The capacity of laboratory animal research to be translated to interventions for human health is dependent on research reproducibility, replicability, and generalizability (RRG). It is often the case that challenges with RRG arise from differences in scientific protocols, unknown or uncontrolled confounding factors, incomplete communication of scientific protocols, and unintentional misapplication of statistical principles, rather than the oft-discussed research misconduct. Many of these challenges with RRG involve practical and theoretical aspects of statistics and study design that go beyond what is taught in basic statistics courses or textbooks. We therefore propose creating a set of animated vignettes to educate early career, laboratory animal investigators (namely, graduate students, postdoctoral fellows, and beginning investigators) about these issues. We specifically propose modules on variability, power, randomization, hypothesis testing, and inferences that address common, and yet not routine-textbook needs of laboratory researchers. Each module will be composed of several short, animated vignettes that will 1) discuss a statistical or design principle that is important for RRG; 2) foster dialogue and collaboration between statistical and laboratory animal scientists; and 3) model diversity of statistical and laboratory animal scientist with respect to expertise and demographic characteristics. Each vignette will also be evaluated for content validity, face validity, and educational value by consulting with statistical experts, experienced lab animal researchers, and early career investigators, respectively. Finally, vignettes will be reinforced with additional online instructional content, including tutorial literture and self-assessment quizzes. Our team will widely disseminate the instructional materials leveraging our experience and resources creating and sharing online educational content, and we commit to maintain the materials in an openly available web portal without cost to end users. By supporting principles of RRG in this way, we contribute broadly to the mission of the NIH in fostering fundamental creative discoveries, innovative research strategies, and their applications; developing, maintaining, and renewing scientific human resources; and exemplifying and promoting the highest level of scientific integrity in the conduct of science. We specifically contribute to the mission of the NIGMS by training the next generation of scientists, in enhancing the diversity of the scientific workforce, and in developing research capacities throughout the country.
Improving the reliability of published laboratory animal research is imperative to being able to translate basic science to interventions for human health. The proposed training modules will support this mission by creating educational materials that address commonly encountered but unexpectedly complex statistical and research design issues.
| Brown, Andrew W; Kaiser, Kathryn A; Allison, David B (2018) Issues with data and analyses: Errors, underlying themes, and potential solutions. Proc Natl Acad Sci U S A 115:2563-2570 |
| Kroeger, Cynthia M; Allison, David B; Thomas, Diana M et al. (2018) TO THE EDITOR. Spine (Phila Pa 1976) 43:E492-E493 |
| George, Brandon J; Beasley, T Mark; Brown, Andrew W et al. (2016) Common scientific and statistical errors in obesity research. Obesity (Silver Spring) 24:781-90 |
| Goldsby, TaShauna U; George, Brandon J; Yeager, Valerie A et al. (2016) Urban Park Development and Pediatric Obesity Rates: A Quasi-Experiment Using Electronic Health Record Data. Int J Environ Res Public Health 13:411 |
| Brown, Andrew W; Li, Peng; Bohan Brown, Michelle M et al. (2015) Best (but oft-forgotten) practices: designing, analyzing, and reporting cluster randomized controlled trials. Am J Clin Nutr 102:241-8 |
