Sean Kelly - Research and Grants

The project advances cyber-enabled, teacher analytics as a new genre of technology that provides automated feedback on teacher performance with the goal of improving teaching effectiveness and student achievement. The innovation aims to help teachers develop expertise on multiple dimensions of classroom communication and will be developed and tested in 9th grade classrooms in Western Pennsylvania. The team will first generate initial insights on how teacher discourse predicts student achievement via a re-analysis of large volumes (128 hours) of existing classroom audio. Next, they will design and iteratively refine hardware/software interfaces for efficient, flexible, scalable audio data collection by teachers. The data will be used to computationally model dimensions of effective discourse by combining linguistic, discursive, acoustic, and contextual analysis of audio with supervised and semi-supervised deep recurrent neural networks. The model-based estimates will be incorporated into an interactive analytic/visualization platform to promote data-driven reflective practice. After refinement via design studies, the impact of the innovation on instructional improvement and student literacy outcomes will be evaluated in a randomized control trial.

The main purpose of this project is to explore teaching practice through the lens of classroom composition. In particular, we will investigate how teaching practice affects student learning and engagement differentially depending on the level and dispersion of students’ initial achievement. We will explore how teachers adapt to different classroom compositions and whether these adaptations are productive in improving student outcomes. The research will have important implications for educational inequality, as teaching practice and classroom composition are dimensions of schooling that are both impactful and unevenly distributed.

A grant to develop and validate CLASS 5.0, a software system for the analysis of classroom discourse. This grant builds off of my previous research with Martin Nystrand at the National Research Center for English Learning and Achievement. We are working with artificial intelligence researchers to develop a voice-recognition and automatically coding version of CLASS and study its use in assessing teachers’ enactment of dialogic instruction. Note: the total funding for this grant to all participants is $1,599,828