Developing computational means to perform microgenetic analysis
NSF Award #: DRL-1252350
Microgenetic learning analytics is a new research method we are developing with funds from the National Science Foundation. We are developing computational means for performing microgenetic analysis of discourse data. Microgenetic analysis seeks to understand conceptual change over short time periods (minutes, hours, days). Our discourse data corpus is derived from middle school aged girls working in collaborative problem solving groups to solve robotics problems. The challenge of our work is finding effective means for interpreting contextual discourse that widely features indexical and pronomial data, as opposed to a specific content-based vocabulary. Microgenetic analysis is viewed by educational researchers as one of the most robust methods for understanding how human learning happens. Microgenetic analysis is a labor intensive method that is usually performed in a case study format involving one or a few students. Our goal is to develop a method that will expand possibilities for performing microgenetic analysis over larger data sets, and to expand the scope of research questions that may be asked and answered with such data sets. Our video will provide information on the current state of our solution to this educational research problem using computational means.