|
The Danish survey
The Danish survey randomly samples 6,011 children among all children living in Denmark. The children in the sample are born between September 15 and October 31, 1995, by mothers with Danish citizenship. A small minority of the mothers were naturalized Danes; less than two percent did not have Danish as their native language. The fathers’ citizenship (or whether a father lived in the family at all) was not taken into account when the sample was drawn, i.e. ‘mixed marriages’ are possible in the Danish survey.
Detailed information about the response status of the children in all waves of the Danish survey is available here.
The survey is intended to follow the children all the way from infancy to adulthood. In the early waves of the survey, where the children were not old enough to answer questionnaires themselves or to be interviewed, the mothers were the primary informant.
As seen from the table below the mothers have been interviewed in all waves of the survey, the fathers were interviewed in 1996, 2003 and 2007, and the children were interviewed for the first time in 2007. The table links to the relevant questionnaires.
|
|
Questionnaires in the Danish survey
In all waves of the survey, information from the mothers has been gathered through “standardized interviews”. This means that the questions from the questionnaire are asked to the mother by an interviewer from SFI who, in turn, fills in the questionnaire. The standardized interview ensures a relatively high data quality (for example, it minimizes the occurrence of missing answers) because the interviewers are able to clarify the questionnaire if necessary. In 1996 and 1999, the interviewing was done according to paper questionnaires, while it has been computer based (CAPI) since 2003.
The questionnaires to the fathers have been for self completion in all waves of the survey. The fathers could complete the questionnaire while the mothers were being interviewed (and give the questionnaire to the interviewer) or they could do it at a later time (the interviewer would leave a stamped envelope).
The questionnaire to the children, in 2007, was answered on the internet. To get the children’s perception of their own lives (rather than the parents’), it was preferred that the children answered their questionnaire on the family computer while the mother was being interviewed (preferably in another room). If, for some reason, internet was not available, the interviewers brought a computer based (CAPI) questionnaire.
|
|
1996
The data was collected in the period January-May, 1996, by the interview staff of SFI. At that time the leave-of-absence was coming to an end for most of the mothers who were about to return to work. The interviewers were instructed to try to visit the families after the children were four months old but before the end of the 24 week leave-of-absence.
The focus of the first wave was the time after the birth. The questionnaire to the mothers described the children through questions about breast feeding, crying, colic and the children’s development. There were questions about health/illness and about the children’s temperament and mood. Further, to describe the possibilities that the parents were able to supply during childhood, there were questions about child care and the parents’ health, education, vocational education, work and economic circumstances.
The interviewers visited the families at home where they used a questionnaire to interview the mothers and – if an interview was obtained – delivered a questionnaire for the fathers that they could fill out themselves, either during the interview with the mother or later. If the fathers preferred to answer the questionnaire later, a stamped addressed envelope was supplied. During the interview, preferably, the interviewers were alone with the mothers but if this was not possible, the interviewer noted who had been present also.
Response rates in 1996
| | Mother questionnaire | Father questionaire | Pct. | Total | Pct. | Total | | Response | 90.3 | 5,429 | 78.3 | 4,087 | | Non-response | 9.7 | 582 | 21.7 | 1,135 | | Total | 100.0 | 6,011 | 100.0 | 5,222 |
A total of 6,011 children were sampled to the first wave of the survey. The response rate on the questionnaire to the mothers was 90.3 as interviews were completed with the parents of 5,429 children (5,425 mothers and four fathers). Of these children, 207 lived with a single parent while 5,222 lived with two parents.
The questionnaire to the fathers was only administered to the fathers of the 5,222 children who lived with two parents. 4,087 of these questionnaires (78.3 pct.) were completed. This response rate is a bit overestimated because it does not take into account the fact that the questionnaire was only administered to the father if an interview was obtained with the mother, i.e. if the mothers did not answer, the fathers response rate was 0 pct. The questionnaire to the father was given to the mother’s partner (if they lived together) no matter if this person was the children’s biological father or not.
|
|
|
1999
The data was collected in the period January-April, 1999, by the interviewers of SFI. At this time the children were about 3 ½ years old and most of them attended some kind of day-care facility.
The focus of the second wave was the children’s development, their social relations and daycare. The questionnaire to the mothers described the children’s development by the use of a number of questions about the children’s ability to perform (or not perform) several tasks. There were questions describing the children’s temper and relations to other children, and the type(s) of day-care arrangement(s) used in the period from the first wave in 1996. Again, the parents were asked about their social conditions. A new block of questions asked divorced parents specifically to the divorce.
Prior to the visit by the interviewer, the families received a letter of introduction encouraging the mothers to participate in the survey. In order to maximize the response rate (and because not all analyses require response in all waves of the survey) the letter was sent to all the families that had been selected to the survey in 1996, not just to the ones that had been interviewed successfully.
Response (mother questionnaire) in 1999
| |
Pct.
|
Total
|
| Response |
88.0
|
5,288
|
| Non-response |
2.2
|
132
|
| Refused/exited from sample |
7.7
|
463
|
| Total |
100.0
|
6,011
|
In the second wave of the survey, interviews were obtained with the parents of 5,288 children corresponding to a response rate of 88.0 percent of the original sample.
However, some of these parents had not responded in the first wave. Only 4,991 children (83.1 percent) participated in both waves. The number of paternal respondents (to the maternal questionnaire) increased from four in 1996 to 55 in 1999. Most (31) were single fathers; the rest (24) was in families that preferred the father to be the respondent.
There was no questionnaire to the fathers in 1999.
|
|
|
2003
The data was collected in the period January-March, 2003, by the interviewers of SFI. At this time the children were around 7 ½ years old and most of them had started school.
In the third wave, again, the mother-questionnaire asked questions about the children’s development, health and social relations. But this time a new block of questions asked about their school attendance. The mothers were also requested to answer a special questionnaire focusing on the children’s physical development (height, weight) and medical history. If the mothers were unable to answer the questions in the special health-questionnaire, permission was requested from them to obtain the information from the children’s general practitioner.
Not all the originally selected families were contacted in 2003. Families were not contacted if they, in 1999, stated that they did not want to participate in the survey, if they had moved abroad or if the children were not eligible because of handicap. The response rates to the mother questionnaire are calculated below in the sample not allowing for attrition.
Response rate (mother questionaire) in 2003
| |
Pct.
|
Total
|
| Response |
82.7
|
4,971
|
| Non-response |
1.9
|
113
|
| Refused/exited from sample |
2.5
|
151
|
| Total |
100.0
|
6,011
|
The parents of 4,971 children – 82.7 percent of the original sample – answered the maternal questionnaire in the third wave of the survey. The questionnaire to the fathers was answered by 3,502 of 4,326 married/cohabiting fathers (81.0 percent). It was only used if the person answering the maternal questionnaire stated that she/he lived with a partner.
By 2003, the parents of 4,644 children had participated in all waves of the survey, corresponding to a response rate of 77.3. The father answered the maternal questionnaire in 91 families (45 single fathers and 46 from families that preferred the father to be the respondent).
Supplementary data collections
The mothers were also requested to answer a special questionnaire focusing on the children’s physical development (height, weight) and medical history. If the mothers were unable to answer the questions in the special health-questionnaire, permission was requested from them to obtain the information from the children’s general practitioner.
|
|
|
The data was collected between January and October 2007 by interviewers from SFI Survey. For the first time, the survey included a questionnaire to the children. It was internet-based but the interviewers also brought a disk for use in computers without access to the internet. The questionnaire was independent of reading skills because it was possible to have the questions read aloud if the child preferred so (and if the family computer had the necessary audio facilities). In order to ensure more open/honest answers from the children, it was preferred to have them answer their questionnaire alone while the mothers were being interviewed.
Response rates in 2007
| |
Mother questionnaire
|
Father questionnaire
|
Child questionnaire |
|
Percent
|
Total
|
Percent
|
Total |
Percent |
Total
|
| Response |
79.9
|
4,802
|
76.9
|
3,078
|
76.0
|
4,568
|
| Non-response |
16.5
|
992
|
12.1
|
925
|
24.0
|
1,443
|
Refused/
exited from sample |
3.6
|
217
|
-
|
-
|
-
|
-
|
| Total |
| |
