Nat'l Academies Pre-print Report on Discipline-Based Education Research

This 270 page "synthesis study" collects the literature and assesses research on approaches to undergraduate science education.  I haven't read the report, just skimmed the 19-page Executive Summary. 

Most reports by the National Academies can be read online for free, including this one.  From the Exec. Summ.:

DBER scholars have devoted considerable attention to effective instructional strategies and to students’ conceptual understanding, problem solving, and use of representations. Key findings from DBER are consistent with cognitive science research and studies in K-12 education.

To gain expertise in science and engineering, students must learn the knowledge, techniques, and standards of each field. However, across the disciplines, the committee found that students have incorrect understandings about fundamental concepts, particularly those that involve very large or very small temporal and spatial scales. Moreover, as novices in a domain, students are challenged by important aspects of the domain that can seem easy or obvious to experts, such as problem solving and understanding domain-specific representations like graphs,
models, and simulations. These challenges pose serious impediments to learning.

DBER clearly shows that research-based instructional strategies are more effective than traditional lecture in improving conceptual knowledge and attitudes about learning. Effective instruction involves a range of approaches, including making lectures more interactive, having students work in groups, and incorporating authentic problems and activities.

To enhance DBER’s contributions to the understanding of undergraduate science and engineering education, the committee recommended:

• Research that explores similarities and differences among different student
  populations.
• Longitudinal studies—including studies of the K-12/undergraduate transition— to better understand the acquisition of important concepts and factors influencing retention.
• More studies that measure outcomes other than test scores and course performance, and better instruments to measure these outcomes.
•  Interdisciplinary studies of cross-cutting concepts and cognitive processes.