Colombia: 1973
France: 1962, 1968, 1975, 1982, 1990
Kenya: 1989
Mexico: 1960, 1970, 1990, 2000
United States: 1960, 1970, 1980, 1990
Vietnam: 1989, 1999

Colombia 1973: Persons age 10+, with a job or experienced unemployed.
France 1962-1968: Persons age 14+, in the labor force.
France 1975: Persons age 16+, in the labor force.
France 1982-1990: Persons age 14+, in the labor force.
Kenya 1989: Persons age 10 or more who worked for pay or profit, on leave/sick leave, or working on family holdings during the last seven days.
Mexico 1960: All persons.
Mexico 1970: Persons age 12+ who worked the previous year.
Mexico 1990-2000: Persons age 12+, employed.
United States 1960-1970: Persons age 14+ who had worked within the previous ten years; not armed forces, not new workers (see note).
United States 1980: Persons age 16+ who had worked within the previous five years; not armed forces, not new workers (see note).
United States 1990: Persons age 16+ who had worked within the previous five years; not new workers (see note).
Vietnam 1989: Persons age 13+.
Vietnam 1999: Persons age 13+ who worked at least six months in the previous twelve months.
NOTE: "New workers" are persons seeking employment for the first time who had not yet secured their first job.

OCC records the person's primary occupation classified according to the system used by the respective national census office at the time. For someone with more than one job, the primary occupation is usually the one in which the person had spent the most time or earned the most money, although this may not have been explicit in the instructions for a specific census.

The classification of occupations differs across countries and within countries over time. For a consistent classification of occupations at a general level, see the OCCISCO variable.

The age of the persons to whom the question applies varies across countries. In some samples, the question was asked only of those who were employed at the time of the census. In others, persons were asked to report the last occupation they had, even if they were no longer employed (e.g., Colombia 1964-1973, Mexico 1970, Vietnam 1989-1999, United States 1960-1990).

Some datasets (e.g., Kenya 1989, Mexico 1970) have numerous undocumented values.

For Kenya, only a small fraction of persons responded to the occupation question.

At the time of this writing, the universes are not imposed by programming. The theoretical universes for the samples are specified above, but the data may not match these specifications. Many persons too young for the occupation question receive a code in OCC.

NOTE: The original samples sometimes have fewer digits in their codes than the four allotted in OCC. In such cases, the codes are justified to the right, and the leading zeroes on the left should be ignored. See the link to "codes" above.

The question asked for the current main occupation for employed persons and the last one for the unemployed.

Colombia 1964 has an occupation variable that we coded in OCCISCO. There were many undocumented values, however, and the detailed codes are not yet reported in OCC.

OCC is fully comparable across the French samples. Unemployed persons report occupations in the French samples. In 1962, their occupation information was derived from their "metier" (profession), which is not exactly the same as occupation. The data have a separate code for new workers. The 1990 sample provides additional information about the occupation of retired persons.

The question applied to employed persons, regardless of whether they were at work, and persons on family holdings.

Despite the stated universe for the question, occupations are also reported for some persons seeking work, students, retired persons, disabled persons, homemakers, and others.

NOTE: Only a tiny fraction of persons actually reported an occupation.

The 1990 and 2000 samples used the same classification, but each of the earlier censuses used its own system. In 1970 the question applied to the occupation the person had during the previous year, 1969, regardless of status at the time of the census. In 1990 and 2000, the question was addressed only to employed persons.

The U.S. samples are fully comparable, aside from the differences in the universe with respect to age, armed forces, and experienced workers not in the labor force. 1980 and 1990 used the same classification, which differed greatly from the ones used in 1960 and 1970.

The 1999 census classified occupations in accordance with the 1988 ISCO system. The 1989 census used a radically different scheme that is non-hierarchical and highly industrial in character. In 1999, the question applied to those persons who worked at least six of the previous twelve months.