Back

Variation in CURE instruction and limitations of CURE studies undermine what can be concluded about CURE effects on student outcomes

Lantz, W. N.; Dolan, E. L.; Barvalia, S.; Apraku, S.; Kandru, H.; Senthamizh Selvan, M.; Majewska, A. A.

2026-08-05 scientific communication and education
10.64898/2026.08.04.741794 bioRxiv
Show abstract

Research suggests that students realize a range of cognitive, affective, and motivation-related outcomes by participating in course-based undergraduate research experiences (CUREs). However, the overall effects of CUREs on student outcomes are unclear. We sought to address this knowledge gap by embarking on a meta-analysis of CURE studies using the Population, Intervention, Comparison/Control group, and Outcomes framework. We conducted a literature search to produce an analytic sample of 353 studies of the effects of CUREs instruction on undergraduates. We characterized the CUREs being studied in this sample, including their dose, duration, and design. We also examined each study to determine if outcomes were measured before and after the CURE and with a comparison or control group so that effects could be attributed to CURE. We found substantial limitations in studies of CUREs, which precluded meta-analysis. CUREs varied widely and were unevenly described, making it difficult to determine what could or should constitute CURE instruction as an intervention. Many outcomes were studied, often without a comparison or control group and with inconsistent measurement and reporting. Our results indicate that advancements are needed in the design and reporting of research on CUREs to gauge their effects on student outcomes. Highlight SummaryMany studies of CURE instruction aim to show effects on student outcomes. Yet, systematic review of these studies reveals limitations in study design, methods, and reporting that preclude meta-analysis and limit any cross-study conclusions.

Matching journals

The top 1 journal accounts for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.