Instructors: Dav Clark, Javier Rosa & guests from industry, the social sector and academia
Oakland has many neighborhoods and schools that are underserved from the perspective of educational opportunities and resources. Many of the children in these neighborhoods will grow up in communities characterized by higher than average dropout rates, incarceration rates, violent events, etc.
To help change this trend, Rachel has helped start a program where Berkeley students will mentor young children in STEM topics in underserved schools in Oakland. To ensure this program is effective, we need to perform assessment of the impact the program has. This has potential to grow into a new research area.
Dav has a nascent research program with a Japanese tutoring company with over 20,000 kids, many of which are engaged in highly comparable curricula. Thus, we have the potential for MOOC-like scale, while still having more traditional “direct” access to students and teachers. SHO Zemi also has a remarkable dedication to evaluation and assessment within the teacher community.
Dav is also currently the maintainer of UC Berkeley’s EdX data. There’s a lot we could do with it, but the data ingest and management questions are quite challenging! We might work with this data as a use-case to round out privacy and other concerns in our data management approaches.
Develop measurement and data management approaches for educational interventions. A particular focus will be on assisting Prof. Rachel Slaybaugh with a project where UC Berkeley undergrads tutor second grade students in underserved schools on Oakland (DeCal course website). Additional topics may include improving UC Berkeley’s (and others’) handling of EdX data, and working with a Japanese tutoring company (SHO Zemi) with over 20,000 students. We will engage in design processes over the course of the semester, including assessment of available technologies (e.g., making use of whatever phones, etc. kids happen to have). Technology will likely include mobile approaches to experience sampling, adaptive / automated testing over the web, and perhaps sensor- or video-based affective measurement.