Name: Occupational data (Socialgruppstillhörighet)
Period(s): 1953, 1963 (SMS)
Code book: II


By Carl-Gunnar Janson (1980) Register Data II. A code book. Project MetropolitanResearch Report No 15. Stockholm.

The occupations refer to 1953 and 1963, respectively; in 1953 to the day of birth of the cohort member and in 1963 to November 1. However, in the occupational classification for 1953 also information on certain conditions up till 1958 has been considered.

For the 1953 classification occupations possible to categorize were obtained for all but 578 cohort members, of whom 242 were born abroad and not Swedish citizens. The information, mostly on the father's occupation, was taken from the delivery records in the first place as stated by the mother at the time of the delivery. However, as pointed out in the presentation of the delivery data, these records refer only to 12,121 deliveries in the metropolitan area. Furthermore, the relevant occupation was not always stated in the records or was not given in specific enough terms for classification. When an adequate occupational title was missing for a parent living in Stockholm City, the 1953 population register of the city was searched. If the information was not found in that register, because the parent was not registered as living in the city on November 1, 1953, or if a specified occupation was not stated, the excerpts from the social registers were searched. If the parent was not found there in 1953, an inquiry was made to the parish in which the cohort member was registered in the parish book of births. About 700 such inquiries were made. If even this failed a last attempt was made to find the parent in the 1954 population register of Stockholm City. In this series the most important sources of information after the delivery records were the 1953 population register and the parish books of births.

In most cases the father's occupation was used. The mother's occupation was used if the mother was single and not cohabiting in 1953 and was still single in 1957, if the father's or stepfather's occupation was missing or if the mother had an upper-class or upper-middle-class occupation whereas the father was a worker. The stepfather´s occupation was used if the mother was single and not cohabiting in 1953 but married another man than the biological father of the cohort member before 1958. If the cohort member was adopted before 1958, the adoptive father´s occupation was used, unless information on it was missing. Then the adoptive mother´s occupation was used, if not missing. The frequencies of these different types can be seen in FAMCODE. The problem whose occupation shall determine the social class of the family will come up also in a survey, although probably the information on which to base the decisions often is slightly better there.

For 1963 the occupations are taken from the 1963 population register and refer to the head of the family except when the information is missing for the husband but not for the wife, and when the husband has a working-class occupation but the wife has an occupation in the uppermiddle-class or higher. The family is that which the cohort member is registered as living with, whether or not the member is registered with parent(s) or other people. Only 412 families could not be classified.

All information used goes back to statements made by the mother or the family head either orally or in writing. The occupation in the delivery record was given to the staff of the delivery ward, where the interest in this piece of information most likely was not great in most cases. This resulted in a higher than usual rate of missing or unclassifiable occupational titles. In terms of response incentive this item probably is the one that comes closest to the conventional survey conditions, although here other register data substitute for the reluctant or avoiding answers. Occupations found in the social registers were given to case workers who easily could check them if they appeared too widely off the mark. The item in the parish book of births was duly reported by the mother to the vicar or some other presumably conscientious parish official who could check it against other information in the parish books. Filling out the yearly population registration questionnaire was mandatory to the head of the household. Even here, as with the social register and the parish books, the occupational item does not appear to have been of first rate interest to the agency. In 1963 the population register had ceased recording minor occupational changes. This is assumed to affect the broad classification of occupations into social classes only slightly. It may be mentioned that the local income-assessing boards use the population registration lists. All things considered, it can be surmised that the available information on occupations is at least of the same order of accuracy as ordinary survey information, e.g. that obtained in the Family Study. In the School Study the cohort members were asked about father´s occupation, but the answers are thought to be inferior to the register data. The most serious limitation of the register data on
occupation probably is the absence of supplementary information on education of the type available in both the Family Study and the School Study. For 1953 and 1963 data on family composition were extracted from the registers, and for 1963 assessed incomes are available.

The occupational classification is an elaboration of a system used by the National Central Bureau of Statistics (SCB) in its official reports on general elections from 1911 to 1956. Voters and people entitled to vote were classified in three "social groups", which roughly corresponded to upper and upper-middle class (social group 1), lower-middle class (social group 2), and working class (social group 3), each social group consisting of several subcategories according to a two-digit code. In the report on the 1952 parliamentary election the classification was used only sparingly, in the report on the 1956 election the tables on social class were not published, and after that the "social groups" were abandoned. By then they had become widely used in the social sciences and in general discussions. The Statistical Department of Stockholm City continued to use them until the mid'60's, and even the surveys carried out by the SCB and the Department of Political Science at the University of Gothenburg on voting in each parliamentary election since 1960 divide the interviewees in socioeconomic categories that can be collapsed into the three "social groups". The social groups were never used in the censuses of population and housing.

Social group 1 consists of leading personnel in big private enterprises and within the public sector, and of professionals. If a certain level of formal education is normally required for the occupation, it is usually some kind of academic education. In social group 2 we have lower rank employees, usually with some high-school education (realskola or gymnasium in the old system), office workers, shop assistants, and small-scale entrepreneurs, including most farmers, whereas manual workers and the lowest ranks of employees, such as bus drivers and superintendents, are placed in social group 3.

Project Metropolitan classifies occupations according to a two-digit code, of which the first digit defines ten categories, the tenth of which is a left-over category of economically inactive persons, such as military conscripts, housewives, the unemployed, and retired people, so far they can not be placed in other categories (cf Research Report No 3:24,72). Categories 1 and 2 with a tenuous division in private and public sector, make up social group 1. Social group 2 is divided in two parts. The first part is category 3, which contains the employees, and the second part consists of categories 4 and 5, small-scale entrepreneurs within urban branches of
industry and farmers and gardeners, respectively. For obvious reasons the agricultural category is very small in the metropolitan cohort. Social group 3 is also divided in two, categories 6 and 7 and categories 8 and 9, respectively. The first part includes the low-rank employees (category 6) and the skilled workers (category 7), the second part the unskilled urban workers (category 8), and the farmhands (category 9), who again are very few in the cohort. The distinction between skilled and unskilled labor was not made among the sub-categories of the election-statistics classification. SOCGR53 and SOCGR63 show the frequencies of the five socioeconomic categories. Note that there is no clear rank order between the two lower-middle-class categories.