The goal of this research is to make progress towards using supervised machine learning for automated content analysis dealing with complex interpretations of text. For Step 1, two humans coded a sub-sample of online forum posts for relational uncertainty. For Step 2, we evaluated reliability, in which we trained three different classifiers to learn from those subjective human interpretations. Reliability was established when two different metrics of inter-coder reliability could not distinguish whether a human or a machine coded the text on a separate hold-out set. Finally, in Step 3 we assessed validity. To accomplish this, we administered a survey in which participants described their own relational uncertainty/certainty via text and completed a questionnaire. After classifying the text, the machine’s classifications of the participants’ text positively correlated with the subjects’ own self-reported relational uncertainty and relational satisfaction. We discuss our results in line with areas of computational communication science, content analysis, and interpersonal communication.
10 November 2017 in Management Communication Quarterly
For years, organizational communication scholars have been interested in the mechanisms that influence the formation of communication networks. One way to gain a deeper insight into the factors that shape such networks is to model them using exponential random graph modeling (ERGM). However, ERGM has only been applicable to binary networks, reducing communication to something that is either present or not. This article illustrates valued ERGM for organizational communication networks that have a weight associated with each tie. Using a data set on friendship strength between collaborative scientists, results show there are important differences when the network is modeled as binary versus when modeled as valued. In particular, the valued model showed that scientists are more selective regarding friendship (less outdegree activity). Moreover, there were several differences regarding how popular certain disciplines were over others. An online appendix with the R code and data is also included.
Although structuration theory (ST) has endured a considerable influence in communication research across various domains, there remains a paucity of quantitative empirical research using ST as a viable framework. The purpose of this article is to develop an adapted structurational framework to explain the emergence of communication networks. The adapted framework distinguishes between relational states/events and internal/external structural rules. For analysis, we introduce multipanel relational event modeling, a technique using processual inference that can empirically demonstrate the recursive nature of system and structure. As an introductory example of both the framework and method, we examined communication, using cell-phone data, of students living in the same dormitory from September 2008 through February 2009. Our results demonstrated how different structures (i.e., perceived social relations and internal/external structural rules) are reproduced and transformed over time. Our research has implications for communication research dealing with recursivity, event-driven network analysis, the changing nature of the phone call, and the promise of computational communication science.
16 June 2016 in Journal of Applied Communication Research
Persistent school shootings have generated ongoing pressure to assess and enhance crisis communication effectiveness via strategies such as resilience-building and post-crisis restoration. A network perspective offers a robust, multilevel approach for examining complex information flows among community, campus, and individual actors affected by crises. Drawing on relational and structural embeddedness and Uncertainty Management Theory, we offer nine propositions that help explain structural and relational changes in a network over time. Although theoretically driven, these propositions are primarily functional, offering practical implications for safety officials and administrators trying to understand, prepare for, and respond to crises. Thus, we suggest measures and recommendations to improve information flow, uncertainty management, and resilience before, during, and after a crisis.