No abstract

sets an ambitious vision for science learning by emphasizing that for students to achieve proficiency in science they will need to participate in the authentic practices of scientists. To realize this vision, all students will need opportunities to learn from high-quality curriculum materials where they engage in science practices. We report on our study of a middle school curriculum called Project-Based Inquiry Science that has some design features that match well with the new directions in science education. To measure the impact of these materials, we conducted a randomized controlled trial in sixth grade science classrooms across 42 schools in an urban school district. We randomly assigned schools to either a treatment condition where teachers implemented the project-based science curriculum or a comparison condition where teachers implemented the district-adopted textbook. Teachers in both conditions received professional development on the Framework. Students who participated in the project-based science curriculum outperformed students in the comparison curriculum on outcome measures that were aligned to core science ideas and science practices in the Framework. Importantly, the results show that project-based curriculum materials that incorporate science practices along with disciplinary content can help students achieve next generation science learning outcomes when there is coherence with district guidance about instruction. The study findings suggest that curriculum materials, district involvement, and support for teachers' implementation of new forms of instruction are important for realizing the vision and key principles of the Framework in the context of a large and diverse urban school district.

This work investigates whether nonlexical information from speech can automatically predict the quality of smallgroup collaborations. Audio was collected from students as they collaborated in groups of three to solve math problems. Experts in education annotated 30-second time windows by hand for collaboration quality. Speech activity features (computed at the group level) and spectral, temporal and prosodic features (extracted at the speaker level) were explored. After the latter were transformed from the speaker level to the group level, features were fused. Results using support vector machines and random forests show that feature fusion yields best classification performance. The corresponding unweighted average F1 measure on a 4-class prediction task ranges between 40% and 50%, significantly higher than chance (12%). Speech activity features alone are strong predictors of collaboration quality, achieving an F1 measure between 35% and 43%. Speaker-based acoustic features alone achieve lower classification performance, but offer value in fusion. These findings illustrate that the approach under study offers promise for future monitoring of group dynamics, and should be attractive for many collaboration activity settings in which privacy is desired.

No abstract

No abstract