Action research (AR) is, at once, both a research methodology or approach to conducting research and a means of effecting social change (Chenail, St. George, & Wulff, 2007). It is a very different approach to conducting research that has been used and is currently being used by multiple disciplines to effect change in healthcare practice arenas and social venues. However, while many disciplines use its tenets, they all have different definitions of what action research is exactly.
Chenail, St. George, and Wulff (2007) state that AR is not a singular method to conduct research, but rather a family of similar approaches. Minkler and Wallerstein (2003) claim that AR is not a methodology in itself, but an orientation to research. Stringer and Genat (2004) define AR as a "systematic, participatory approach to inquiry that enables people to extend their understanding of problems or issues and to formulate actions directed towards the resolution of those problems or issues" (p. 4). Ziegler's definition of a systematic inquiry done by practitioner-researchers who reflect and act on problems encountered in their practice is used by Freeman (2007).
Different disciplines also used different terms for AR. Action research first obtained popularity when it was used and described by Kurt Lewin, a sociologist who conducted action research studies in the 1940s (Stringer & Genat, 2004; Wallerstein & Duran, 2003). Lewin viewed AR as cyclical, dynamic, and collaborative process in which people addressed social issues that affected their lives and used AR to address discrimination ,segregation, and assimilation problems in society (Stringer & Genat, 2004; Wallerstein & Duran, 2003). Since then, education has used it, calling it classroom action research, critical action research, and practitioner research; organization psychology calls it action learning, action science, or industrial action research; psychology and human relations termed it cooperative, mutual, or reflective practitioner inquiry; evaluation research uses constructivist or fourth-generation inquiry; nursing has called it emancipatory inquiry; and public health coined it popular epidemiology (Wallerstein & Duran, 2003). Regardless of the name, action research is designed to bring about change and to bridge the gap between theory and practice.
Four commonly used approaches in this family are action research, participatory action research (PAR), community-based participatory research (CBPR), and appreciative inquiry (AI). While all of these approaches have similarities and differences, as discussed below, they are all intended to identify a problem and develop solutions to that problem (Chenail, et al., 2007). AR usually has an outsider assisting the internal group of stakeholders conduct the research and uses a praxis to develop solutions to problems that don't threaten the existing structures of the organization or community (Chenail, et al., 2007). PAR typically obtains its leadership from within the oppressed group and strives to solve a problem of inequity or unfairness by changing the oppressive structure or rules that govern them (Chenail, et al., 2007). Participatory in both PAR and CBPR means that the "subjects" or participants actually become the researchers and design and conduct the study, then implement the changes (Chenail, et al., 2007). Often these participants are called stakeholders, because they have a stake in both the situation and the outcome. CBPR's characteristics have been identified in Minkler and Wallerstein (2003) as: participatory and cooperative; both the participants and researcher have equal parts to play within the study; both participants and researchers co-learn from each other and the process; it involves systems development and community building; empowering process that allows participants to increase control over their lives; and a process that achieves a balance between research and change. AI, on the other hand, typically uses an outside consultant to help design and conduct the research, but strives to create change by initially focusing on what the organization does right, the processes that are working and running smoothly, in order to help build momentum and confidence in the participants' ability to make change (Chenail, et al., 2007).
The process of conducting AR is quite similar to other research methodologies with the exception of the final step - taking action to resolve the issue (Stringer & Genat, 2004). First, the study is designed. This step involves clearly identifying the issue, defining the research question, planning the data collection methods and getting appropriate approvals, checking the researcher's ethics, and planning for appropriate validity of the study (Stringer & Genat, 2004). Next, the participants actually collect the data; a variety of sources can and should be used, from participant interviews, participant observation and researcher field notes, focus group sessions and findings, records including documents, materials, equipment, photographs, surveys, and literature reviews (Stringer & Genat, 2004). Once the data is thoroughly collected, then it is analyzed to define the key features of the issue (Stringer & Genat, 2004). Stringer and Genat (2004) state that the next step is to communicate the findings to relevant stakeholders; this could be anyone from the nurses on a particular unit to an entire community. The last step is the part that sets AR apart: taking action to resolve the issue.
There are two models that illustrate the AR process. Figure 1 (see Appendix A) demonstrates the process of AR in a linear fashion, while Figure 2 (see Appendix B) describes AR in a more natural, cyclical or helix format. AR should be iterative in nature; many studies go through four or more cycles of data collection, analysis, and action before the issue is fully defined and solved (Stringer & Genat, 2004). In the helical view, researchers first look at the problem to clarify its nature, then think or reflect on its significance and who is affected by the problem, and thirdly act by deciding what to do and then performing the change (Stringer & Genat, 2004). The process begins again when the researchers look at what occurred when they took action, think about those actions and evaluate their effectiveness, and then again act by refining or enlarging the previous action (Stringer & Genat, 2004). The process keeps repeating until a reasonable and effective solution to the problem is defined and implemented.
Another difference between AR and other research methodologies lies in its focus; the goal is not merely increasing the body of knowledge of the discipline or increasing researcher knowledge but on change and improving the practice and behavior of the participants (Stringer & Genat, 2004). Stringer and Genat (2004) list other goals of AR and what they involve: reflection, participation, inclusion, sharing, repetition, practice, and community. Reflection involves people thinking, reflecting, and/or theorizing about their own practices, behaviors, and situations. T he participants have to use participation to change their own practice and behavior and not focus on changing others' practice or behavior. Inclusion starts with querying and starting the change process with those who are the least powerful, later widening the circle to include all of those affected by the problem. People sharing their perspectives with others, while a simple goal, often leads to community building. Understanding involves achieving clarity of understanding the different perspectives and experiences of all those involved in the process. Repetition is the repeating of research cycles until the solution is found. Practice is changed by testing the new understandings by using them as the basis for changing practice or constructing new practice. Finally, communities are developed or built as the end product of the AR process.
The people conducting AR often have multiple roles; these roles can change over time during the AR process (Stringer & Genat, 2004; Wallerstein & Duran, 2003). The academic researcher, who typically holds all the power in the research process, becomes more of a consultant or facilitator as they assist the participants in planning, designing, and conducting the research; participants who typically only provide the data also become researchers as they assist in the design, planning, and data collection (Stringer & Genat, 2004; Wallerstein & Duran, 2003). Both the researcher and the participants are considered stakeholders who are affected by the problem or situation at hand. The issue of power is important in the AR process. Typically, the academic has held all of the power, by having access to funding, knowledge, resources, time, and research assistants to assist them (Wallerstein & Duran, 2003). This power play can inhibit openness among the participants, affect trust, and could actually change the results. It's very important for the researcher to become the equal of the participants in order to avoid this power struggle. The involvement of participants can best be summarized by a statement from Habermas, as cited in Wallerstein and Duran (2003) "in the process of enlightenment, there can only be participants" (p. 32).
Action research can draw its philosophical base from several different theories or perspectives. It is of the naturalistic inquiry, which focuses on understanding the meaning and interpretation of everyday things. The experience and meaning of the phenomena is what's important in the research, not the ability to generalize the findings to the population in question (Stringer & Genat, 2004). Much of AR leans upon critical theory, critical social theory, feminist theory, emancipatory theory, poststructuralism, and/or postcolonialism (Wallerstein & Duran, 2003). All of these theories or philosophies aim to analyze personal lives in relation to the structures, both open and hidden, that victimize and control the power relations of the individual in question (Wallerstein & Duran, 2003).
Chenail, St. George, and Wulff (2007) discuss the philosophical traditions of different approaches of AR. Action research in itself draws from pragmatism and consensus social theory, while participatory action research is more about the power, drawing from critical social theory, conflict theory, and Marxism; appreciative inquiry has a base on social constructionism. Again, looking at the power relations and how to change who has the power is the underlying philosophy behind all forms of AR.
Freeman (2007), using Ziegler's definition mentioned earlier, states that assumptions of AR are as follows: 1) practice is the foundation for action research; 2) researchers are partners with the participants; and 3) action research necessitates the application of a system to inquiry. Stringer and Genat (2004) state that all action research must be participatory, democratic in nature, empowering, and life-enhancing. Participants must have ownership of the problem and the research; they have to take an active role in both the design and planning and the actual conduction of the research study (Stringer & Genat, 2004). The researcher and participant are equal in this democratic process. While the researcher may have the expert knowledge and access to resources, the participants have the expert knowledge of the problem and how it affects their lives; they have the experience of the problem (Stringer & Genat, 2004). The participants are also the ones who have to incorporate the solution into their lives; therefore they have to own the solution as well for AR to work and change the situation. The participants in AR have typically not held the power; they have not had the ability and freedom to change their perspectives and their situation (Stringer & Genat, 2004). As the AR cycle begins to effect change, participants see and understand how powerful their voice really is. This empowerment is paid forward, eventually building learning communities that can enhance the lives not only of the individuals, but of the agencies, organizations, and institutions surrounding the individuals (Stringer & Genat, 2004).
Appropriate situations for use
The action research family of approaches can be used in any situation, whether clinical or community-based, where a problem is found that needs a solution. Change and effecting social change is the main impetus behind AR; therefore it can be used anywhere a problem or social injustice exists. Action research can be an effective methodology to use to help reduce or eliminate health disparities. Giving the oppressed, disparaged, powerless people a voice and showing them how much power they have collectively can empower them. Small changes in policy or community norms can be started as a result of AR.
Action research studies have flourished in the past ten years in nursing. Sample research questions that have been asked and researched range from how a nurse practitioner can impact behavior of a person after myocardial infarction related to the medical regimen (Freeman, 2007) to how to involve homeless persons in the development of a consumer advisory board dedicated to their health (Buck, Rochon, Davidson, & McCurdy, 2004). Depression in Latinas in an emerging Latino immigrant community was examined in one CBPR study (Shattell, et al., 2009). Australians have used AR in the mental health realm to develop a framework to decrease the use of seclusion in an inpatient psychiatric unit (Hyde, Fulbrook, Fenton, & Kilshaw, 2009) and to develop policies regarding the safe transport and delivery of acute psychiatric patients in rural remote settings (Petrie, Schultz, & Pearson, 2009). One particular AR study involves Hmong women and diabetes (Yang, Xiong, Vang, & Pharris, 2009). The authors obtained the stories of Hmong women living with diabetes and developed a drama which they performed for the Hmong women they interviewed. Then the participants were asked their opinion of the drama and whether it truly captured their stories. The dialogue between the participants enabled them to enlighten the community regarding their particular situation in dealing with diabetes and calls for action were issued. All of these studies show how disparaged populations can become empowered and let their voices be heard through AR and CBPR studies.
One method of data analysis of action research data is similar to other qualitative methods: categorizing and coding, identifying themes within the data. However, Stringer and Genat (2004) state that epiphany analysis may be more salient with AR. Epiphany analysis preserves the participant experience and voice by using the "a ha" moments as the primary units of analysis (Stringer & Genat, 2004). The ultimate goal of data analysis is to give voice to these experiences. Key meanings in language are identified through epiphany analysis, which provides ideas and concepts that mesh with the important elements of their experience (Stringer & Genat, 2004). Data analysis should answer the questions: How do people make sense of their experience and how do we use this knowledge to make a positive change in their lives?
Epiphany analysis follows several steps as described here (Stringer & Genat, 2004). First, the researcher must identify key people who have had a significant experience and who represent the entire group. The key group's epiphanies are identified and then deconstructed to reveal salient features and elements. Then the researcher constructs individual accounts describing how each person experiences the issue and interprets the issue. The individual accounts are then reviewed to identify common and uncommon elements. Joint accounts of significant group experiences are then constructed and reviewed for similarities and differences, both across- and between-groups. Lastly, a collective account or overall version is then constructed.
Dissemination of findings
How the findings are presented essentially depends on who the stakeholders are. For AR research using the helical method, the first act that is performed is to present the outcomes of data analysis to the research participants and other stakeholders (Stringer & Genat, 2004). If the primary stakeholders are academics, then dissemination would be expected to take place in journal articles, book chapters, and in teaching content presented to their students (Stringer & Genat, 2004). If the public is the primary stakeholder, then appropriate dissemination outlets would be the media, documentaries, drama performances (Yang, et al., 2009), audiovisual presentations, art, poetry, or even music. Professional or organizational stakeholders would expect for the findings to be revealed through organizational educational methods, such as brochures, posters, pamphlets, education inservice programs, or meeting minutes.
In conclusion, action research is a type of research methodology or approach that can be used when change is needed, when disparaged populations need a voice to effect social change, or to better practices among healthcare professionals. The process is democratic with the participants actually helping design, plan, and conduct the research. Once the data is analyzed, the participants then devise a possible solution and implement it. AR is cyclical or iterative in nature, continuing to evaluate the change and its effects until a reasonable and effective solution has been reached.
- Buck, D. S., Rochon, D., Davidson, H., & McCurdy, S. (2004). Involving homeless persons in the leadership of a health care organization. Qualitative Health Research, 14, 513-525. doi: 10.1177/1049732303262642
- Chenail, R., St. George, S., & Wulff, D. (2007). Action research: The methodologies. In P. L. Munhall (Ed.), Nursing research: A qualitative perspective (4th ed., pp. 447-461). Sudbury, MA: Jones and Bartlett.
- Freeman, E. M. (2007). Exemplar: Practical discourse as action research: Inquiry into post-myocardial behavioral coaching. In P. L. Munhall (Ed.), Nursing research: A qualitative perspective (4th ed.). Sudbury, MA: Jones and Bartlett.
- Hyde, S., Fulbrook, P., Fenton, K., & Kilshaw, M. (2009). A clinical improvement project to develop and implement a decision-making framework for the use of seclusion. International Journal of Mental Health Nursing, 18, 398-408. doi: 10.1111/j.1447-0349.2009.00631.x
- Minkler, M., & Wallerstein, N. (2003). Introduction to community based participatory research. In M. Minkler & N. Wallerstein (Eds.), Community-based participatory research for health (pp. 3-26). San Francisco: Jossey-Bass.
- Petrie, E., Schultz, T., & Pearson, A. (2009). Informing and implementing policy to advance mental health and wellbeing through action research in a rural remote community mental health setting. Australasian Psychiatry, 17, S112-S115. doi: 10.1080/10398560902948373
- Shattell, M. M., Villalba, J., Stokes, N., Hamilton, D., Foster, J., Petrini, R. H., et al. (2009). Depression in Latinas residing in emerging Latino immigrant communities in the United States. Hispanic Health Care International, 7, 190-202. doi: 10.1891/1540-418.104.22.168
- Stringer, E., & Genat, W. J. (2004). Action research in health. Upper Saddle River, NJ: Pearson Merrill Prentice-Hall.
- Wallerstein, N., & Duran, B. (2003). The conceptual, historical, and practice roots of community based participatory research and related participatory traditions. In M. Minkler & N. Wallerstein (Eds.), Community-based participatory research for health (pp. 27-52). San Francisco: Jossey-Bass.
- Yang, A., Xiong, D., Vang, E., & Pharris, M. D. (2009). Hmong American women living with diabetes. Journal of Nursing Scholarship, 41, 139-148. doi: 10.1111/j.1547-5069.2009.01265.x