How to Measure Shared Leadership?

[We’re pleased to welcome G. James Lemoine of the University at Buffalo–State University of New York, Gamze Koseoglu of the University of Melbourne, Hamed Ghahremani of the University at Buffalo–State University of New York, and Terry C. Blum of Georgia Institute of Technology. They recently published an article in Organizational Research Methods entitled, “Importance-Weighted Density: A Shared Leadership Illustration of the Case for Moving Beyond Density and Decentralization in Particularistic Resource Networks,” which is currently free to read for a limited time. Below, they reflect on the methodology and significance of this research:]


What motivated you to pursue this research?

This research started its life as a second-year PhD student seminar paper, with a completely different research question and design. I was very interested in shared leadership and how different patterns of leadership within a team might affect its outcomes. Over several iterations of that paper, though, I became increasingly dissatisfied with the way shared leadership was measured in the literature. Specifically, I wasn’t convinced that the ways shared leadership had been measured – as an aggregation, or as network density or decentralization – could fully capture it in a way consistent with its conceptual meaning. I shared these concerns with a few co-authors who are far smarter than I am, and we agreed that tackling this measurement issue was potentially more important and interesting than my original research question. Further, over the course of the manuscript’s development and with the help of co-authors and reviewers, I soon realized that these measurement issues aren’t limited to the study of shared leadership. In fact, we feel they’re widespread throughout the organizational literature on team properties with particularistic qualities. There are many other team constructs, like shared mental models and advice networks, where a ‘tie’ from one member to another becomes more valuable when the sender is better connected. For instance, someone receiving lots of advice from others should in turn offer better advice, and if someone views you as a team leader, that’s a more powerful link if that person is him or herself seen as a leader by others (and it’s more consistent with the core idea of ‘sharing’ leadership). When we realized how many streams of research might benefit from a network statistic that specifically accounted for these types of team configurations, we hoped that we might make an impact on the field by proposing a potential solution.

What has been the most challenging aspect of conducting your research? Were there any surprising findings?

In order to build a formula for a network statistic that would solve the issues we encountered, it was necessary to do a ‘deep dive’ into the literature on network methodology: You can’t suggest an additional path forward if you don’t understand where the literature has been. This was at times a difficult challenge for us, as none of us are focused methodologists. Speaking only for myself, many older methods papers can be difficult to decipher for a relative layperson like me (one of the reasons I like ORM is that so many papers are written relatively simply).

We tried to build on that research to generate a new measurement tool that would provide added value, and I can’t tell you how many weeks we spent going over and over our formulae to make sure they were accurate and appropriate. I have a stack of Pizza Hut napkins on my desk right now, covered with scribbled math and algebra (a habit which did not amuse my wife). And finally, after we were confident that we’d got it right, the next challenge was to distill it into a manuscript that everyone, not just network statisticians, could understand. Hopefully we did a good job of this.

What advice would you give to new scholars and incoming researchers in this particular field of study?

Don’t be satisfied with the way research is conducted, just because that’s the way research has always been conducted. Just because an assumption has never been seriously questioned does not mean it should not be questioned. Just because an idea or a theory or a method has been printed in an A-journal, doesn’t make it right. Always approach research from the perspective not of how it’s commonly done, but how it should best be done. Along those lines but more specific to our paper, this means carefully examining how a construct of interest is measured. We proposed the importance-weighted density (IWD) statistic for particularistic resource networks, but we acknowledge that what measure you use really depends on your research question. There are some hypotheses for which density or decentralization would be a better fit than our IWD. As always, conceptualization and theory should drive measurement.

Stay up-to-date with the latest research and sign up for email alerts today through the homepage!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s