Beyond P-values and means: Statistical appreciation in the life sciences

Statistical literacy is frequently treated in undergraduate and graduate life science training as a technical competency peripheral to experimental practice rather than as a central component of scientific reasoning. This separation is especially problematic in life sciences and related disciplines, where biological variability is intrinsic to living systems and often gives rise to heterogeneous, skewed, multiplicative and non-normal data. Consequently, the interpretation of biological measurements depends not only on the data themselves but also on how variability, distributions, measurement scale, uncertainty and inference are understood. This perspective argues that stronger statistical appreciation is urgently needed in the life sciences, particularly during early scientific training. Emphasis is placed on the biological origins of variability, the interpretation of distributions, the importance of multiplicative thinking and lognormality, the appropriate use of means, medians, geometric means and logarithmic transformations, and the distinction between variability, uncertainty and effect size. The article further discusses replication, outliers, the limitations of significance testing and why statistical thinking should be embedded within bioscience education as an integral part of scientific reasoning rather than taught as an isolated analytical skill.