Field entry, published 2026-11-03T00:00:00.000Z, updated 2026-11-03T00:00:00.000Z. Group: National variants, labelled as variants
How do national variants differ from ISCO-08 in detail?
| Major group | National variant, not ISCO-08 itself, see text |
|---|---|
| Skill level | See text, ISCO-08, ILO |
| Code ladder | general |
Classification only, not a job decision
Differences described here are about system design. Nothing here sorts a specific job into a variant group or into an ISCO-08 group.
National variants differ from ISCO-08 in detail even when they agree at broad level, because a variant is built to serve a country while ISCO-08 is built to serve comparison across countries. Broad agreement can hide detailed disagreement, and detailed disagreement is where most quoting errors happen. The direct answer is therefore twofold: expect similarity at the top of the hierarchy and divergence in the middle and at the bottom, and check the detail before assuming equivalence.
ISCO-08, published by the ILO, has 10 major groups, 43 sub-major groups, 130 minor groups and 436 unit groups. A variant may mirror that skeleton closely and still differ in almost every local title, because titles are where national life shows up most clearly.
What kinds of difference appear?
| Kind of difference | What it looks like to a reader | Why it matters for comparison |
|---|---|---|
| Finer splits | One ISCO-08 group appears as several national groups | Detail collapses on mapping, and cannot be recovered afterwards |
| Broader merges | Several ISCO-08 groups appear as one national group | International detail is lost before mapping starts |
| Renamed titles | A familiar title covers a different scope | Title matching alone misleads |
| Extra local groups | A group exists nationally with no neat ISCO-08 match | Mapping must record a best fit or a residual |
| Different boundaries | Supervisors, helpers or related roles sit elsewhere | Totals shift even when titles look stable |
| Language differences | A translated title suggests a scope it does not carry | Definition, not title, should govern |
These patterns are general, not claims about any one country, and country detail must be checked in that country’s publication.
Why do splits and merges happen?
Splits happen when a country needs to track a kind of work closely, for policy, training or survey reasons, and a single international group would hide the distinction that matters nationally. Merges happen when a distinction that matters internationally is too small, too rare or too hard to code reliably in national surveys to justify a separate group. Both moves are rational for national use, and both complicate international use in opposite ways.
A split creates a mapping problem of collapse. Several national groups map to one ISCO-08 group, so the mapped result is correct at international level but carries no trace of the national distinction. A merge creates a mapping problem of expansion. One national group spans several ISCO-08 groups, so mapping must either choose a best fit, split the total by an auxiliary rule, or report the result at a broader level where the merge no longer matters. Each choice should be documented, because each choice changes what the mapped figure means.
How do skill boundaries shift?
ISCO-08 organises groups by skill level and skill specialisation, considering task complexity, field of knowledge, tools, materials and outputs. A national variant may weight those factors differently, because its training system, licensing rules or workplace organisation draw lines in different places. A role that sits at one skill level in ISCO-08 may therefore sit near a different boundary nationally, not because one system is wrong, but because skill is judged through national institutions.
This shift is easy to miss when titles match. A reader who assumes that skill level travels with the title will carry a national skill judgement into an international table without noticing. The safer habit is to compare group definitions, not titles, and to note any skill difference in the method note for the table.
What about titles that look identical?
Identical titles are the most deceptive case. Two systems can use the same words for groups that include different tasks, exclude different neighbouring roles, or treat part time, seasonal or supervisory work differently. A title is a label for a definition, not a definition itself, and the definition is where scope lives.
When a title looks identical across systems, a careful reader slows down rather than speeding up. Check the definition in each system, check the examples given, and check the groups listed as included and excluded. If those checks cannot be done from the material at hand, the honest course is to present the two figures side by side with their system labels, and to avoid claiming that they measure the same thing.
How should detail differences be handled in practice?
Handle them by level. Use mapped comparison at the broadest level where the mapping is documented as reliable, and keep national detail in its national label where finer comparison is not supported. Record splits, merges and unmapped groups in the method note, rather than forcing every national group into an ISCO-08 box. A residual or unmapped category, openly reported, is more trustworthy than a forced match that hides uncertainty.
Above all, do not repair a detail difference by renaming it. A mapped national group remains a national variant group, because mapping is a bridge, not a change of identity, so the label should survive in the table, the chart and any later quotation.
Sources checked
Back to the National variants, labelled as variants group or browse all occupations.