Rethinking “Contribution”

Theory

Much has been written about theory, and the quote above from Kerlinger (1986) always struck me, even as a graduate student, as somewhat strange. However, over the years, I have come to realize the ubiquity of theory, in that it infiltrates every aspect of our lives. We discipline our children based on some implicit or explicit theory about how to change their behavior. We invest our pensions based on some implicit or explicit theory about what generates wealth. We drive to work based on a theory as to which route we think will take the shortest time (based on traffic). In science, we try to systematically gather data to test the accuracy of our theories and to revise them when the data suggests. Van de Ven wrote, “Good theory is practical precisely because it advances knowledge in a scientific discipline, guides research toward crucial questions, and enlightens the profession of management” (1989: 486).

But what do we mean by theory? Sutton and Staw (1995) describe theory as follows:

Theory is the answer to queries of why. Theory is about the connections among phenomena, a story about why acts, events, structure, and thoughts occur. Theory emphasizes the nature of causal relationships, identifying what comes first as well as the timing of such events. Strong theory, in our view, delves into underlying processes so as to understand the systematic reasons for a particular occurrence or nonoccurrence. (p. 378)

So far, so good; we can all agree on the importance of theory to both research and practice. However, having papers rejected when the leading comment from the Action Editor dealt with the “lack of theory” probably rates as one of my biggest professional pet peeves. Recognizing that this may not have been the only reason for the rejection, I still seethe when that criticism leads off a decision letter for a few reasons. First, sometimes the lack of theory stems more from a different view of what constitutes theory than from a true lack of it. For instance, having grown up in a micro tradition, I was indoctrinated into a mind set of explaining the relationships between the variables one measured based on some overarching theory (e.g., expectancy, goal setting, etc.). However, when I was initially exposed to more macro researchers who kept talking about “telling a story,” I felt like I had been dropped on a new planet. These stories usually consisted of taking an independent variable at one end, a dependent variable at the other end, and then proposing a bunch of flowery processes and variables in between that somehow happen but are not measured. Neither approach is right or wrong; they are both just different and may serve as an obstacle to bridging the macro-micro divide (Wright, Coff, & Moliterno, 2014).

Second, the lack-of-theory concern may just mean a disagreement around which theory is most appropriate. A particular phenomenon could be explained by signaling theory, sense-making theory, or a host of other theories, yet because a reviewer doesn’t like the theory chosen by the author(s), the lack-of-theory criticism rears its ugly head.

Third, and most importantly, I believe that the fact that there is “no theory” should never, in and of itself, be reason enough to reject a manuscript in its initial submission. Theory can be added or changed, that is, as an editorial comment for authors to deal with in a subsequent revision. I should note, however, that in that subsequent revision, if the “lack of theory” still exists, then it may signal that the authors do not possess the capability to adequately leverage theory, and that then may be the time for such an omission to result in rejection.

Contribution to Theory

A more reasonable, yet potentially problematic, criticism of manuscripts stems from the lack of a “contribution to theory.” No top-tier journal can afford to waste valuable space on papers that simply reiterate what the field already knows, yet what the field “knows” may vary based on the perceptions of a reviewer. And to require that every paper contribute to theory may be detrimental to the field. In fact, Hambrick wrote, “The requirement that every paper must contribute to theory is not very sensible; it is probably a sign of our academic insecurity; and it is costing us in multiple ways. These ways consisted of facts having to await theories, contorted, ponderous prose, and too little regard for simple tests” (2007: 1346).

For instance, if the literature had discussed a theoretical issue for a number of years and possibly even comes to conclude something about it based on logic and/or thought experiments, is a paper showing empirical support for the conclusion a contribution to theory? Certainly, few would argue against the necessity for theory testing. Yet, Kacmar and Whitfield (2000) found that only 9% of theoretical presentations in Academy of Management Review articles were ever tested. More problematic would be if their data actually reflected that empirical tests had been published for only 9% when, in fact, a higher percentage had been tested but not published because of a “lack of contribution to theory.”

Or what about the case where authors make a somewhat minor tweak to a research design to answer a slightly different question that has already been answered in the literature? Whetton (1989: 493) stated, “Theoretical insights come from demonstrating how the addition of a new variable significantly alters our understanding of the phenomena by reorganizing our causal maps,” but that still leaves the evaluation of how much altering must take place for it to be considered a contribution. These areas present dilemmas for reviewers and Action Editors, as they must make their own judgments about the “incremental contribution” to the literature. At what point does the incremental contribution deserve publication in a top-tier journal?

Returning to our different approaches to theory, what about a study that presents an interesting narrative regarding what might be happening in a data set, but the data set itself has significant problems? I remember being part of a panel discussion a few years back where in response to my assertion that I am much more concerned about the quality of the data than the quality of the theoretical discussion, one of my macro colleagues stated, “For us it’s all about the theory, because our data is terrible.”

This illustrates the challenge of managing an interdisciplinary journal. Again, these judgment calls on the part of reviewers and editors cannot be pigeonholed into easy checklists. Rather, the guiding principle I suggest is that for studies with great data, we ask, “Does the theory provide enough explanation for the credibility of the data to come through?” and for those with poor data, the guiding principle is “Does the data provide enough support for the credibility of the theorizing to come through?”

Incremental Data Contribution

Finally, an emerging area that creates difficulty for judging contribution lies with multiple papers emerging from the same data set. We all know the difficulty of gaining data and, when possible, the value in gathering as much as possible. One should not spend months of person-hours putting together a data set to then be limited to one paper from it.

However, I have also seen papers submitted that had almost all the same variables as an existing paper, just with those variables configured differently (i.e., variables that were controls in one paper become independent variables in another). Thus, as an editorial team, we have to make judgments regarding at what point the data overlap between (or among) multiple studies becomes so great as to question the incremental contribution of the additional paper.

In order to do so, we have instituted a new step in the submission process requiring authors to note if other papers use the same data set and, if so, to delineate the overlap in variables in a comparative table. We certainly hope that authors feel free to justify the differences between papers in the cover letter so that we can make these judgments with more information regarding the larger context of the stream of research coming from the data. However, I believe that authors have the responsibility to self-report any data overlap, and not doing so will result in an immediate rejection.

I am blessed to have a great editorial team in whom I have great confidence. Cathy Maritan and David Allen have taken on the Senior Associate Editor roles and have been doing a great job. Sucheta Nadkarni, Anne Parmigiani, Bill Schulze, William Wan, and Devi Gnyawali serve as the Associate Editors for all the macro manuscripts; and Eden King, Fred Oswald, Craig Wallace, Bianca Beersma, Brian Hoffman, and Christopher Porter oversee all the micro manuscripts. Michelle Duffy and Dan Beal served as Associate Editors during the first 6 months of the transition, and I thank them for their contributions to the journal. As our team goes forward over the next 3 years, I do not suggest that we will get every call right, but I am sure that the collective expertise in this group will result in vastly more right decisions than wrong ones.