This incredible resource, “It’s the Effect Size, Stupid” from University of Durham (UK) professor Robert Coe, is an excellent overview of effect size. It gets wonky, but even if you prefer to avoid the math and extra explanations, bookmark this link for the table that interprets effect sizes, allowing you to estimate a percentage effect (similar to relative risk) from Cohen’s d, for example. Also skim the Conclusions at the bottom.
The Psychiatry journal article “Estimating the Size of Treatment Effects” discusses five types of effect sizes: Cohen’s d (aka standardized mean difference), relative risk, odds ratio, number needed to treat and area under the curve.
One way to deal with attrition in a study is to do an intent-to-treat analysis. To better understand the rationale behind this method and its advantages and disadvantages, check out “Intention-to-treat concept: A review” and “The Intention-to-Treat Principle” from the JAMA Guide to Statistics and Methods.
This article “Interpretation of Cost-Effectiveness Analyses” from the Journal of General Internal Medicine looks at some of the challenges of doing cost-effective analyses and using QALYs, including the influence of different variables (such as funding sources).
This free article “Calculating QALYs, comparing QALY and DALY calculations” gets pretty technical with the math of how quality-adjusted life years (QALYs) and disability-adjusted life years (DALYs) are calculated, but for those wanting to know the nitty gritty details, it covers them well.
“Quality Adjusted Life Years,” a slideshow from Johns Hopkins Bloomberg School of Public Health, explains QALYs a bit more gradually and more accessibly.
Statistics How To is a blog and resource site that explains the basics on statistics and probability with an simple index of links and even a page of calculators where you can enter whatever raw data you have to see if you get the same results on a method the authors stated they used.
Each of these articles from Explorable and Statistics How To explain convenience sampling, why it might be used and its advantages and disadvantages.
To better understanding types of sampling in clinical trials, “Types of Samples” from University of California at Davis explains the difference between probability sampling and non-probability sampling as well as some examples of each.
Validity, reliability, and generalizability in qualitative research: To better understand generalizability in a study as well as how to assess the reliability and validity of study findings, this article, “Validity, reliability, and generalizability in qualitative research,” briefly discusses five published studies to illustrate how each of these concepts applies.
To better understand p-hacking, this Nature article dives into the possible statistical errors in research.
While knowing the five basic steps to a systematic review is helpful, this more in-depth article goes into detail on each of the steps.
The eight stages of systematic reviews and meta-analysis (done together) are outlined in detail in this article from the Journal of the Canadian Academy of Child and Adolescent Psychiatry.
Sensitivity and specificity can be challenging to understand, and this article clearly describes the differences between them and how they relate to false positives, false negatives, positive predictive value and negative predictive value. It also walks you through concrete examples.
Forest plot explained: This concise explainer of forest plots offers several helpful visual examples.
Deciphering a forest plot for a systematic review: This one-page diagram of a forest plot(PDF) identifies the key parts of it and provides a quick reference for journalists.
The Number Needed to Treat calculator provides journalists with a tool for figuring out the NNT even if it’s not reported in a study, as long as the study provides the raw data on outcomes (absolute risk instead of only relative risk).
Sometimes it’s helps to go through an actual lesson plan with sample problems to understand certain biostatistical concepts. “A beginners guide to interpreting odds ratios, confidence intervals and p values” is 20-minute tutorial over those three topics.
“Explaining Odds Ratios” explains what odds ratios are and what the mathematical formula is for them, including several illustrative examples.
Don’t understand the difference between incidence, prevalence, rates, ratios or other measures of disease occurrence? Check out this helpful cheat sheet from the University of Ottawa in Canada.
Statistics Glossary is an easy-to-use cheat sheet to help you remember what important statistical concepts mean, from the Dartmouth Institute for Health Policy and Clinical Practice.
Compendium of Primers is a collection of articles understanding statistics and statistical methods in medical research. It was originally published by now-defunct journal Effective Clinical Practice, a publication of the American College of Physicians.
The Cochrane Collaboration has put together this entertaining tutorial about P values and statistics.
“Epidemiology and how confounding statistics can confuse,” by Marya Zilberberg, M.D., M.P.H.
“News and Numbers, a Writers Guide to Statistics 3rd Edition,” by Cohn, Cope, and Runkel. Wiley-Blackwell. 2011
“Know Your Chances: Understanding Health Statistics,” by Woloshin, Schwartz, and Welch. University of California Press, 2008