Emergency Bells in 17,131 Korean Toilets

A third of Korean public toilets have a bell on record

Korean public toilets increasingly carry an emergency call bell — a button that alerts staff or, in many installations, connects directly to police. It is common enough that Korea’s national open-data standard has a dedicated field for it. Across the 51,228 toilets in the national catalogue, 17,131 — 33.4% — have that field marked yes.

As with every field in this dataset, that is a registration rate. The column cannot be left empty, so an unanswered question is stored as “no”. We count only the explicit yes values, and a toilet with no entry has not been assessed as lacking a bell.

Bells sit between the two other safety-adjacent fields we have measured: accessible facilities are recorded at 41.3% of toilets, changing tables at 11.2%. The interesting part is that the provinces do not rank the same way across the three.

Emergency bell registration by province

Jeju tops the table at 70.4% — more than seven in ten of its toilets document a bell. Gyeongbuk sits last at 25.3%. That is a 2.8× spread, narrower than the 3.9× we found for accessible facilities and far narrower than the 6.8× for changing tables.

The rankings do not agree with each other

This is where the three fields become more useful together than apart. If some provinces were simply better at open data, the rankings would broadly match. They do not.

Jeonbuk is the clearest case. It is last of fifteen on accessible facilities and fourteenth on changing tables — but climbs to ninth on emergency bells, comfortably mid-table and ahead of Gyeonggi, Gangwon and Gyeongbuk. Whatever is going wrong with Jeonbuk’s record-keeping is not going wrong uniformly.

Sejong runs the other way: first on accessible facilities, sixth on bells. Gyeonggi is fourth on accessible facilities and fifteenth on changing tables. Across all three fields, only Jeju is consistently near the top.

The practical conclusion is the same one we keep arriving at from different directions: completeness is a property of a field, not of a province. Judging a dataset by one well-filled column tells you almost nothing about the next one.

Rate and volume point in opposite directions

Registration rates answer “how completely is this province documented”. They do not answer “where are the bells”. Those two questions have almost opposite answers here.

Gyeonggi records 3,489 bells — twice as many as anywhere else and 4.5 times Jeju’s total — while sitting 13th of 15 on rate. Jeju tops the rate table with 774. The four provinces with the most bells on file are all in the bottom third by completeness, simply because they are large.

Which number matters depends entirely on your question. If you are asking whether a province’s open data can be trusted, use the rate. If you are asking where you are likely to find a bell, use the volume — and remember that both are floors, because unrecorded is not the same as absent.

What an emergency bell in a Korean toilet actually is

If you have not seen one, it is usually a red or yellow button mounted on the cubicle wall at around chest height, sometimes with a pull cord instead. Installations vary: some ring an alarm locally, some alert a facility manager, and many in stations and parks connect to a monitored line.

They were rolled out substantially in response to safety concerns in isolated public toilets, and their presence in the national data standard reflects that policy priority. That is worth knowing when you read the registration rates: this is a field the government specifically wanted collected, which makes a 25% return in some provinces more notable than it would be for an incidental attribute.

Seoul offers the useful counter-example, though it has to be read carefully. Its separately maintained dataset records a bell at 2,392 of 4,655 toilets — 51.4%, well above the 33.4% national figure and higher than every province except Jeju and Ulsan. But Seoul’s file is curated location by location rather than bulk-submitted, and it registers every field more completely, not just this one. The honest reading is that Seoul documents its toilets better, which is not the same claim as Seoul equipping them better.

• Do not press one to report a broken tap. It is an emergency channel, and in many installations it reaches police.
• If you are in trouble and cannot find a bell, the emergency numbers are 112 (police) and 119 (fire and ambulance). Both handle calls in English, and 119 covers medical emergencies.
• Subway station toilets are the most consistently equipped, and are staffed during operating hours.
• An unlabelled toilet is not an unequipped one. In Gyeongbuk, Gangwon and Gyeonggi the field is left blank about seven times in ten.

One habit worth forming, and it costs nothing: look for the button when you go in, not when you need it. Installations are not standardised in position — wall-mounted at chest height in most newer builds, but low near the floor in some, and occasionally a pull cord rather than a button. Ten seconds of orientation on the way in is the entire preparation this requires, and it is the kind of thing nobody does until the first time they wish they had.

Sources and method

• Primary source: 전국공중화장실표준데이터 — Nationwide Public Toilet Standard Data, Ministry of the Interior and Safety. Emergency bell installation is one of the standard’s defined fields.
• Counts taken from our own catalogue on 6 August 2026: 51,228 records, of which 31,529 arrive from the national standard dataset and the remainder from provincial and municipal feeds using the same schema.
• Seoul’s 4,655 records come from the Seoul Open Data Plaza and are reported separately rather than merged, because the two datasets are collected differently.
• A toilet counts as having a bell only where the field is explicitly marked yes.