Making Sense of Statistics

A Guide to Understanding & Explaining Data

Fundamental Statistical Concepts

Statistics help us describe data and draw inferences. These are the building blocks.

Mean (Average)

The sum of all values divided by the number of values. Can be skewed by very high or low numbers (outliers).

Median

The middle value in an ordered dataset. It's often a better measure of the 'center' when outliers are present.

Standard Deviation

Measures how spread out the data points are from the mean. A small SD means data is clustered tightly.

Sample vs. Population

A population is the entire group of interest. A sample is a smaller, manageable subset of that population.

How We Make Claims from Data

Inferential statistics allow us to make educated guesses (inferences) about a whole population based on a smaller sample.

P-value & Hypothesis Testing

Researchers start with a 'null hypothesis' (a statement of no effect). The p-value tells you the probability of seeing your data if that 'no effect' idea was true. A small p-value (e.g., < 0.05) suggests the effect you saw is probably real and not just random chance.

Confidence Intervals (CI)

This gives a range of plausible values for the true effect in the whole population. A 95% CI means we're 95% confident the true value lies within that range. A narrow CI is more precise than a wide one.

The Most Important Distinction

Statistical Significance (p-value) just means an effect is unlikely to be a fluke. Practical/Clinical Importance (Effect Size) tells you if the effect is large enough to actually matter in the real world. A tiny, useless effect can still be 'statistically significant' in a large study!

Translate a Stat

Paste a statistical finding (e.g., from a paper's abstract) and let our AI helper translate it into plain English.

Risk Explainer

Headlines often use "Relative Risk" which can be misleading. Use this tool to see the real-world "Absolute Risk".

Communication Cheat Sheet

  • Focus on the 'So What?' - what does this number mean for people?
  • Use analogies. 'Unlikely to be a fluke' is better than 'statistically significant'.
  • Always provide the Absolute Risk alongside the Relative Risk to give context.
  • Explain the Confidence Interval as a 'range of plausible results'.
  • State the limitations. What does the study *not* tell us?