My interest in “Action Research” arose not from my experience as an educational researcher, a teacher in the school system, or a principal or superintendent. Instead, my interest in Action Research was tweaked by my experience as an administrator of three primarily social science institutional review boards (IRBs)Footnote 1 to which researchers presented research protocols that purported to be minimal risk studies of teacher practice where the “teacher–researcher” was the “research subject.” Recent discussions in Canada and the United States have exposed many of the problems that educational, social, and behavioral science researchers encounter as their research methodologies are reviewed by and in the oversight mandate of the IRBs (Pritchard, 2002; Sieber & Baluyot, 1992; Sieber, 1992).Footnote 2 These discussions focus on the divergence between the IRB’s role and assumed bio-clinical predisposition and the ability of behavioral and social science researchers to have their research methodologies and research understood and appreciated by IRB members.

In this article, I explore some of the dilemmas confronting IRB members and administrators in the review and administration of the action research protocols, particularly those that involve vulnerable populations (e.g., children) and which, from the practitioner–researcher’s perspective, focus on the practitioner–researcher as the object of the research. This paper continues a dialogue between the action researcher community and the IRBs.

Action research protocols present IRB members with issues that the practitioner–researchers often do not directly address. These issues include, but are not limited to, power relationships between the practitioner–research and their students, the conflict of interest position in which practitioner–researchers are placed when conducting research in their classrooms, and the tensions of obtaining (and maintaining) meaningful informed consent/assent throughout the research processes. University faculty researchers and graduate students often are caught between two sets of ethics reviews: one imposed by the institution at which they are faculty or studying and the second by the school authority that governs access to the research site. The requirements for two ethical review processes may be quite divergent. My view is that the university has the responsibility for the ethical conduct of its researchers, no matter where the research is conducted, and is obliged to review all research involving human subjects conducted by faculty and students.

School boards also have regulations and guidelines for the conduct of research with their students, teachers and administrative staff and legitimately require internal scrutiny of the proposed research.Footnote 3 Thus, teacher–researchers, who may desire to conduct action research projects as part of their professional life as well as an educational program leading to a higher degree find themselves subject to two, consecutive review processes. It is here that key problems arise: teacher–researchers often do not distinguish between their complementary but sometimes conflicting roles – as teacher, administrator, and researcher in the context of their role as professionals within the school system and as student–researcher conducting research as part of a degree program.

Literature Review

The literature on action research focuses on (1) the definition or concept of action research, (2) the underlying philosophical concerns of theorists and practitioners, (3) the methods by which individuals or teams of practitioner–researchers undertake their research programs, (4) the personal as well as professional growth that emerges from engaging in action research, and (5) the practical outcomes – how action research projects have improved learning environments for students. Few address specifically issues of the ethics review of the research. When scholars or practitioners address issues of the ethical conduct of research with human subjects, much of the discussion, legitimately, highlights obtaining consent and maintaining confidentiality. Yet much of the discussion is superficial and does not address underlying ethical issues – why free, informed and continuing consent and assent is required prior to the initiation of and during the research. As a result, it is not unusual that individual researchers are confused, hurt, and antagonistic when IRBs ask them to present their research for review or for modifications to the protocol.

Defining action research will, to some extent, identify some of the issues that create problems for IRBs. Without dwelling on definitions of Action Research not exhaustively reporting on the literature, I have drawn on two definitions. First, Arhar, Holly and Kasten (2001) propose:

Action Research (AR) is a process of theorizing and testing our own, as well as other people’s, ideas and theories in practice. It is grounded in ideas about how knowledge is generated (AR is the process of developing theories by practitioners who test those theories against their own values) and in the relationship between the cnr and what is known (the research and what is being studied are connected in an interactive way – the values of the researcher influence what is learned.)

Second, Kemmis and McTaggart (1988) defined action research as,

[a] form of collective, self-reflective inquiry that participants in social situations undertake to improve: (1) the rationality and justice of their own social or educational practices; (2) the participants’ understanding of how these practices and the situations in which they carry out these practices. Groups of participants can be teachers, students, parents, workplace colleagues ..., that is any group which a shared concern and the motivation and will to address their shared concern. (cited in Altrichter et al., 2002: 125).

Altrichter et al. (1990) further note the axiomatic elements of this and other definitions: “action research is about people reflecting upon and improving their own practice, by tightly inter-linking their reflection and action, and making their experiences public to other people concerned by and interested in the respective practice” (p. 128). For my purposes, the key elements of defining action research are: the researcher intends the proposed research to be self-reflective, where he or she is also the participant, and whereby the focus of the research is the researcher–participant’s professional practice, and the research occurs within a professional/public setting. Thus, within the context of the philosophy of action research, it is not surprising that researchers, especially teacher–researchers, are taken aback when an IRB questions their research protocols. Yet, the methods by which data are collected are those that typically engage others as participants, including, focus groups with colleagues and key informants, surveys of parents, questionnaires to parents and/or students, journals by students, interviews of students (present or past), photographs of classroom activities, or video-taping of classroom practices. Since most practitioner–researchers learn about action research through professional development opportunities or through their graduate research programs, it is useful to assess what textbooks and manuals state about research ethics and compare these to the requirements of international standards for ethical review of research (e.g., 45 CFR 46 or the Common Rule, and MRC et al., 1998).

Brown (1998), in his review of narrative inquiry, asserts that “research ethics requires that participants be protected at every stage, that no harm shall come to them as a result of participation in the research.” McMillan and Schumacher (1993) argue that researchers need to balance the costs of using questionable methods against the benefits of conducting a study. They note that in educational, social, and behavioral science research “costs include injury or psychological difficulties, such as anxiety, shame, loss of self-esteem, and affronts to human dignity, or they may involve legal infringement on human rights.” These costs/harms are to be balanced or weighed against potential benefits, including “gains for the research participants like increased self-understanding, satisfaction in helping, and knowledge of research methods, as well as more (or less) obvious benefits to theory and knowledge of human behavior (p. 182).” McMillan and Schumacher then summarize the ethics of research, employing American Psychology Association (APA) ethical principles as the basis of their discussion (McMillan & Schumacher, 1993: 182–5). Borg and Gall (1989) address ethical issues in research (1989: 84–93), focusing in greater detail on issues of confidentiality (p.87–89) and participant observation (p. 391–397), ethnographic interviews (p. 397–399), case studies (p. 402–404), and data collection strategies employed in action research.Footnote 4 Since qualitative researchers tend to view “reality as socially constructed ... [where] the researcher interacts with participants in order to understand their social constructions,” qualitative researchers, including action researchers, “must continually ask what it means to be ethical within their research role” (1989: 109–110). Citing the Council of American Anthropological Association’s ethical principles, “where there is a conflict of interest, [subjects or participants] must come first,” Glesne and Peshkin remind qualitative researchers that their relationships with the research participants are asymmetrical, “with power disproportionately on the side of the researcher” (Glesne & Peshkin, 1992: 111, 117). This is the case in teacher-as-researcher environments.

With specific reference to action research, Arhar, Holly and Kasten (2001) note that since action research

involves a wider audience than does traditional teaching[,] ... ethical dilemmas ... arise ... particularly around procedural issues (e.g., how, when and from whom to gain permission); legal and institutional issues (e.g., compliance ..., turning over documents such as journals to legal authorities in cases of child abuse and neglect, etc.); relational issues (e.g., building trust, using data to build up the community as opposed to using data only to further one’s career); and role confusion (e.g., ... multiple responsibilities of action researchers to research, to students, and to parents). (Arhar, Holly & Kasten, 2001: 169).

For many educational researchers, especially teacher–researchers, who are often graduate students in faculties of education, McNiff’s definition and description of action research is most attractive:

In action research, researchers do research on themselves. Empirical researchers enquire into other people’s lives. Action researchers enquire into their own. Action research is an enquiry conducted by the self into the self. You, a practitioner, think about your own life and work, and this involves you asking yourself why you do the things that you do, and why you are the way that you are. When you produce your research report, it shows how you have carried out a systematic investigation into your own behavior, and the reasons for that behavior. The report shows the process you have gone through in order to achieve a better understanding of yourself, so that you can continue developing yourself and your work. (McNiff, 2002).

While action researchers argue that students, colleagues and others are “collaborators” in their research, not research subjects, Arhat et al., drawing on Zeni (1998), identify the multiple roles of the teacher–researcher (e.g., power and trust relationships with students) as problematic, but not impossible to overcome by using appropriate processes to obtain informed consent and voluntary participation. Their ethics guidelines for teacher action researchers are useful, if limited in the discussion of obligations to student participants.

Arhar, Holly and Kasten (2001) suggest that there may be limits to the voluntary participation of students in the research process. Acknowledging that “students need to know that they may withdraw from the study at any time without fear of negative consequences,” this right is constrained by the claim that “if the study includes teaching activities and curriculum that are part of the regular teaching process, withdrawing is not an issue,” that is, students cannot withdraw. From an ethical perspective, this raises significant, but not unresolvable, issues.

Canadian and US Regulations and Guidelines

In Canada and the United States, the ethical review of research involving humans is governed, respectively, by the Tri-Council Policy Statement (TCPS) on the Ethical Conduct for Research Involving Humans or 45 CFR 46, known as the Common Rule. Both sets of ethical regulations have been constructed with the underpinnings of the Declaration(s) of Helsinki. The “guiding principles” of TCPS, for example, are

  • Respect for Human Dignity

  • Respect for Free and Informed Consent

  • Respect for Privacy and Confidentiality

  • Balancing Harms and BenefitsFootnote 5

  • Respect for Vulnerable Persons

  • Respect for Justice and Inclusiveness.

These principles encompass the three principles of the Belmont Report, the foundations for 45 CFR 46 (DHHS Policy for Protection of Human Research Subjects) and 21 CFR 50 (Informed Consent).

The Ethical Dilemmas

The primary purpose of the IRB review is the protection of the research subjects by ensuring that their participation is informed and voluntary, that they give consent freely given and are not coerced, and that the benefits/harms balance is appropriate to the research and to the subjects. The ethics review of the research (and the role of the IRB) is to the benefit of the subjects, not the researcher (although it also does the latter).

By taking a “subject-centred” focus, the researcher and the IRB will view not only the research processes but also the research ethics review differently. They will view research, as action research purports to do, as a mechanism for improving the educational process, the knowledge on which pedagogy is based, and thereby the learning experience in which students find themselves. Both the researcher and the IRB member will also ensure that the research will not harm unintentionally the research participants – teacher–researchers, students, colleagues, and parents.

Some researchers in the social sciences view the ethics review process as a burden or barrier to surmount. Some view the ethics review process as one that does not benefit the research participants but instead hinders the research and the researcher. Few view the ethics review process as the equivalent of peer review, with a focus on ethical issues related to the research and its impact on human participants rather than the scholarly or scientific nature of the study. To assess this perception and the ethical issues identified by IRBs, I conducted a study of research protocols submitted to an IRB at a Canadian university to identify those areas around which disagreement occurs and to identify ways in which the IRB and the researcher resolved any ethical or methodological conflicts.

Data Collection and Analysis

In my review of more than 100 action research protocols submitted to the Brock University Research Ethics Board (IRB) over three academic years 1999–2002, the concerns identified by the IRB were: data collection, voluntary participation, informed consent, and confidentiality of participants’ data and images during and after data collection. The researchers collected data employing a wide range of methods (videotaping classroom teaching practices, collection and analysis of student-developed journals, interviews with students and colleagues, focus groups with colleagues, and questionnaires administered to students, parents, colleagues, administrators). The practitioner–researchers included consent forms for parents (usually) and assent forms for students (infrequently). Some consent forms identified the purpose of the research, any time commitment of the student or parent that was over regular classroom requirements, and any measures that protected the students.

Case Studies

Two examples, each with a slightly different focus, highlight some of the common problems encountered by the IRB and action researchers.

First Case Study

In this case study, a teacher proposed to assess the effectiveness of her strategies for teaching and improving writing skills to attempt to effect an improvement in student writing skills, and evidence of the degree of influence that these strategies among her school students. To carry out the research, the teacher proposed to collect data including: her own written records of classroom teaching; samples of student writing at various stages during her research (to demonstrate changes over time); questionnaires completed by students about their writing skills and the ways in which she assisted them; video and audiotapes records of her teaching; and transcripts of interviews with volunteer students regarding their writing skills.

While it is probable that the teacher would have collected many of these data points as part of the regular teaching and evaluation processes, the IRB determined that the proposed study posed a number of ethical issues and that the information letter to parents and consent forms did not adequately address potential risks/harms to students. Specifically, the IRB reviewers were concerned about those elements that related to the students (and parents) understanding the nature of the study (i.e., separating out the study elements from the day-to-day pedagogical processes), the understanding of free and informed consent, without implied coercion (i.e., student’s and parent’s agreeing to participation while fearing that non-agreement might have some unforeseen negative impact on the student’s interactions with the teacher and/or on the student’s assessment); and issues of confidentiality of the student records and other data. The IRB also requested that the teacher–researcher address (or at least acknowledge) any conflict of interest or power imbalance that might exist in the research study. The IRB was particularly concerned about proposed video- and audio-taping of classroom behavior, which, while well-justified (to capture the actual test of the interaction and nuances such as non-verbal cuing, tone-of-voice, and pauses) held out the possibility for abuse. Students and others become rapidly accustomed to the presence of a video-recorder in a classroom and may resume regular patterns of interaction, including some behaviors that might be viewed as disruptive. There were no safeguards put in place to protect such students if these behaviors went unrecognized during the instructional periods but were revealed during the researcher–teacher’s review of the tapes for research purposes. How would the researcher–teacher propose to handle such revelations?

There was also an implied conflict-of-interest that required clarification. The IRB questioned the position that “there is no conflict of interest in my being both researcher and classroom teacher.” The potential conflict (and potential coercion) arose if an insufficient number of students/parents consented to participate (the optimal “N” had not been identified in the original protocol.) The conflict was, of course, that if insufficient numbers of students/parents agreed to participate, then the researcher–teacher would (1) need to find another population to study, (2) change the study, and (3) delay her progress through the program and potentially professional progress (it was not clear if there completion of the degree was required for salary advancement.)

Not surprisingly, the IRB requested clarifications. Specifically, the IRB requested the teacher–researcher to clarify how she would recruit participants so that the perception of coercion or conflict of interest would be avoided or mitigated, that issues of confidentiality were addressed and that assurances of anonymity were softened. While the IRB did not question the issue of storage of data specifically, some issues have been raised is other venues about the storage of data (audio-tapes, video-tapes, student work, etc.) in the researcher–teacher’s home office for a period of 3 years to allow for re-analysis and possibly publication of the findings.

Responding to the concerns of the IRB, this researcher–teacher modified the information letter and consent forms to demonstrate that there might be other forms of recruitment, beyond her classes. Significantly, and perhaps one might argue more appropriately, she proposed two modes of recruitment: (1) volunteers from her classes and (2) volunteers who would sign up for special noon-hour workshops that would be open to a broader student base. In addition to mitigating the issue of undue inducement or coercion, the combination of strategies broadened the range of participants, reduced potential harms to non-participating students, and provided a “control” group, which is not common in action-research strategies. If the researcher–teacher had insisted on using only her students, the IRB might have required the withholding of the names of the volunteers from the researcher–teacher until after her interaction with the class had ended (e.g., end of term.) The names of individuals who had agreed to participate were not made available to others.

Second Case Study

Other teacher–researchers took a more active approach in their responses to the IRB. A number of the researcher–teachers were or planned to become school administrators and were undertaking studies designed to assess and improve their practice as school administrators. Employing similar data collection methodologies (e.g., journal-keeping, voice and video-recordings, existing documentation and data such as staff meeting minutes, email correspondence, ‘interviews’ with staff members, interviews and meetings with students, discussion with parents or parent council members), these teacher/administrator–researchers were affronted when asked to clarify how their position of authority might influence whether staff members or students agreed to participate in their research study. Consent forms and information letters also contained information assuring confidentiality, anonymity, and voluntary participation.

The IRB encouraged the researcher to modify consent forms to briefly outline various strategies for data collection, to include video/audio-taping and photography and why this was required as a form a data collection as well as a separate permission or release form to use videos or photographs, and to address the serious power imbalance that existed between the researcher and the informants. In the IRB’s view, almost all of the prospective participants were, “in some way,” subordinates, thereby “raising a strong possibility of the perception of coercion.” The IRB asked how the research would “reduce the risk that prospective participants will feel coerced into participation (add no bearing on performance evaluations, student records, etc.).” The IRB also required the researcher to inform participants in focus groups or other forms of group discussion that there would be limits on anonymity and confidentiality.

The researcher expressed a concern that the IRB was questioning his professional ethics and referenced two sets of professional codes of ethics for Ontario teachers to which he adhered. Moreover, as with others who wished to access schools to conduct their research, this researcher highlighted the need to obtain approval quickly in order to complete the data collection prior to the end of the school term.

Perhaps to mollify the IRB, the researcher chose not to include forms of data collections and modified the consent forms to adhere, generally, to the direction of the IRB.

Lessons Learned

Canadian IRBs adhered to the TCPS guidelines with regard to the ethical review of research, employing a proportionate approach. That is, IRBs reviewed each proposal on the principle that “the more invasive the research, the greater should be the care in assessing the research (1.7).” Taking a “subject-centred” perspective, IRBs asked what were the potential benefits and what were the risks/harms. Recognizing that the participants, as well as other children in the schools in which the teacher–researchers worked, would potentially benefit from the research through enhanced practice of the teacher–researchers, IRBs did, appropriately, consider the risks/harms. These risks, as with most social science research, are quite different from those presented by medical or clinical research. The primary harms to individuals in action research fall into the categories of loss of privacy, embarrassment, emotional distress or psychological trauma, and loss of privilege. Power relationships between the researchers and subjects were concerns of the IRB, leading to the argument that students, parents and other staff members might not view their proposed involvement to be entirely voluntary and without fear of loss of privilege.

There appears to be a divide between the objectives of the ethics review process (“the protection of the research subjects”) and the interests of the researcher (improve her/his professional practice and enhance the quality of instruction and the learning environment in the classroom). Drawing on the published thesis of a cohort of action research–teachers, we learn that they were focused on their research and their objective: to reflect on their professional practice as a means to improving the education of students. These teacher–researchers believed that IRB did not understand the nature of action research and was driven by a positivist, non-qualitative approach to research. One teacher–researcher noted: “action research is not a ‘neat and tidy’ process (Knill-Griesser, 2001, Chapter 4: 2).” Others felt that the IRB questioned their professional and personal ethics (Suderman-Gladwell, 2001) and argued that “action research is, by its very nature an ethical process (Black, 2001, Chapter 23: 2) In her published thesis, a cohort teacher–researcher asserted: “as a member of this [school] board, I have the permission of the school board and the blessing of the [Ontario] College of Teachers to engage in this type of research. Both organizations encourage teachers to reflect on their own practice, thus taking responsibility for their own professional development (Black, 2001, Chapter 3: 4).” Yet, IRBs do not question the personal or professional ethics of researchers, but instead expect that researchers adhere to professional codes of ethics, whether they are anthropologists, physicians, engineers or teachers.

We can surmount this divide between the action researchers and the IRBs. The American Association of University Professors’ American Association of University Professors (2001) report “Protecting Human Beings: IRBs and Social Science Research” underscored “an inescapable interdependence among university administrators, members of IRBs, scholars, and students, but explaining the relationship calls for better communication among these components.” The AAUP recommended that administrators “help” social science researchers “through campus-based seminars ... to which would be invited past and current IRB members, social science researchers who have gone through an IRB review, and researchers likely to face one.” Primarily, the AAUP encouraged the demystification of the ethics review process and making the process by which decisions are made transparent. IRBs need to become familiar with the methodologies that researchers utilize and the traditions of research in those divergent disciplines. While understanding disciplinary traditions, IRBs need also to focus on their primary purpose, the rights of the subjects or participants. Hence, IRBs have an educational mission that needs to be fulfilled. That is, IRB chairs and administrators need to devote considerable time to active education of researchers in all disciplines about their purpose and the reasons why they seek to protect the interests of the subjects. While some faculty and students suggest that the primary purpose of the IRB is to protect the institution from liability, IRBs instead need to demonstrate that the hierarchy of interests falls as follows: (1) subjects/participants, (2) parents/guardians (those responsible for the participants), (3) the community/society, (4) the researcher, and then finally (5) the research institution.

Researchers do have rights to be able to conduct research that challenges the frontiers of knowledge and social or pedagogical practices. They also have obligations to respect the dignity of the subject/participants and to recognize that the research that they do undertake potentially poses risks to their subjects. Faculty and student researchers must recognize that while schools and professional bodies encourage self-reflective research, once teacher–researchers become part of a university system, research for the purpose of a degree is not what they do in their day-to-day professional life. The systematic collection and analysis of data that leads to generalizable knowledge in the form of a thesis, project report, and contribution to the professional body of knowledge and practice is additional to that which they do through the self-reflective processes of assessing one’s practice.

Most action research projects reviewed required clarifications on the things that we do prior to data collection (i.e., identifying a data collection strategy, justifying the numbers and sample population, ensuring that consent and assent is truly voluntary and informed); and clarifications of what occurred after the collection of data, with particular reference to (1) lingering terms of consent and assent; and (2) confidentiality of data/records. Since subjects/participants were usually minors – children in schools – and thus vulnerable and ‘captive’ populations, the burden of care is on the researcher and the IRB to ensure that these individuals are not subjected, even inadvertently, to harm as a result of the data collection and analyses processes.

Implications for Researchers and Research Ethics Board

The AAUP (2000) report underscored “an inescapable interdependence among university administrators, members of IRBs, scholars, and students, but explaining the relationship calls for better communication among these components.” The AAUP then went on to recommend that administrators “help” social science researchers “through campus-based seminars, symposia and the like to which would be invited past and current IRB members, social science researchers who have gone through an IRB review, and researchers likely to face one.” Primarily, the AAUP encouraged the demystification of the ethics review process and making the process by which decisions are made transparent. Other recent articles and reports suggest that the ethics review process is exceptionally burdensome for social and behavioral science researchers, often using a bioclinical model and an overly conservative concern about potential harms to individuals and collectivities.

Instead, I argue that the research ethics or institutional review boards do not adequately educate those researchers about the potential harms that might arise from research that purports to be benign and poses minimal risks to the participants. Research ethics boards need to devote considerably more effort in working with researchers to demonstrate that harm arising from social and behavioral science research is, as recent the Office of Human Research Protections’ (OHRP) presentations suggest, real. These potential harms may be social, psychological, and economic rather than physical as in the case of bioclinical research, but they are real nonetheless. Moreover, to ensure that the public’s support of all research is not jeopardized further, REBs/IRBs and researchers need to assess potential harms and benefits of research activities and ensure that research participants – children, students, and professional colleagues – participate freely, voluntarily, with an understanding of the research and its potential benefits to advancing professional practice.

REBs/IRBs need, as well, to ensure that the research methodologies of the behavioral and social sciences are well represented on committees that assess these protocols. It is as unreasonable to expect an IRB that is primarily bioclinical in its orientations to assess research protocols emerging from educational practice as it is for a IRB composed primarily of social scientists to assess the risks and benefits of a clinical protocol. Moreover, IRBs ought not be overly conservative in their assessment of risks and harms involved in educational and social science research. Hence, I would argue, there is a requirement for mutual education between IRBs and social and behavioral science researchers. From my experience as a researcher and research administrator, these are a small number ways in which action research can be facilitated while adhering to strict research ethics standards, guidelines for the ethics review and ethical conduct of research employing action research methodologies.

This article is a plea for improving the communication between researchers and REBs/IRBs, to removing barriers of understanding between the two, and to ensuring that the focus of each is on the interests of those who are the research subjects. Only through enhanced communications can these hurdles be cleared and high quality research be supported.