Discourse Analysis

One of the major cross-cutting thrusts of my work is identification of conversational constructs that predict important individual difference variables including motivational constructs and participation goals as well as individual and group outcome measures of success such as learning, knowledge transfer, trust, and stress reduction. The theoretical contribution of my work in this area is the reinterpretation of largely qualitative frameworks from sociolinguistics and discourse analysis from a computational perspective, with a particular focus on frameworks characterizing interpersonal dynamics within the theory of Systemic Functional Linguistics (SFL). This work comes together in a multidimensional framework referred to as SouFLé, published prior to my tenure review in the International Handbook of Collaborative Learning and featured in analyses of two out of five focal corpora in my co-authored edited volume on Productive Multivocality in Analysis of Collaborative Learning Interactions. Published analyses of the framework applied to corpora demonstrate the predictive value of the framework in connection with these external success measures. Recognition of this work has continued since my tenure review in the form of invitations to write handbook chapters in three different research and practice communities.

As a first example, the work was featured in an event hosted by Educational Testing Service (ETS) on innovative approaches to Assessment of Collaboration, and then later as a chapter in a handbook on Innovative Assessment of Collaboration, with select chapters from the ETS sponsored event. Building on this, the host from that event invited me to be a featured speaker representing this work at the recent Association for Test Publishers: Innovations in Testing conference in order to further disseminate my work to assessment practitioners. A write-up of this presentation was invited to be featured in an in preparation ETS Research Report publication. These opportunities follow on inroads into the assessment community that began at the time of my tenure review when a collaborative problem solving assessment was under development by the Program for International Student Assessment (PISA). PISA is a worldwide study conducted by the Organization for Economic Co-operation and Development, with the purpose of informing development of educational policies worldwide. Its findings are used to investigate what causes difference in achievement across nations. As part of this assessment development effort, a workshop was organized for CSCL 2013 where Art Graesser, leader of the assessment development task force, presented the draft assessment plan, and a panel of experts, on which I was a speaker, offered feedback. This invitation signified that at that time my group’s work was beginning to contribute to change in thinking about international assessment practices.

Subsequently, I was invited to author the chapter on Discourse Analytics for two different handbooks, including the third edition of the International Handbook of the Learning Sciences and the International Handbook of Learning Analytics, where work on automated analysis of collaborative processes are featured as important areas of work. The primary focus of my pre-tenure work on interaction analysis was within pairs and small groups. As my post-tenure work has worked towards supporting collaborative interactions within online learning communities, I became aware of the need to understand how local interactions within pairs or small groups may lead to emergent behavior at the community level and how this affects individuals’ experiences and outcomes within a larger community context. Since my tenure review, the focus of my analytic work has targeted analysis of discussion in Massive Open Online Courses. 23 invited talks since my tenure review have focused on my work in a MOOC context, many of these integrating work across the three strands of my research. This includes both academic meetings and ones hosted by industrial MOOC platforms, such as edX and Coursera.

My discourse analysis work draws insights from rich theoretical models of interaction from sociolinguistics and discourse analysis, and operationalizes them in ways that capture the most important essence for achieving impact. This approach sets my work apart as distinctive among researchers in the Computational Linguistics community working on analysis of large scale social interaction. My recently accepted Computational Linguistics journal article that defines a vision for the emerging field of Computational Sociolinguistics describes the theories, methodologies, and modeling technologies that my work integrates, builds on and extends.

Carolyn Penstein Rose (cprose@cs.cmu.edu)/ Carnegie Mellon University