Abstract
“Power to the people” is a popular political slogan that is used to convey the essence of democracy and participation. This paper derives from an action research intervention to implement participatory evaluation in a medium-sized private hospital in Nigeria, where the author was a training consultant. While the original objective of the intervention was to improve effectiveness of training through participatory evaluation, the study enabled an important outcome: it helped in clarifying and resolving social power differences prevalent in evaluation. The paper therefore contributes to the theory and practice of action research by describing how a researcher may better understand and relate to others with regard to social power. This plays a potentially useful role especially in participatory evaluation.
Introduction
This paper derives from an action research intervention to implement participatory evaluation in a medium-sized private hospital in Nigeria where the author was a training consultant. The original objectives of the intervention were to deepen insight and understanding about evaluation of training from the perspectives of stakeholders and implement a change of evaluation system in the hospital. However, underpinning the study in social constructionism produced an additional important outcome: it helped in clarifying and resolving social power differences prevalent in evaluation. The paper therefore focuses on the procedural aspects of the study and concludes that how the researcher understands and relates to others with regard to social power could contribute to understanding the role of action research especially in evaluation.
Context
In the process of studying the transition process to a participatory evaluation method in Zenith Medical Centre, I gradually began encountering various empirical complexities in the field. This sparked immense interest in me and led to the development of critical and methodological reflections about my 18 months of action research experience on the intervention. Although the intervention started with a single apparently impersonal focus on evaluation, the process around sharing power interpersonally as a key to the success of the evaluation became much more focal as our effort developed. That process is the focus of the rest of this paper.
Ordinarily, my background would hardly lead anyone to consider the subtle issue of social power as important. Although an external management training consultant to Zenith Medical Centre among other hospitals, organizations and institutions, I am a professional accountant with obviously strong quantitative background. I am also a male from an African culture where being a male is traditionally equated to superior power. In addition, my designation as the director of program of my training firm and prior role as the chief executive officer of a health maintenance organization in whose network the hospital was a provider placed me in a position to regard most of the research participants as inferior. Moreover, as a reputable United Kingdom medical school postgraduate, it might be right to say that I had everything going for me with regard to social power in such a hospital environment at the time of the intervention.
However, my exposure to qualitative methods at both scholarly and practice levels over the past decade has made me become increasingly interested in the kinds of power relations which exist in the workplace and the implications these have on the effective organization and management of the working environment. I started doing action research as a requirement for completing a critical action learning-based doctorate degree in business administration. This coincided with the client management’s desire to change the training evaluation system of the hospital.
The issue of social power differences arose in relation to access, politics and ethics in unique ways peculiar to the situation and relationships with and within the organization. This included my researcher role as partly an insider and partly an outsider. Power was also concerned with the attitudes, norms and relationships existing between the different levels of the organization hierarchy: between doctors and nurses; between medical and administrative staff; and between hospital staff and customers. Besides being a hospital setting, social power differences are prevalent in evaluation, for example, of training. This could be due to the multiple stakeholders in training (Nickols, 2005). The resulting role ambiguity and conflict usually faced by insider action researchers have also been a recurrent problem in action research (Coghlan, 2001; Coghlan & Brannick, 2010).
Conventional power hierarchy in Nigeria
In Africa, particularly in Nigerian health institutions, there are acute power struggles between different healthcare professionals with medical doctors claiming superiority over all others. It is even more worrisome in the private sector where small to medium-sized hospitals are usually owned and managed by medical doctors. Others serve as subordinate staff or employees. Moreover, the issue of gender hierarchy in Nigeria where, traditionally, males are considered more powerful than females adds to doctors’ domineering posture, especially as there are fewer female doctors than males and fewer male nurses than females. Thus, I was concerned that while my familiarity with all the ranks of the hospital staff would provide easy access to data, it might be difficult to get nurses to speak up in the presence of doctors who were co-participants. Also, I feared that perception of my existing closeness to the top management of the hospital who are all male doctors might present ethical issues for me to address (Brydon-Miller et al., 2006). For example, nurses might see me as one of “them” (directors/doctors) and tend to withhold information. Conversely, directors/doctors might take me either as one of “us” or merely as an external training consultant who perhaps desired only to justify his continuous engagement with the hospital. The research design, which is rooted in social constructionism, proved helpful in changing how I understood and managed these power relations to produce practical value to the participants, the organization and the action research community.
Subjectivism and social constructionism
The values researchers hold and the ideological perspectives that guide them can be powerfully influenced by the choices they make in the course of inquiry (Brown & Tandon, 1983). The underlying assumptions of my action research influenced how I went about conducting the research and the criteria by which I wish it should be evaluated (Cassell & Johnson, 2006). Of particular interest, they also helped produce outcomes that are the subject of this paper.
I considered Cunliffe’s (2011) typology of knowledge problematics, as perhaps a more useful way to articulate my ontological stance on subjectivism rather than the age-long interpretive or functionalist paradigms. My position was hinged upon the basic understanding of subjectivism as historical, social, and linguistic construction of reality (Cunliffe, 2011). This view assumes that knowledge, meanings, and sensemaking are embedded in particular contexts, time and place as constructed by people in their everyday interactions in relationship with their surroundings. Hence, humans are viewed as intentional and reflective subjects, as constructors of social realities, interpreters, and sensemakers in the research process (Cunliffe, 2011). Subjectivists also extend the view of the world as where there are “truths” instead of one truth. This assumes there is no universally accepted understanding of any phenomenon and therefore no one person or group of persons has a claim to “correct” understanding above others.
My epistemological stance was social constructionism, which acknowledges the first principle of action research: that the self is relational. This means that both inquirer and co-inquirers benefit from diversity of voices (Bradbury, 2015). From the social constructionist perspective, all data are seen as mediated by both researcher’s reasoning and those of participants (Johnson & Duberly, 2000).
Subjectivism and social constructionism share some common basic philosophical assumptions. As in the subjectivist ontology, we start the constructionist research design with the assumption that there is no absolute truth. Action researchers do not expect real-world problems to come well-formed (Schon, 1992), so we try to “establish how various claims for truth and reality become constructed in everyday life” (Easterby-Smith, Thorpe, & Jackson, 2008, p. 93). Gergen (2015) maintains that this is a shift from the traditional social science effort at mirroring to re-making the world. Of course, this ideological shift is not new. At least from the time of Emmanuel Kant in the 18th century, it has increasingly been recognized that what leads us to believe that a world exists “out there” is but a set of interpretations of our perceptions and experiences. The social constructionist perspective assumes that the authority of knowledge or how we come to know ultimately derives from a “knowledge community” of people at any point in time. Kuhn puts it more pointedly: “Knowledge is intrinsically the common property of a group or else nothing at all” (1970, p. 210).
The difference between this constructionist view and the more traditional views of knowledge as what individuals believe distinguishes Piaget’s “constructivism” from Papert’s “constructionism”—terms which are sometimes used interchangeably. Gergen and Gergen (2008) explains that while constructivism is associated with consideration of meaning-making as taking place in the individual mind (cognition), constructionism is considered as a product of human relationships. This does not mean that individuals have no ideas but that such ideas must ultimately be given meaning by their social context. This social constructionist epistemology fitted the intentions of my research because what constitute training evaluation data are socially constructed.
The implication of this stance was that my research data should be analyzed in such a way as to invent training evaluation phenomenon through meaning-making and reflexivity. Participation and democratization were also implicated as the research process must ensure that the usually unheard voices of participants from the female-dominated, non-director levels of the organizational hierarchy were included. The end result would neither be the generation of universal models of training evaluation practice nor a stipulation of “best practice” standards, but the construction and co-creation of knowledge and understanding with rather than over the participants/stakeholders.
Methods
I followed an action research process that involved collecting evaluation data from training stakeholders who were purposively selected as research participants, through focus group and one-on-one interviews. This approach was chosen because action research “brings together action and reflection, theory and practice, in the pursuit of practical solutions to issues of pressing concern” to people (Bradbury, 2015, p. 1). The research design therefore involved going through the iterative action research cycles of constructing, planning, acting, and evaluating in collaboration with participants (Coghlan & Brannick, 2010). Each of the four cycles culminated in an intervention.
In the first cycle, I engaged participants in constructing the action research intervention. My choice of research site was firstly informed by the unique challenges hospital settings present to evaluation of novel interventions, particularly in capacity building (Wharton & Alexander, 2013). Secondly, I considered power relationships among the various professionals in healthcare an interesting area to learn about. Thirdly, I preferred Zenith Medical Centre because of my familiarity with or pre-understanding of the political dynamics of the hospital as an external training consultant for more than five years prior to the project. Together with the hospital management, I identified shallow use of training evaluation data as the “red-hot” issue confronting the organization and participatory evaluation was considered a possible means of resolving the issue. Participatory evaluation was chosen because it incorporates some elements of the other methods of evaluation that are relevant to the research question in terms of function, control of decision making, selection for and depth of participation (Cousins & Whitmore, 1998; Cullen & Coryn, 2011).
To determine the intentions or desired outcomes of the training stakeholders, I drew a table of key focus group and one-on-one interview questions for the second, third, and fourth cycles. To qualify and select participants, a rigorous ethical approval process was followed and obtained after about six months. Ten representatives of training stakeholder groups in the organization were identified as potential participants. The eligibility criteria were that a participant must have contributions to and inducements for training, be willing to participate voluntarily and have the intention to “self-study in action” with the researcher (Coghlan & Brannick, 2010). These equal inclusion requirements or ground rules were to ensure that everyone’s opinion would count and that no one would dominate conversations. Eight out of the 10 sampled stakeholder representatives volunteered to participate by signing the informed consent. They consisted of three males and five females. Two of the male participants were medical doctors—in fact, co-medical directors who represented executive management. The third male was a laboratory scientist and head of lab services. The five females were made up of three nurses/midwives, one lab technologist, and one administrative officer. As the facilitator of the training, my researcher role as partly an insider and partly an outsider was negotiated at this stage.
In the second cycle, I considered it critical that obtaining rich evaluation data was a key to addressing the research question. Deciding on the methods of collecting and analyzing data and the roles of participants in this process called for the application of Kitzinger’s (1994, 1995) contextual constructionism of focus groups. Focus group has been hailed as a means of generating insightful and useful rich data in timely and cost-effective manner (Davies, 2008). In the constructionist perspective, having heterogeneous participants (with marked status differences) increases the generation of divergent views likely to deepen understanding. Homogenous groups (or groups of equals) may lead to easier conformity among group members and inhibit detailed discussion (Freeman, 2006). However, what Freeman did not recognize is that in some cultures, voices of less powerful group members can be limited or silenced when they don’t share the opinions of their superiors (Verhezen, 2010).
The first focus group for the purpose of generating evaluation data was held at the hospital’s boardroom. It lasted for about 2 hours during which time the stakes of participants in training and their key contributions and inducements were identified. This was achieved through artful use of semi-structured questions and probes around the key focus group questions. I collected data using audio recording which I personally handled after duly explaining the procedure and getting participants’ agreement on the ground rules. The use of an audio recorder enabled me to analyze my own communication during the discussions. I also kept field notes where I jotted side talks and non-verbal clues that could not be captured through audio record.
I listened several times to the audio-taped interviews before transcribing soon after each cycle so as to vividly remember and note the non-verbal behaviors associated with the comments. In line with template analytical approach to discourse as a style of thematic analysis (King, 2012; Taylor & Ussher, 2001), I familiarized myself with the transcripts and discussed with the participants to correct any errors in transcription. I then identified initial discursive themes and coded these along the margins of the transcript where they appeared. I later grouped the themes together, checking for emerging patterns, for variability and consistency, and for the function and effects of specific discourses. I conducted the interpretation of these themes through a process of reading and re-reading as well as referencing relevant literature and discussing emerging sensemaking with the participants in the last action research cycle. To further ensure the correctness of my interpretations, I checked and rechecked the themes/codes against the focus group and interview transcripts to form primary interpretive categories, some of which included two or more “secondary discourses” or “sub-codes” (Taylor & Ussher, 2001). Lastly, I examined the data for differences and commonalities within and across code categories to make apparent discourses underlying systems of meaning.
From the stakeholder perspective, Nickols (2005) has identified training evaluation data as falling under three categories, namely, stakes in training, contributions to training, and inducements for training. Based on this categorization, I asked participants to answer three key questions from their perspectives. Every comment was in turn discussed by group members in the form of dialogue, probing or asking clarifying questions in a social atmosphere. I ensured that my personal opinions minimally impinged on the conversations.
In the third cycle, I engaged participants in an actual participatory evaluation in which the increased data generated from the analysis were fed into the design and facilitation of the ongoing customer service training. This was done by ensuring that identified stakeholders’ contributions were incorporated as inputs to the redesign of the training such that their inducements could be realized. This action process represented pragmatic philosophy which addresses practical outcomes (Shotter, 2010). Hence, I used the results to drive productive conversations with participants through one-on-one interviews that lasted between 15 and 30 minutes each at different times within four months. The conversation that ensued enabled deeper insight into participants’ world-life that produced the desired change in the evaluation of training in the organization and in other areas not originally contemplated by the intervention. It also provided the team an opportunity for “reflection-in-action” (Schon, 1992).
The unstructured interview questions concentrated on three main issues, namely: (1) the observed or experienced impact of the training on the participants’ work, others or organization, (2) the effect participants thought the increased use of evaluation data had on the perceived effectiveness of the training, and (3) whether and why participatory evaluation should be recommended for future improvement of training. I urged every interviewee to relate (judge) the evaluation as much as possible in relation to their stakeholder perspective. As a result, some of the responses were quickly followed by examples derived from the workplace after the training.
Clarifying power relations
What it means to have a stake in training
Template of a priori themes (stakes) in the first focus group discussion.
Understanding of participants’ stakes was not straightforward from individual perspectives. Divergent opinions related to participants’ positions in the organizational hierarchy were expected given that dissention is a critical component of action research (Bjorn & Boulus, 2011). For example, when a management executive asked, “They said we are line managers…?” another management executive interrupted and described line managers as represented by unit heads who supervise the work of others. There was also the question of “stake in what?” by a line manager. After prompting several suggestions, a management executive offered to summarize what he then understood by “stake”, without anyone countering: “Your stake is how you are involved…talking about why are you here – as an employee or a trainee? Your stake may be that you are a line manager” (Management executive).
With “stake” understood in this context, four stakeholder groups were identified as trainers, trainees, line managers, and management executives. However, there are other training stakeholders identified by Garavan (1995) and Nickols (2005) which were not found here. That only four groups were identified could be attributed to the size or structure of the organization.
I expected critical views also about what contributions to training meant to participants. Again, the same management executive who previously summarized interpretation of “stake” offered to interpret contribution in his own words while others merely watched. I intervened by inviting everyone to freely express themselves, reminding all about our commitment to participation and democratization. The key themes participants used to describe their contributions reflected mostly input values (cost) in training return on investment measurement (Phillips, 1997). Management executives still assumed ownership of training with the word “we” while referring to the other stakeholders as “you”, thus heightening the social power conflicts among the group. “We’re spending a lot of money even though some of you will not see the physical cash, but we’re actually spending a lot of money if you monetize those things” (Management executive).
Understanding the “insider–outsider” role duality
At this point, I realized that we needed to widen the scope of discussion on inducements as expectations from training so that all voices could be heard. This opened up deeper understanding, critical insights, and peer probing but led trainee stakeholders to address me as the owner of training in the organization. This was perhaps to contrast management executive’s view regarding training “ownership,” thus introducing a conflict to my dual role as insider–outsider. Without taking sides, I tried to clarify the conflict by referring to our earlier agreed negotiation on my dual roles as training consultant and researcher.
Top of the themes that emerged as representing general categories of inducements for the different stakeholders included enriched knowledge, enhanced skill and attitude change, although management executives still named return on investment expectations such as client retention and increased revenue/profitability. Interestingly, the discussion of “better me” used by a management executive to describe his expectation generated a debate. A line manager asked if by wanting to make himself a better person the executive was presently having a problem with his personality which he needed to work on. The management executive gladly replied (and the gratitude was registered in his countenance while the shift from his relaxed to upright sitting position perhaps either signaled that he welcomed the challenge or was surprised by it). Other participants joined in the dialogue, explaining how the intervention could make them better as individuals and the hospital as a whole.
Resolving social power differences in the group
Gradually getting to convergence, trainee stakeholders equally shared the client retention sentiment of management executives as an inducement for the training. One of them expressed how this evaluation data related to her job as a nurse and to the hospital as her employer. “We want training that will teach us how to deal with [patients] so that they go out and tell others to start coming here” (Trainee).
Another trainee captured conversations about attitude change as follows: “I want a situation whereby there should be a change after this project. Even if I leave this place to another place, they should see a difference from where I was coming” (Trainee).
Responses regarding the evaluation data analysis and actual utilization gave evidence that the social power gaps had narrowed. This naturally led to participants’ recommendation of participatory evaluation. For example, a trainee participant who would want participatory evaluation extended to all future trainings in the organization said that it had become clear that participatory evaluation would help give more information about the training beforehand. To her it was “So far so very good.” Another trainee participant thought that participatory evaluation helped everyone to get prepared and not “jump” or rush into the training. Management executives illustrated how the intervention, by breaking the barriers of social power differences among the ranks, created relationships that significantly improved individual and hospital performance: By the improved interrelationship of staff: relationship between staff and management, interrelationship between management staff, attitude of staff to clients and the happiness of the clients, the profitability [or goodwill] of the business has gone up so much after the intervention. (Management executive)
From evaluation over to evaluation with stakeholders
The issue here was how to do an evaluation in the Nigerian culture where conventional power arrangements reflect social distance which is considered by most to be an unproblematic norm. This seems to have happened at the same time that I became aware of my own power and impact. So my awakening helped awaken the system I was serving.
Evidence from the data analysis showed that by increasing the quantity and quality of evaluation data and utilizing such data by all stakeholders, implementation of participatory evaluation resulted in more effective training outcomes for stakeholders and the hospital. Participants expressed their learning and insight from the process as manifesting in change of perceptions regarding training, evaluation and customer service. They also named better quality training and change in participants’ behavior that resulted in improved customer service. These were the desired outcomes or effectiveness of the training as primarily intended by the study.
Interestingly, a significant outcome of the intervention was that it helped clarify and balance social power differences in the group. This important result was produced by constructing the action research in such a way that encouraged open communication and required understanding of the phenomena as co-created, context bound, relational, and situated (Susman & Evered, 1978). The process conformed to the assumptions of social constructionism hinged on the subjectivist view of reality as a creation of human relations (Cunliffe, 2011; Gergen & Gergen, 2008) and involves participation and democratization as key elements. “Authentic participation in research”, explains McTaggart, “means sharing in the way research is conceptualized, practiced, and brought to bear on the life-world. It means ownership – responsible agency in the production of knowledge and the improvement of practice” (1991, p. 171). Hence, it generates commitment (Dick, 2002). Democratization is important for collaboration in action research. This addresses the African culture that discourages open contribution to conversations with perceived superiors (for example, children amongst adults, women amongst men, etc.) and which transcends organizational hierarchies, especially in multiprofessional situations such as hospitals. It also recognizes post-structuralists’ concern for, or Foucauldian perspective of, “the micro-politics of power that shape society” which action researchers share (Ozanne & Saatcioglu, 2008, p. 425).
Specifically, the constructionist action research design features of participation and democratization were activated to enhance the often suppressed nursing and non-management staff to have their say in identifying, clarifying, understanding, and resolving social power differences in the group. From the first focus group when participants were categorized according to their stakes in training, template analysis showed that the group was literally divided into owner-managers (all doctors) and employees (mainly nurses/mid-wives) in the organizational hierarchy. The line manager and researcher-trainer stood in-between the two levels. Although heterogeneity was recognized in the constructionist design, it was also identified as a source of social power conflicts which I had to deal with. In action research, solutions to problems are negotiated among the interests of stakeholders with different power and resources (Herr & Anderson, 2005).
However, I was concerned when a management executive’s voice was veering to dominate other voices, such that his own interpretation had to be heard before others commented, if at all. I tried to adjust this overbearing tendency by personally pointing at individuals one after the other to speak, referring constantly to the ground rule which grants equal opportunity to every participant to be heard during conversations. Despite this effort, I was surprised by the way management executives still assumed ownership of training with the words “we” and “our.” I realized at this stage that although I spent a lot of time educating participants on the process of action research, I had not discussed the underlying assumptions that guided the research with them. I decided to include that important education in the second, reflective focus group.
I was initially jolted by a participant’s references to “my constituency” and “somebody from the outside,” as I thought about my role duality and associated issues usually encountered by action researchers (Coghlan & Brannick, 2010; Evered & Louis, 1981). I was not certain that he was referring to me as an outsider who could not “see” or understand his constituency. However, I quickly made a note of this and paid closer attention to the rest of responses, recognizing that active listening was a key requirement from me as the interviewer. At this stage, when employee/trainee participants were gradually beginning to voice their opinions on conversations, I was addressed as the owner of training in the organization—a title management executives already claimed. This proves that “self-referentiality” is inevitable in insider action research (Kristiansen & Bloch-Poulsen, 2004).Especially, when I felt that explaining the underlying ideas and rules guiding the research to participants would help drive discussions reflexively, unwittingly, my multiple roles as trainer-instructor and researcher-interviewer was becoming more obvious.
Contrary to my expectation, I was challenged by the positive reaction of all levels of participants when I realized that by embedding self, the insider–outsider role was not perceived in bad light. This changed my perception from the dominant view of “insider–outsiderness” as a thorny path in action research that needs treading with caution (for examples, see Humphrey, 2007; Williander & Styre, 2006). Rather, I found my position as an advantage from both sides of the hyphen.
Deeper understanding of my insider–outsider role helped me in the last stage of finally breaking the barrier to group cohesion. Initially, the risk of making sensitive comments prevented employee—predominantly female—participants from expressing themselves freely, given the social power differences in the team. Management executive participants were more enthusiastic at the beginning while trainee participants almost had to be coerced before I co-constructed the research design with all, emphasizing the ground rules derived from social constructionism. It took the first challenge to a management executive by a line manager on “better me” claim to practically break the pervading culture of silence (Verhezen, 2010), which initially prevented participants from challenging the opinions of superiors and open up dialogue. I had a mixture of surprise and excitement when the trainee participants started being so vocal that they were even expressing personal plans such as “leave this place to another place.” These are topics not normally discussed in the presence of superiors in similar organizations in Nigeria.
Willingness of participants to act as “gatekeepers” to knowledge was also crucial to applying the constructionist principles, accommodating each other’s opinions irrespective of the official status of the individual (Ashton, 2004). Most of the participants agreed it was the first time they engaged in a collaborative research where directors and employees discussed as equals, and that their perceptions of taken-for-granted practices changed significantly.
First person change benefits second and third person view of power
Equipped with the theoretical and procedural backgrounds with which the research was designed, I engaged training stakeholders as participants in action research cycles. Not only were its originally intended outcomes achieved but it also produced valuable outcomes. Social construction of the intervention encouraged research with rather than on people to clarify and resolve social power differences prevalent in African hierarchical organizations and which has also been a recurring issue in evaluation and action research. Practical results included improved relationships, which led to increased performance of individuals and the hospital. The paper has not only demonstrated that social constructionism complements action research but it also contributes to the repertoire of knowledge about how the first person change resulting from social construction of the action research intervention has implications for second- and third-person research and practice (Chandler & Torbert, 2003; Reason & Bradbury, 2008).
The first-person change was that the intervention helped meet the researcher’s objectives to deepen understanding and provide professional development for his practice. Besides improved understanding of the “insider-outsider” role of the action researcher, I have also developed an increased capacity for carrying out action research projects including how to work with a heterogeneous team in planning, acting, evaluating, and dealing with challenges. The second-person implication was that it deepened understanding and provided improved day-to-day practice for the participants who intended to self-study with me in action. It also provided practical solutions and organizational learning for the client. In the third-person inquiry, specific knowledge has been provided about how researcher relations with regards to power could contribute to understanding the role of action research, particularly in a hierarchical African situation where the action research practice is new.
Footnotes
Acknowledgements
I wish to thank Dr. Lisa Anderson and Dr. Claire Rigg for their advisory input in the draft version and the two anonymous reviewers of the original version of this work. Should there be any comments/reactions you wish to share, please bring them to the interactive portion (Reader Responses column) of the website:
.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
