An important part of conducting sound science involves interpreting data correctly. Unfortunately, we don’t do that very well. For example, we are fooled by regression to the mean, we report findings when there are none, and we are overconfident about statistical power and significance.

As scientists and lay persons, we want to be certain about research findings. But statistics only gives information on probabilities, and how we intuitively interpret probabilities can lead us astray. This is problematic. As scientists, we need some competance in statistical rigor so that we don’t draw the wrong conclusions. But this can be a challenge. Furthermore, for some of us (myself included), the whole idea of “statistics” is scary when one dreads the lion called “mathematics”.

Thankfully, smart people have done the hard work of explaining difficult concepts simply. This series will summarize a few chapters on key concepts in biostatistics from the very readable book “Intuitive Biostatistics” by Harvey Motulsky. Specifically, we will cover:

1. From sample to population (Ch 3)
2. Confidence interval of a proportion (Ch 4) and what a “confidence interval” means
3. Quantifying scatter (Ch 9) and why the SEM does not quantify scatter in data
4. The Normal distribution (Ch 10)
5. Confidence interval of a mean (Ch 12)

We will aim to keep the summaries short, the language readable, and spend time on key concepts as needed. Hopefully this series contributes to the foundation on understanding key concepts of biostatistics, or serves as a good refresher.

Reference

Motulsky H (2018) Intuitive Biostatistics. A Nonmathematical Guide to Statistical Thinking. 4th Ed. Oxford University Press: Oxford, UK.