01 /
Begin with the population definition
Korea Data's population pages use a resident-registration series. Read a change in that series as a change in registered population under its stated coverage. The Ministry of the Interior and Safety describes the register as including residents, people with unknown domicile, and registered Korean nationals residing abroad; foreign nationals are excluded. It is therefore not interchangeable with every other published measure of Korea's population.
A census, a population estimate, and a registration table can use different coverage and reference dates. Do not subtract a census total from a registration total and call the result annual growth. Start with Korea Data's national population page or a regional profile, confirm the source and unit, and keep the same series for both ends of your comparison.
02 /
Choose month-to-month or year-to-year deliberately
A monthly change compares a selected month with the previous month. A year-over-year change compares it with the same month one year earlier. A value labeled latest refers to the latest available observation, not necessarily the current calendar month. Record the two observation dates before calculating anything.
The difference between the two population counts is the absolute change. To express that difference as a rate, divide by the earlier population and multiply by 100. Keep the earlier count as the denominator in both positive and negative cases. Reversing the dates changes both the interpretation and the rate's base.
Hypothetical monthly and annual comparisons
An invented region has 100,000 residents in September of Year 1, 101,500 in August of Year 2, and 102,000 in September of Year 2. September's monthly change is +500, or about +0.49%. Its year-over-year change is +2,000, or +2%. Neither is a forecast for the following year.
03 /
Read the count alongside the percentage
A small region can have a large percentage change from a relatively small number of people. A large region can gain more residents while showing a lower rate. Use Population rankings to understand scale and Population growth to inspect the displayed year-over-year change, then return to the relevant profiles to see the underlying periods and counts.
The table below illustrates why the fastest-growing region and the region adding the most people need not be the same. If your question concerns additional demand in absolute terms, the number of people matters. If it concerns growth relative to the starting population, the rate matters. Neither alone establishes demand for a particular service or explains who moved.
| Region | Earlier population | Later population | Change |
|---|---|---|---|
| Small region | 10,000 | 10,500 | +500 / +5% |
| Large region | 500,000 | 505,000 | +5,000 / +1% |
04 /
Use additional evidence to explain the change
A population total records an outcome. By itself it cannot separate births, deaths, people moving in or out, and registration adjustments. A growing total does not demonstrate that the birth rate increased, and a declining total does not demonstrate that everyone leaving moved to the same nearby city. Those explanations require the relevant component data with compatible coverage.
Administrative corrections can be material. In an official KOSIS response about March 2021, Statistics Korea explained that a review of long-term unknown-domicile registrations removed people from the register and contributed to a sharp change in the oldest age group. That documented example is a reason to read the notes around a break, not a reason to assume every unexpected movement is a correction.
Boundary changes create another complication. If a district is transferred or a municipality merges, before-and-after counts may describe different areas. Consult the geography notes and the administrative-region guide before interpreting a sudden jump as migration. Preserve any explanation supplied by the statistical source rather than filling the gap with a plausible story.
05 /
Handle baselines and aggregation carefully
A percentage is undefined when the earlier population is zero. Missing data are also different from zero and should not be silently used as a baseline. For very small groups, report the count as well as a rounded rate so that the percentage does not exaggerate the scale of the movement.
When combining non-overlapping regions, add their earlier counts and later counts before calculating a combined rate. An ordinary average of regional percentages gives a small county the same weight as a large city. Do not include a province total and its constituent areas in that sum, because the residents would be counted twice.
06 /
Save a claim that another reader can check
A useful note names the region, series, dates, earlier and later counts, absolute change, and calculated rate. For the invented September example, write that registered population rose from 100,000 to 102,000 between the two Septembers, an increase of 2,000 or 2%. Keep this separate from any proposed explanation.
Include the source link, the date you consulted it, and any revision or boundary note. If an observation is later revised, you can identify why your earlier result differs. Before sharing a chart, make sure its title says monthly or year-over-year change, its unit is clear, and every region uses the same comparison window.
Sources and further reading
These references support the definitions and limits discussed in this guide. Worked scenarios labeled as examples are illustrative.
- MOIS: resident-registration population statisticsDefines registered-population coverage, including the exclusion of foreign nationals, and provides population-change categories.
- KOSIS: explanation of the March 2021 register changeThe official response explains how registration cleanup, in addition to births and deaths, can affect monthly counts. It is a historical example, not a current population figure.
- KOSIS: differences in foreign-population coverageAn official explanation of why census and administrative foreign-population counts differ in date and coverage; supports checking definitions before combining series.