Name: The Family Study (Familjeundersökningen)
Period(s): 1968 (SMS)
Code book: I


By Carl-Gunnar Janson (1980) The Family Study. A code book. Project Metropolitan Research Report No 4. Stockholm.

The field work for the family study was carried out between March and June, 1968. The respondents were a sample of 4,021 mothers or substitute mothers. The cohort studied in Project Metropolitan consists of those persons who satisfy both of two criteria, i e they were born in 1953, and they were registered as living in the Stockholm Metropolitan area on November l, 1963. The cohort numbers 15,117 persons, 7,398 females and 7,719 males.

For the purpose of drawing the sample of the family study the cohort (or population) was divided in five strata. First, the population members still in the metropolitan area as of November l, 1967, were listed. Some 525 members had been lost since November l, 1963. Of the remaining those who had complete mental test scores from the school study were distributed in three strata (High, Medium, and Low) according to score. The "High" stratum comprised those with the five per cent highest scores, and the "Low" stratum held those with the five per cent lowest scores, whereas "Medium" had the intermediate ninety per cent. The cutting points on the score line were fixed separately for girls and boys and duly corrected for those few who had the test in 1967 instead of in 1966. Actually both end strata were made a little larger than five per cent. Those who had not participated in the school study were put in a fourth non-response stratum. Finally, a small fifth stratum contained those with incomplete test scores.

In the High and Low strata all members were included in the sample. In each of the other strata every fifth was drawn. The total sample comprised 4,021 cohort members, 1,972 girls and 2,049 boys, as can be seen in the first table of the report. For each sampled cohort member the mother or substitute mother was taken as the person to be interviewed.

For the field work the Survey Research Institute (UI) of the National Central Bureau of Statistics (SCB) was engaged. Its interviewers were informed about the project and study by me both in writing and orally and carried out test interviews.

Before the field work started an introductory letter was sent to each respondent. It was clear from the letter that the interview would contain questions on family conditions. That, of course, meant that a higher rate of refusals was likely than the Survey Institute usually encountered in its surveys of less delicate matters. However, the refusals at first contacts were fewer than expected, less than 12 per cent (unweighted), but still many enough to deserve special handling. Most of those who had declined to be interviewed received a letter, in one of three versions, asking them to reconsider their decision. Some even got a second letter or were called on the telephone. The measures brought the refusals down to 7 per cent (unweighted).

The field work started in March, 1968, with 75 interviewers. It was scheduled to last through April. However, the interviewers got more involved than the Survey Research Institute had reckoned with in its other projects. This, together with the refusals, made the institute call upon additional interviewers and extend the field work in time. In fact, interviewing dragged on until June 10.

Then 3,651 respondents had been interviewed. Of the original sample 21 families had left the area. For fifteen children who lived permanently outside home there was no interview. Twenty-eight respondents were ill, either at home or in hospital, and could not be interviewed, twenty-five were never reached, and four were not interviewed for other reasons. Finally, 277 declined to be interviewed. Of the total sample the non-response was 9 per cent. If those who had left the area were subtracted from the sample, the non-response became somewhat less than 9 per cent.

The rate varied between strata. It was highest for those without mental test score, next to highest for the low-score category, and lowest for the high-score category, which hardly is surprising. The rates were 19 per cent, 13 per cent and 5 per cent respectively. The nonresponse was the same for mothers of cohort boys as for mothers of cohort girls. In addition to the all-over non-response there was, of course, missing answers on each question. In the first section of the interview the respondent and her role in relation to the child was identified. In little more than one per cent of the cases another person than the original person sampled was found to have the role of the child's mother and thus was substituted. A good one per cent of the respondents were men.

The second section of the interview dealt with occupation and education of the interviewee, her husband and children, her parents and parents-in-1aw. For other cases than married couples living together, "husband" was extended to denote former husbands and men living together with interviewees on a more or less permanent basis. According to this extended definition practically all respondents had or had had a husband and parents-in-1aw. In later sections some questions on the husband's opinions or interactions with the child or interviewee were not app1icab1e in all cases. For instance, questions on present interaction within the family did not relate to former husbands. It should be noted that the husband referred to in the interview was not necessari1y the father (biologically or from adoption) of the child in the cohort, although it must have been clear to the interviewee a1ready from the first part of the interview that she was interviewed as the mother or substitute mother of that child.

In the third section there were questions on educationa1 and occupational plans as to the child, on parental involvement in the child's schooling, on attitudes towards school, and on evaluation of occupations. The attitudinal items were the same as those put to the chi1d in the school study, although some omissions and modifications were made because they now were put to an adult person. Occupations were evaluated first as future occupations of the child according to the interviewee's opinion of her own, then as the interviewee thought most parents would feel about their daughter or son getting into them. All respondents were asked about ten occupations of which three were the same to all. These occupations served as points of reference and were farmer, meta1 worker, and secondary school teacher. For the remaining seven occupations the sample was divided into ten sub-samples, each of which had their own set of occupations to evaluate. Some occupations were not the same when the child was a girl as when it was a boy.

The fourth section comprised various questions on norms as to the child's behaviour and on relations between family members. Finally, the fifth section consisted of three sets of attitudina1 items on up-bringing, politics, and conception of the future, respectively. The code book presents the answers to most but not all questions of the interview. In order to make the sample representative of the population in the study, one needs to poststratify the material either through the means of multiplying the intermediary category in the stratum variable (FAM7) by the value of 5 or by using the variable vikt1. The former is suitable for presenting descriptive statistics while the latter is more appropriate for the purpose of drawing inferences.