The research design for this project is longitudinal, comparative, descriptive and causal. This study uses mixed qualitative (e.g., field observations, videography, interviews, case studies) and quantitative methods using ANCOVA and regression approaches to analyze data from quasi-experimental studies based on within-class comparisons of treatment and quasi-experimental groups randomly assigned. In the experimental conditions teachers in groups will: 1. conduct mathematical investigations; 2. discuss and reflect on those investigations; and 3. through the Video Mosaic, study and discuss videos of multiple cases of children engaging in the same investigations at different stages of development. The comparison treatment will include components 1 and 2, but rather than having teachers in groups view videos of children’s developing reasoning, teachers will spend equivalent time engaged in another activity, such as studying only artifacts of students’ written solutions to the investigation, or viewing video not directly connected to the intervention problem set.
Instruments under development include a pre-post assessment instrument to measure participants’ ability to identify student reasoning from sample cases, a survey about course factors based on Bielaczyc's Social Infrastructure Framework, and a survey on teacher beliefs.