Illinois

Collaborative Research: Learning Linkages: Integrating Data Streams of Multiple Modalities and Timescales

Principal Investigator:

Bruce Sherin

Project Overview

Background & Purpose:

Our exploratory project aims to:

Collect diverse sources of data in the context of classroom enactments that make use of two computer-supported learning environments, ChemVLab+ (chemvlab.org) and the Carnegie Mellon Algebra Cognitive Tutor
Compare analyses of log-file data with analyses of integrated datasets to determine whether the use of the integrated data produces more robust models of student learning and proficiency, and to understand the limits of analyses of log data alone.
Develop new ways to represent and visualize information about student proficiency and engagement to make the knowledge actionable.

Setting:

Middle and high school classrooms in Pennsylvania.

Research Design:

The project uses a comparative research design and will generate evidence that is descriptive [case study, observational] and associative/correlational [interpretive commentary, educational data mining]. Original data are being collected on middle and high school students using assessments of learning and observation [personal observation, videography, Web logs], and survey research [self-completion questionnaire, semi-structured or informal interview]. Mathematics instruction is being designed to address the learning needs of children and will be adjusted on a week-by-week basis. The project is evaluating ChemVlab+ online activities and Carnegie Learning Algebra Tutor.

Instruments or measures being used include: (1) computer work, (2) whole-class discussions, (3) small group work, (4) individual assessments and seat work, and (5) individual interviews. These data will be collected in three modalities: (1) log files, (2) video and audio recordings, and (3) written artifacts. We will (1) conduct automated analyses of log data, and (2) perform qualitative analyses of the multiple data modalities to manually create what we call learning narratives.

Findings:

Findings will be posted as they become available.

Discipline:

State:

Target Population:

Research Design: