CURE-like, not cure-all: Varying broad relevance in experimentation labs produces similar student outcomes

Problem Statement
Physics education reform aims to modernize undergraduate curricula by engaging students in authentic experimental practices, moving away from "traditional" verification labs toward "experimentation-based" formats. A prominent model for this is the Course-Based Undergraduate Research Experience (CURE), which incorporates five key components: use of scientific practices, discovery, broad relevance, collaboration, and iteration. While CUREs are shown to benefit student outcomes (such as persistence and identity), they are resource-intensive and difficult to implement in physics. The primary barriers are the requirements for "discovery" (novel results unknown to the instructor) and "broad relevance" (results of interest to the wider scientific community). These requirements necessitate a continuous pipeline of novel, publishable research projects, which is rare in introductory physics.

A critical gap in the literature is the lack of evidence isolating which specific CURE components are essential for student gains. It remains unclear whether a "CURE-like" course that maintains scientific practices, discovery, collaboration, and iteration—but relaxes the "broad relevance" requirement (i.e., the work is not directly relevant to external researchers)—can produce student outcomes comparable to a full CURE or standard experimentation-based labs.

Methodology
The authors conducted a quasi-experimental study over two semesters in a large introductory experimental physics course at Cornell University (N = 1,114 students). The study contrasted two lab conditions that shared identical learning goals, instructional frameworks (open-ended, iterative, collaborative), and levels of "discovery" (questions unknown to students, often unknown to instructors, but not necessarily publishable). The conditions differed systematically in "broad relevance":

1. SALT (Skills-based Lab): A standard experimentation-based lab using equipment typical of introductory physics (springs, pendulums, circuits). The concepts and apparatus were not relevant beyond the classroom.
2. PEPPER (Particle Physics Lab): A "CURE-like" lab using tabletop cosmic ray muon detectors (adapted from MIT CosmicWatch). The concepts and apparatus were applicable to contemporary experimental particle physics research, though the specific student results were not directly relevant to the broader physics community. In Semester 2, "partner particle physicists" were introduced to explicitly communicate the relevance of the work to the scientific community.

Data Collection and Analysis
Student outcomes were measured via pre- and post-semester surveys assessing five constructs:

• Experimental Critical Thinking: Measured using the Physics Lab Inventory of Critical Thinking (PLIC).
• Attitudes and Identity: Measured via self-efficacy, perceived agency, belonging, and recognition.

To address missing data (ranging from ~15% to 40% in post-semester surveys), the authors employed multiple imputation (m=50) using the mice R package with predictive mean matching. They utilized hierarchical linear modeling (HLM) to compare student outcomes between the two lab conditions, accounting for clustering by section and semester.

Key Results
The study found no statistically significant difference in post-survey scores between students in the SALT (low broad relevance) and PEPPER (high broad relevance) conditions across all five measured constructs:

• Experimental critical thinking skills (PLIC)
• Self-efficacy
• Perceived agency
• Belonging
• Recognition

The results held true across demographic groups, with no evidence of differential outcomes based on the varying relevance of the lab content.

Significance and Claims
The paper argues that increased levels of "broad relevance"—defined as work having impact and action beyond the classroom and fitting into a broader scientific endeavor—may not inherently improve gains in student learning or attitudes when compared to well-designed experimentation-based labs that lack this specific component.

The authors claim that their findings challenge the notion that CUREs are uniquely positioned to achieve superior student outcomes over other well-designed experimentation-based labs. They suggest that the "broad relevance" component, while valuable, may not be a necessary condition for improving student outcomes in introductory physics. Consequently, relaxing the requirement for results to be directly relevant to external researchers could substantially lower the barriers to implementing "CURE-like" courses, making authentic scientific practices more scalable and accessible without sacrificing student learning gains. The study concludes that future work should focus on identifying which specific components of experimentation-based labs are most strongly associated with student gains, rather than assuming the full CURE model is required.