Reading path

Reading Statistics and Uncertainty in Medical Evidence

Statistics can feel like a wall built to keep you out, but a handful of ideas do most of the real work. This path walks you through what a p-value, a confidence interval, and an effect size actually mean, and why a result can be statistically significant without mattering to a patient. By the end you will be able to read the numbers in a study with a steady eye, and see honest uncertainty as a mark of careful work rather than a weakness.

The path, step by step

  1. 1 What a P-Value Really Means, and the Four Things People Think It Means but It Does Not

    5 min read

    Start with the most misunderstood number in medicine, so you know what a p-value does and does not claim before anything is built on top of it.

  2. 2 Statistical Significance Versus Clinical Importance: Why a Small P-Value Does Not Mean a Result Matters

    5 min read

    Once you know what significance is, the next lesson is that a small p-value can still describe a change too tiny to matter to a real person.

  3. 3 What a Confidence Interval Is Not, and How to Read It as a Measure of Precision

    6 min read

    Move from a single yes-or-no verdict to a range, and learn to read the interval as a measure of how precise or shaky an estimate really is.

  4. 4 How to Read Antidepressant Effect Sizes Without Overclaiming

    4 min read

    Now put a size on the effect itself, using a worked example that shows how to read how big a benefit is without inflating it.

  5. 5 Multiple Testing: Why Twenty Questions of One Dataset Produce a False Positive

    5 min read

    With the basics in hand, see how asking a dataset many questions manufactures false positives, the first way significance quietly misleads.

  6. 6 P-Hacking: The Hidden Choices That Manufacture Significance

    5 min read

    Follow that thread into the flexible choices researchers make, and how ordinary decisions can conjure a significant result from noise.

  7. 7 The Fragility Index: How Many Events Separate a Positive Trial From a Null One

    5 min read

    Test how sturdy a positive trial really is by asking how few events would flip it, a concrete way to feel the uncertainty behind a headline.

  8. 8 The Ethics of Uncertainty in Medicine

    5 min read

    Close by reframing everything above: acknowledging what we do not yet know is not a failure of the evidence but the honest core of reading it well.

Each step is a full article on the Reading the Evidence blog.

Back to all reading paths