Methodology — The Culture Shock Index

This index has been rebuilt twice. You are reading wave 3 (6 September 2026). Wave 1 measured twelve everyday-infrastructure surprises on two axes (how unexpected, how good) and would have published a ranking once 50 visitors had rated six or more items — a threshold we had already lowered from 200, and said so. It closed with zero qualifying responses. Three independent attempts to seed it from outside data (search volume, search entities, a 56-document media-mention corpus) failed at the same niche items, because no external source covered all twelve.

Wave 2 replaced those twelve items with the eleven infrastructure categories of the national International Visitor Survey, because that survey measures all of its own categories. It lasted one day. The items were wrong for the question this index asks. Satisfaction with immigration or accommodation is a judgement about something you already expected; culture shock is about what you did not. Those eleven items and their measured satisfaction were not discarded — they moved to their own index, the Travel Infrastructure Index, where the name matches what the numbers say.

Wave 3 is the original twelve items, ranked by visitor votes alone. There is no official-statistics panel here, and there will not be one until a source covers all twelve — publishing only the items that happen to have data would be picking the ranking after seeing it. Wave 2 ended with zero votes, so nothing was merged or lost.

What this measures
“Which of these surprised you most about Korea?” — What visitors to this site actually pick when asked the question above — nothing more. It is a running poll, not a quality rating: the ranking is the count of picks.
Who answers
Visitors to this site, unpaid and self-selected — a convenience sample, not a probability sample. The question is about everyday decisions in Korea, so respondents skew toward people living in or engaged with Korea, and the results carry that skew.
Where the options come from
The twelve items frozen in `docs/item-naming-decisions.md` (18 August 2026), unchanged — the same list as the twelve articles in the hub's Reading section, derived mechanically from `items.ts` so the two cannot drift apart. This index has no measured panel: three independent attempts (search volume, search entities, a 56-document media-mention corpus) confirmed that no external statistic covers all twelve items, and publishing only the items that happened to work would be selection after seeing the results. The ranking is one hundred percent visitor votes.
When the ranking opens
Tallies are live from the first vote, with the vote count always shown. Rank numbers appear only once the poll has collected 100 votes in total — a figure fixed before any data existed. Below that, the tallies are early signal, not an order, and the page says so.
Formula
Plain vote counts, shown with each option's share of the total. No weighting, no smoothing, no editorial adjustment.
Editorial Top 5 shown before the poll
On 24 September 2026, Rankorea began showing a reviewed editorial Top 5 above the poll on published topics. It is selected by the Rankorea editorial team under the stated criteria, not calculated from votes. Seeing it first can affect later choices, so votes before and after that date must not be read as a preference change alone. Editorial order never enters the tally, rank, or tie handling.
Options that are not on the list
Visitors can write in what they would have picked. Write-ins never enter the ranking — an "other" bucket would collect every stray preference and float to the top. They are stored separately, only their count is public, and once 20 or more say the same thing, that option becomes a candidate for the next wave. A new wave starts a fresh tally rather than merging into this one, and is announced here.
The official-statistics panel
Where the official statistic that gave us the option list also ranks those same options, we show that ranking above the poll, with its source — measured, not voted. It is never blended into the poll; the two are kept apart so you can see where visitors here differ from the population at large.

Limits worth knowing before you cite this