Interactive-engagement (IE) techniques consistently enhance conceptual learning gains relative to traditional-lecture courses, but attitudinal gains typically emerge only in small, inquiry-based curricula. The current study evaluated whether a "scalable IE" curriculum-a curriculum used in a large course (∼130 students per section) and likely adoptable by a wide range of physics departments-could produce significant attitudinal benefits relative to a traditional-lecture curriculum. This study included data across three years, 10 instructors, over 30 sections, and over 1100 students, and our analytic strategy allowed us to isolate the effects that were due to the curriculum itself rather than other potential factors such as instructor differences or preexisting differences among students. Results revealed that our Active-Physics curriculum, which is based on Moore's Six Ideas That Shaped Physics, produced significant attitudinal and conceptuallearning benefits relative to our traditional-lecture physics curriculum. Further, the Active-Physics curriculum, for the most part, benefitted males and females equally, and relative to the Fall semester alone, the benefits of Active Physics became more robust when viewed across the entire two-semester sequence of introductory physics. Our data highlight that some (though not all) of the attitudinal benefits of small, inquiry-based courses may be achievable in larger course with scalable IE curricula that can potentially reach a large proportion of introductory physics students.

Increasingly, studies are investigating the factors that influence student discourse in science courses, and specifically the mechanisms and discourse processes within small groups, to better understand the learning that takes place as students work together. This paper contributes to a growing body of research by analyzing how students engage in conversation and work together to solve problems in a peer-led small-group setting. This qualitative study evaluates video of Peer-Led Team Learning (PLTL) sessions in general chemistry, with attention to both the activity structures and the function of discourse as students undertook different types of problems across one semester. Our findings suggest that students talk their way through the problems; practicing a combination of regulative and instructional language to manage the group dynamics of their community of peer learners while developing and using specific disciplinary vocabulary. Additionally, student discourse patterns revealed a focus on the process of complex problem-solving, where students engage in joint decision-making by taking turns, questioning and explaining, and building on one another's ideas. While students in our study engaged in less of the deeper, meaning-making discourse than expected, these observations about the function of language in small-group learning deepens an understanding of how PLTL and other types of small-group learning based on the tenets of social constructivism may lead to improvements in science education, with implications for the structure of small-group learning environments, problem design, and training of peer group leaders to encourage students to engage in more of the most effective discourse in these learning contexts.

Student attitudes, defined as the extent to which one holds expertlike beliefs about and approaches to physics, are a major research topic in physics education research. An implicit but rarely tested assumption underlying much of this research is that student attitudes play a significant part in student learning and performance. The current study directly tested this attitude-learning link by measuring the association between incoming attitudes (Colorado Learning Attitudes about Science Survey) and student learning during the semester after statistically controlling for the effects of prior knowledge [early-semester Force Concept Inventory (FCI) or Brief Electricity and Magnetism Assessment (BEMA)]. This study spanned four different courses and included two complementary measures of student knowledge: late-semester concept inventory scores (FCI or BEMA) and exam averages. In three of the four courses, after controlling for prior knowledge, attitudes significantly predicted both late-semester concept inventory scores and exam averages, but in all cases these attitudes explained only a small amount of variance in concept-inventory and exam scores. Results indicate that after accounting for students' incoming knowledge, attitudes may uniquely but modestly relate to how much students learn and how well they perform in the course.