Krista Galyen

Network analysis of trace data for the support of group work: activity patterns in a completely online course

A 16-student, completely online software design course was studied using social network analysis and grounded theory techniques. Bi-directional (read and post) log data of user activity was recorded to understand how small group networks change over time with activity type (individual, peer-to-peer, and small group). Network structure was revealed through sociograms and triangulated with discussion board topics and interview data on group experience. Results show significant differences in network structure across activity types, which are supported by open coding and axial coding of the text of member discussions and editing patterns of member work products. It is also indicated that bi-directional log data, contextualized to specific activities and artifacts, revealed a more accurate and complete description of small group activity than ordinary, uni-directional log data would have. Our findings have implications for tool development revealing group structure and software design for completely online group work.

Connecting performance to social structure and pedagogy as a pathway to scaling learning analytics in MOOCs: an exploratory study

This exploratory study focuses on the design and evaluation of teaching analytics that relate social learning structure with performance measures in a massive open online course MOOC prototype environment. Using reflexive analysis of online learning trace data and qualitative performance measures we present an exploratory empirical study that: a rigorously evaluates a novel, multi-dimensional performance construct; b describes differences in small group dynamics and structure; and c draws a connection between learning performance and group structure. Performance is operationalized using a combination of knowledge construction measurement from discussion boards, rigorous analysis of student work products and several indicators of small group identity in the course examined. Interview and observational data are used to develop an approach for deriving and validating a model of the social structure of students in the course using traces of interaction data. The connection between performance and structure is developed at the small group unit of analysis. Implications for MOOC design, scaling MOOC analytics and a vision for developing social sensors in MOOC environments are presented in the discussion and conclusion.

Social Ability in Online Groups: Representing the Quality of Interactions in Social Computing Environments

We propose Social Ability as a construct that will contribute to the development of social computing models. Completely online group experiences vary according to the composition of tools, tasks and individuals. This conditionality creates challenges for social computing researchers who seek to model social behavior online. We use social ability as a frame to describe how members of completely online groups (COGs) experience the social nature of their interaction and how the nature of their social computing changes over time. We show how social ability measures for COG participants change during collaboration and describe how two social ability factors - Social Presence and Perceived Written Communication Skills - are related to participation and contribution in online group activities. The findings from this mixed methods study show that members who participate in online groups experience increases in perceived written communication skills and peer social presence, suggesting that completely online group work influences social ability.

Smart 3D collaborative virtual learning environments: A preliminary framework

With increases in access to powerful computing and high-speed networks, 3D virtual learning environments are being envisioned and developed as places for collaborative learning. These new environments for collaborative learning have promise for great authenticity in experience, great presence with others, and the monitoring/sensing of broad ranges of human cognition and behavior inferred from the actions of avatars. In the process of designing and building such an environment, iSocial, to develop social competency for youth with Autism Spectrum Disorder the authors have explored the potential for developing a smart system. The paper provides a framework for conceptualizing and implementing a smart 3D collaborative virtual learning environments based on 3 key constructs: environmental scaffolds, social affordances and coaching. The framework and constructs are illustrated using experiences and functionality developed for the iSocial system.

Distance Learning for Students with Special Needs through 3D Virtual Learning

iSocial is a 3D Virtual Learning Environment (3D VLE) to develop social competency for students who have been identified with High-Functioning Autism Spectrum Disorders. The motivation for developing a 3D VLE is to improve access to special needs curriculum for students who live in rural or small school districts. The paper first describes a number of innovative features developed to translate a targeted curriculum and to undertake teaching and learning to meet special needs in a 3D Virtual Environment. Secondly the paper describes results from a field test of iSocial at 3 sites with 11 students. The results show promise for improvements in social competence, but also challenges for distance learning delivery of 3D VLE.

Leveraging open source software and design based research principles for development of a 3D virtual learning environment

Design based research (DBR) has been acknowledged as a productive approach for advancing educational technology. Coincidentally, open source software has been found to be a good fit for implementing design based research. This report presents a case study of a software project using a design-based research approach and free/open source software. The project, iSocial, is developing a 3D virtual environment for youth with autism spectrum disorders to develop social competence. The study illustrates how the flexibility and community features of FOSS fit with the iterative nature of design-based research to benefit the development of iSocial, and the report also discusses challenges of implementing FOSS and DBR.