The Mode Is the Only Average That Works on Blood Types
Published 11/17/2026 · 3 min read · Everyday calculators
Data comes in kinds, and each kind supports fewer summaries than the last. Numbers allow a mean, a median and a mode. Ranked categories like small, medium and large allow a median and a mode but no mean, because there is no arithmetic between the labels. Unranked categories — blood types, eye colours, postcodes, favourite fruit — allow only the mode, since neither adding nor ordering means anything. The mode is the humblest of the three and the only one that survives all the way down, which is why survey software reports it for questions where the other two would be nonsense.
You cannot average A, B, AB and O, and you cannot rank them to find a middle. Counting which appears most often is the only summary left.
The mode can be missing or plural
A set where every value appears once has no mode at all, and a set where two values tie has two. Neither situation breaks anything, but both break software written by someone who assumed a single answer, which is a common source of quiet bugs in reporting tools. Numeric data almost always needs binning before a mode means anything — the most common exact height in a class of thirty is meaningless, while the most common five-centimetre band is informative. Choosing the bin width chooses the answer, and honest reports say which width was used.
Where the mode is the decision
A shop ordering one size of a garment orders the modal size, not the mean size, because it must sell actual items to actual people. A blood bank plans around the frequency of each type rather than an average. A factory setting a single machine for one product runs the most common specification. In each case the summary that gets used is the one that can be acted on, and a mean that lands between two real options is a number with nowhere to go.
| Data | Example | Usable summaries |
|---|---|---|
| Numbers | Height in cm | Mean, median, mode |
| Ranked labels | S, M, L, XL | Median, mode |
| Plain labels | Blood type | Mode only |
Frequently asked questions
- Can a mode be misleading?
- Easily, when the most common value is still rare. If a hundred people name a favourite fruit and apples lead with twelve votes, the mode is apples and 88% of the room disagrees. Reporting the mode without its frequency is the mistake: twelve out of a hundred is a very different statement from sixty out of a hundred, and both are described by the same word. The tool prints the count alongside the value for that reason.
- Why do textbooks teach three averages?
- Because they answer three different questions that ordinary language calls by one name. The mean answers what each would get if the total were shared out, the median what the middle person has, and the mode what is most typical. Asking which is correct is the wrong question; asking which one the decision needs is the right one. A salary negotiation, a stock order and a tax projection want different ones, and quoting the flattering one without saying which it is has been standard practice for a long time.
Articles you may find interesting
All guides →Related tools
Sources
Spotted a mistake in this article?