Shedding Light on Pseudoreplication in Statistics for Life Science Students

Statistical errors in life sciences research may lead to inaccurate and misleading findings and therefore contribute to reproducibility concerns in science. One common error is pseudoreplication. This is a form of model misspecification where models assuming independent observations are fit to dependent data. In this paper, the impact of pseudoreplication is explored through results of recent simulation studies that were motivated by published life sciences studies. Implications for teaching and learning statistics in life sciences and future research are then discussed.