Authors:
Mary B. Dunn – St. Edward’s University
Candace Jones – Boston College

Interviewers:
Bjoern Mitzinneck – Cornell University
Diego Soares – Queen’s University

Article link: http://asq.sagepub.com/content/55/1/114

Question 1. Your study examines the dynamic tension between the science and care logics in medical education over time, contributing to our understanding of professions that espouse plural logics. Through a mixed methods study, you identify inter-professional rivalry between medicine and public health, intra-professional contestation among physicians, as well as exogenous forces of waxing and waning public attention to managed care as influencing the balance between the science and care logic. How did you identify this interesting research context and how did the framing for this study emerge? Did you know from the beginning that you would be conducting a mixed methods study?

The interesting context came out of Mary’s dissertation research (Candace was the dissertation chair) that examined a top, science oriented medical school program and sought to understand the role of social and mentoring networks in physician-scientist innovation as measured by both patents and academic articles. The field research for the dissertation revealed a strong focus on science as the basis for medical care in this organization, and faculty physicians expressed concerns with the increasing care orientation prevalent at other medical schools especially as they perceived it potentially affecting their own scientific curriculum. We wondered how quality care became defined in a particular way, and sought to understand the contrasts physicians used between science and care as distinct or complementary approaches to medicine.

When we started analyzing the history of medical education in the United States, we saw quite clearly that science was the foundation of medical training, and the dramatic rise of a care logic corresponded temporally with a shift in women attending medical schools. We wondered if these trends were related. Our original submission to ASQ was a historical narrative examining the changing logics in the profession of medicine and gender dynamics. The editor and reviewers found our study interesting, but not convincing enough so they urged us to supplement the historical narrative with multivariate statistical tests of our hypotheses and demonstrate our insights. This meant that we needed to collect systematic data on changes in the environment of medical schools to ascertain what was driving the shifts in the medical professional logic and in the definition of what care meant over time.

Question 2. Your study is one of the comparatively few mixed-methods studies in the leading journals of our field. Do you have recommendations for junior researchers aspiring to similarly rigorous, method-spanning work and successfully steering it through the review process?

We did not start out as a mixed method study; it evolved, as noted above, in response to the editor’s and reviewers’ queries and recommendations to improve the study. The hypothesis testing portion of the study took another year in data collection and additional analyses. We think this really strengthened the study, answered the questions of the editor and reviewers, and enabled us to move from a high risk R&R to a conditional accept in one revision. In hindsight, it would have been helpful to start as a mixed methods study. Thus, we suggest that junior and more senior scholars start from a framework of mixed methods especially when their research questions involve phenomena anchored within a specific context such as institutional logics. We both tend to use mixed methods in our studies (as noted above Mary’s dissertation was mixed method) since such an approach enables in-depth insights about the dynamics at play in the research context and allows researchers to test their qualitative insights more rigorously. Of course mixed methods studies can take significant time and effort to execute, and require expertise in both qualitative and quantitative analytical techniques, but the results can be well worth it. Mixed methods approaches can answer editors’, reviewers’ and readers’ challenges and help the authors strengthen their arguments.

Question 3. Based on content analytic methods, your paper is pioneering a method of measuring the attention to institutional logics. Coding annual reports published by the Liaison Committee on Medical Education, you are using standardized relative frequencies of inductively identified keywords to measure the relative attention of the science and care logic in medical education in North America over time. What would be your suggestions to other researchers who would like to use similar content analytic methods to capture logics? What are important data requirements, what are pitfalls to watch out for, and for what levels of analysis (society, field, profession, organization) may this method be used?

We think using content analytic methods to capture logics provides strong advantages. First, it has descriptive validity. In our study, content analytic methods demonstrated that science and care were distinct logics because the texts revealed how medical professionals defined a science and care approach. Second, content analysis gave voice to the actors in the context, and it provided a clear chain of evidence from data to theory. When there is a dominant and persistent logic, such as science, the content analytic methods and reliability of the logic as a construct were relatively easy and pretty straightforward. In contrast, the care logic was much more challenging because the logic was the confluence of many distinct perspectives—physician-patient relationship, public health which focused on prevention and statistical analysis of health issues, and managed care, which focused on efficient use of resources. The hybrid nature of the care logic, with its competing voices and changes over time, created a more challenging task of identifying what is similar and stable among disparate voices that co-create the hybrid logic. Thus a pitfall of content analytic methods is that they pick up and reflect the confusion and messiness of dynamic environments with multiple perspectives. In a plural environment that has multiple competing voices contending over the logic, then the researcher may need to shift to a theory driven rather than inductive, bottom up approach. For instance, in another research project on the rise of modern architecture (Jones et al., 2012 Organization Science), the bottom up approach revealed competing voices whereas the use of binary contrasts revealed both the competing factions and meaning and change in meaning of modern architecture over time.

We believe that content analytic methods are an approach that can be used with any level of analysis as long as the data represent well that level of analysis. For example, content analysis of individual journals and diaries can capture individual sense making and identity dynamics. Analysis of organizational memos or meeting minutes among committee members or top management can capture organizational concerns and issues, and professional journals can capture field level dynamics.

Question 4. We noted in your acknowledgement the diverse set of scholars and field participants you engaged with in the process of writing this paper. Could you tell us a little bit about this process and your experiences on how to make the best use of discussions with field participants, peer-feedback, and conferences in preparing manuscripts for submission?

We sought to ensure from the outset that our analysis represented the medical profession. We discussed our research with physicians, both practicing and academic, asked for their input into the dynamics of the field, and had them read drafts of our paper. The physicians were super engaged by the study. They acted as a validity check for our understanding of the medical profession and its dynamics over time. We presented the paper at the EGOS conference and received a strong positive response from colleagues who suggested that we submit it to ASQ. Once we received the invitation to revise the paper, our colleagues were incredibly generous and important in helping us move the paper from high risk R&R to a conditional acceptance in one revision. We specifically asked a mix of colleagues skilled in different areas to read and comment on our paper including people with theoretical expertise in institutional theory, domain expertise in medicine, and methodological expertise in longitudinal statistical techniques. When we had dramatically reduced the number comments, questions, and confusions from our collegial readers, we knew it was time to resubmit the paper to ASQ.

Question 5. We were intrigued to see so many topics in your article from 2010 that are receiving much attention in current logics research. For example, you highlight the importance of paying attention to field heterogeneity, a call that was heard frequently this year in the sessions of the Academy of Management Meeting in Vancouver. You also discuss hybridization in your paper, a topic that grew enormously in recent years. Can you compare the perspective on hybridization in your article, with the more recent developments? Generally, how do you see the evolution in the logics literature in these past five years and where would you like to see it going?

Thank you for your observation! We are pleased that our ASQ paper foreshadows so many topics important to studying institutional logics and institutional theory. We believe that deep immersion in the context, such as professions, helps to reveal important issues, such as field heterogeneity. Professions are contested sites, as seminal sociological analyses of professions reveal (e.g., Abbott, 1988; Edelman & colleagues, 1999; 2001; Friedson, 1970; 2001; Larson, 1977). The professionalization project is fundamentally about who and what is included in the profession, and who has authority and control over its work and its meaning.

Field heterogeneity and hybridization are related; when a field is heterogeneous, then multiple voices seek to influence the profession and its logics, creating multiple logics and blends of voices within a logic. Most of the recent theorization and empirical studies of hybrid logics have been within the organization, which presumes that the demands for coordination and control may be greater within an organization and that hybrid logics may be more likely and more challenging within an organization. The degree of coordination, however, may be quite demanding between independent organizations, such as in the medical care of a patient. The challenge and likelihood of hybrid logics may result more from the type of task interdependence, such as pooled, sequential or reciprocal in James Thompson’s terms, than from where organizational boundaries reside. In medicine, the nature of interdependence among professionals may differ significantly depending on the areas of specialty and the division of labor within specific areas. Scientific medicine tends to be more specialized and perhaps more buffered from invading interpretations of medical care, whereas clinical medicine tends to comprise more primary care physicians and those working to coordinate care with other professionals such as nurses, public health professionals, and social workers.

We agree that the relationships between logics and the evolution of hybrid logics are an important area for future research. While hybridization and competition over logics may play out in organizational settings, they are ultimately rooted in the broader institutional and professional fields. In our study, we found that both logics of science and care are needed for effective medical care, but that the relative balance between science and care has shifted over time. Logics are continually evolving and being shaped by the actors that use and promote them.