Clicked Gallery

What's the Difference Between an Odds Ratio and a Relative Risk?

Highlighted from a real textbook passage. Explained by Clicked.

Used in a sentence

University Course Reader · STEM

The paper reported an odds ratio of 2.7, and the press release turned it into nearly three times as likely, which the relative risk of 2 never said.

The reader highlighted one term — in an article or a paper PDF. Clicked made the statistics term “odds ratio” easy to understand:

Explained in three depths

Same facts, different vibe — Slang mode 😎

The Clicked way

●○○

Overview

An odds ratio and a relative risk both say how much more likely something is in one group than another, and they are not the same number. Relative risk compares plain chances, so a relative risk of 2 means twice as likely. An odds ratio compares for-versus-against counts, the way a bookmaker quotes a race. Reading an odds ratio as times more likely overstates the effect whenever the outcome is common, and headlines do it constantly.
●○○

Overview

One comparison, two honest numbers, and the bigger one gets quoted. Relative risk speaks in probabilities, where 2 is a straight doubling, end of story. An odds ratio speaks in wins-and-failures, a 1-in-5 shot becomes 1 to 4, and ratios of those run hot when the event is everyday stuff. A headline that swaps one for the other has quietly upgraded the finding. 😎

A quick take — often all you need.

●●○

Detail

Relative risk and the odds ratio are two ways of comparing how often something happens in two groups. Relative risk divides the two chances. An odds ratio divides the two odds, and odds count for-versus-against, so a 1-in-5 chance is odds of 1 to 4. Say 40 people in 100 get an upset stomach on a new drug, against 20 in 100 without it. The chances are 40% against 20%, so the relative risk is 2, twice the risk. The odds are 40-to-60 against 20-to-80, and dividing those gives an odds ratio of 2.7. Same trial, same stomachs, both numbers true, and quoting the 2.7 as times more likely inflates a doubling into nearly a tripling. Shrink the outcome to something rare, 4 stomachs in 100 against 2, and the relative risk is still 2 while the odds ratio falls to 2.04. Researchers still lean on odds ratios because some study designs cannot make anything else. One familiar design gathers 100 people who already have a disease and 100 who do not, then looks back at their exposures. Those totals were the researchers' own choice, so nothing in the data says how often the disease strikes. Only the odds within each group survive, and their ratio is all such a study can report. So when a claim says times more likely, check which number sits underneath it before repeating it.
●●○

Detail

Relative risk is the normal-person number. Take each side, work out the percentage the thing happened to, and divide one by the other. An odds ratio splits each side into happened-or-not first, then compares those. Watch them drift apart. Say 75 players in 100 rage-quit game A, and 25 in 100 rage-quit game B. As percentages, 75% next to 25%, so the risk tripled. As odds, it is 75-to-25 on one side and 25-to-75 on the other, which is 3.0 and 0.33, and dividing them comes out at 9. Nobody's chance multiplied by 9. The true jump is triple. Yet 9 is what lands in the press release. The two versions tell one story when the thing is rare, roughly under 1 person in 10 in the comparison crowd, and drift this far apart when it is common. Two traps worth carrying out of here. Papers say risk for whatever they are counting, even the risk of recovering, so the word alone settles zero. And a doubled risk can stay tiny, since 1 in a million doubled is 2 in a million. So when a headline hands you a big multiplier, ask which of the two it came from. 😎

Want more? One click digs deeper.

●●●

Analogy

Two goalkeepers face 10 penalties each. The first saves 8 and lets in 2. The second saves 5 and lets in 5. In chances, that is 80% against 50%, so the first keeper is 1.6 times as likely to make a save. In odds, it is 8-to-2 against 5-to-5, which is 4 against 1, so his odds of saving are 4 times higher. Ten kicks each, and both sentences are true. Quote the 4 as four times as likely to save, and you have invented a keeper nobody actually watched. The gap is this wide only because saves are the norm. If saves were scarce, the two numbers would sit almost on top of each other.
Deepest level reached.
●●●

Analogy

Two airlines, and suppose you flew 1,000 flights with each. One loses your bag once, its rival twice. As probabilities, 0.2% and 0.1%, double the risk. As odds, 2-to-998 for the careless airline and 1-to-999 for the careful one, and dividing the careless odds by the careful odds gives 2.002, the odds ratio. When a thing almost never happens, the flights where nothing went wrong are nearly the whole pile, so the two versions land on the same answer. On a rare outcome like this one, the odds ratio really does mean times as likely. On an everyday outcome, it overshoots. Either number here says switch airlines. 😎
Deepest level reached.

Unfamiliar concept? A real-world example makes it click — fresh analogies on tap.

AI explanations may contain errors · Not professional advice

Formal definition — The same term, explained the usual way

Relative risk (risk ratio) is the ratio of the probability of an outcome in an exposed or treated group to the probability in a comparison group. The odds ratio is the ratio of the corresponding odds, where the odds of an outcome are the probability of its occurrence divided by the probability of its non-occurrence. The two measures approximate one another when the outcome is uncommon and diverge as outcome prevalence rises, with the odds ratio further from unity than the relative risk. Odds ratios arise naturally in case-control designs, which sample on outcome status and therefore cannot estimate outcome probabilities directly.

Want Clicked to explain terms like “odds ratio” directly in your browser — including on PDFs?

Add to Chrome — Free

50 free Explanations · No credit card required