly structured interactions with subjects, reducing trust and flexibility in the field.

A second point was related to assessing not only the magnitude of risk but also the probability of adverse events. Evaluators noted the IRB’s tendency to overstate the probability of harm. A third point related to imposing sanctions on investigators. The evaluators informed the IRB that this was outside its regulatory authority and that this is the role of institutional officials. Finally, it was noted that the IRB inconsistently applied regulatory criteria for approval of proposals. The members were said to apply their own rules and were not consistent across proposals, with each other, or with the federal regulations (Schrag, 2011, notes the commonality of this issue among IRBs).

These were only some of the numerous and substantial recommendations that resulted in a significant overhaul to the IRB system at this institution. Presumably, as a result of the review, the IRB administrative director who “presided” over the case in question resigned. In the end, the faculty senate’s recommendation to suspend the PI’s tenure clock was rejected, and his battle with the IRB, which lasted several years severely damaged his research productivity and tenure decision. The effects of racialized invalidation were quite robust and identifiable at institutional, ideological, and individual levels in this case. The broader racialized context of the university was also salient in this case. With the exception of one administrator, at the time of his tenure denial, the PI was the only U.S.-born Latino PhD in a college of some 45 tenure-stream faculty members. On this point, we make recommendations for the mentorship of untenured minority ethnographers below.

Discussion and Conclusion

This article discussed the expansion of the regulatory authority of university-based IRBs from biomedical, clinical research into the realm of social and behavioral research. One of the negative consequences of this expansion is an increase in the likelihood of causing damage to ethnographic social science research agendas, namely, for minority scholars conducting research classified by the IRB as high risk. It framed this expansion as the tendency of Western bureaucracies to subordinate value rationality to endless rules and regulations pursuing efficient control of more and more human activities. This perspective was synthesized with “racialized invalidation” concepts to illustrate how it results in a mechanism of contemporary institutional racism. In this case, it was shrouded in institution-level bureaucratic dynamics.

The case study showed that the increased power and autonomy enjoyed by today’s IRB come at a significant cost to researchers, at times getting into areas inappropriate to and beyond its scope and function (Bell et al., 1998), and arguably leading to abuses of its authority. One study noted that if the IRB’s suspension or termination powers were not used occasionally, it might signal that protections are inadequate (Bell et al., 1998, p. 64). In the current racialized context of higher education, minority scholars with unorthodox research agendas and field methods prove to be easy targets for administrators seeking to “flex their organizational muscles” in this regard. To protect themselves from these potential pitfalls, we recommend that pretenured, minority scholars aspiring to conduct high-risk ethnography take several precautions. First, one should not attempt to “hit the ground running” with the IRB. It is important for young scholars to gain credibility with their IRBs and advance slowly toward the projects that may be perceived to be high risk. If working alone, as do most ethnographers, they might want to begin by proposing a project that qualifies as low-risk and/or expedited review by the board. We do not mean to stifle ambition or boldness in these research agendas, but at a minimum, young crime ethnographers should seek mentorship for their full board review, high-risk proposals. If a suitable mentor for these projects does not exist within one’s institution (which is often the case with nuanced, avant-garde projects), they should seek the guidance of experts at other universities. It also helps to approach IRB administrators for guidance ahead of submitting a high-risk proposal.

We noted deficiencies of the IRB that are not well addressed in the literature on contentious regulatory issues for field researchers. While data for recent decades are lacking, IRBs appear to be inadequately diversified in terms of racial–ethnic composition. This reduces the board’s ability to reflect and incorporate community standards and provide meaningful review for certain types of proposals, especially those with which the members are relatively unfamiliar. For IRBs that resist ethnic or methodological diversification, yet which regularly review ethnographic research proposals to study ethnic minority communities, we suggest the increased use of outside experts in those fields. Whenever possible, expert preference should be given to members from the same social class and ethnic background as the minority groups being studied to reduce the lack of cultural familiarity that leads to irrelevant concerns regarding human subject protection or those based on dominant group stereotypes.

Other criminal justice scholars have recently noted that the new climate of powerful IRBs results in a stratification effect that is detrimental to field researchers at the rank of assistant professor Jacques & Wright, 2010). The current case study illustrated that for the pretenured ethnic minority scholar who works with vulnerable minority populations, and who did not have good mentorship, the struggle for credibility with the IRB was even greater and limited his chances for tenure by constraining the research agenda. This struggle was tantamount to what some have referred to as “racial invalidation” of the minority member’s ideas, goals, and approach toward research and scholarship.

While the current case study may not typify the experiences of most ethnic minority ethnographers, we sense that many in this small (but growing) community of scholars have also had negative experiences with their IRBs. To our knowledge, there are no available data to document this hunch, but this is an important issue worth investigating. Having scoured the literature on IRBs, we know that scientific efforts to interview a large sample of researchers about their experiences with the IRB are lacking and much needed. The ANPRM we discussed above is a good source of information that may lead to systematic changes, but its limitations are evident. An open comment opportunity is likely to see scholars who have had bad experiences with their IRBs self-select into the process, thus potentially biasing the data the government obtains.

In 2015, the DHHS provided notice of proposed rulemaking for improving human subject protections and “reducing burden, delay, and ambiguity for investigators” (Federal Register, 2015). The period for public comments ended on January 6, 2016, and final modification of the proposed changes will likely occur in early 2017. This will be the first large-scale revision of the regulations since human subjects protections were adopted as the Common Rule for federal agencies in 1981. While the regulatory scope is likely to be expanded to include all studies involving human subjects, some proposed changes may prove helpful for researchers.

Still several stages from finalization, many of the proposed changes respond in part to a substantial history of complaints from social scientists and some in the humanities fields who say the burdensome process can slow their research to a crawl even though it typically poses little threat to participants (Dobrin & Lederman, 2011; Lederman, 2011). Although the proposed changes do not address the issues of racial diversity on the composition of IRBs, this reform may provide some relief to qualitative and other researchers who struggle with the IRB process because their work is atypical, not well understood, or unduly perceived as too risky under the current risk-benefit framework.

Our case study showed that one such research agenda was devalued through ill-conceived institutional processes, resulting in the racial invalidation of a pretenured faculty member’s fieldwork, ultimately muting scientific inquiry and career goals. In order to avoid such outcomes in the future, the issues of relative autonomy, absence of appeal processes, and diversity must be addressed in constructive ways. Finally, and of critical importance, IRBs must resist facile solutions to the structurally induced dilemmas arising from its responsibilities, and it must resist usurping the functions of risk management relative to institutional liability and focus on protecting human subjects.