Data collection

The aim of collecting data for the Main Study was twofold. First of all, the aim was to provide descriptive information about the living arrangements (since 1971, there was no census in the Netherlands) and social networks of older adults (the Main Study). A large-scale survey based on a representative sample was expected to provide such information. Secondly, the Main Study provided the information necessary to carry out the two subsidiary longitudinal studies. Respondents who experienced the transition to widowhood are followed longitudinally (Widowhood Adaptation Longitudinal Study; WALS). A second study is the Network Study. Identification of the respondents to be followed in the Network Study took place on the basis of data collected during the Main Study. Furthermore, during the Main Study, additional information has been collected on selected groups of respondents. Research questions to be addressed in specific projects motivate the selection of these groups.

The Main Study consisted of face-to-face interviews conducted in 1992 with 4,494 respondents. Respondents were interviewed in their homes and personal computer assistance was used in the data collection. This program used a stratified random sample of men and women born between 1903 and 1937, so that the respondents vary in age from 54 to 89. The mean age of the respondents is 72.8 (SD= 10.0). By including these cohorts in the sample, data are available about people who grew up and reached maturity before and during the Depression and during or after the Second World War. These differences are likely to be reflected in the history of the composition of their households (e.g. membership of three-generation households, co-residence with parents as newly-weds due to housing shortage) or in the timing and the likelihood of marriage and the birth of children. In addition, data are available on older adults who, at the time of the interview, find themselves in different age-related circumstances. A large proportion of the youngest respondents face the transition from employment to retirement, and from having a family with children to having an "empty nest". A considerable number of the oldest respondents may be facing the transition from health and independent functioning to physical ailments and restricted independent functioning. They may be facing possible admission into a home for older persons. Furthermore, there is, with increasing age, an increasing chance of widowhood (with its associated changes in the composition of the household, and changes in the social network) and the increasing chance of death of social network members.  The oldest individuals, and in particular the oldest men, were over-represented in the sample. By not introducing additional stratification criteria, it remains possible to generalize the findings in a relatively simple way to the population of the selected regions and municipalities. The sample was randomly taken from the registers of eleven municipalities in the east (one city, Zwolle, and four rural municipalities, Genemuiden, Zwartsluis, Hasselt, and Ommen), the south (one city, Oss, and two rural municipalities, Uden and Boekel) and the west (Amsterdam and two rural municipalities, Waterland and Wormerland). These three regions can be taken to represent differences in culture, religion, urbanization and aging in the Netherlands.

Two municipalities (Amsterdam and Boekel) forced us to adopt a two-step sampling procedure. In the first step, the municipalities asked the people in the sample permission to pass on their names and addresses to the researchers. In the second step, the researchers sent the remaining people in the sample who lived in the two municipalities and all the people in the sample who lived in the other municipalities (N= 7,574) a letter introducing the study and the interviewers approached the prospective respondents. People who were deceased (5%), too ill to be interviewed (7%; for some of them proxies were interviewed), or did not speak Dutch (<1%) were classified as ineligible. Many (28%) refused, and a small proportion (<1%) was not contacted. On the basis of the response results in step 2, the non-response in step 1 was divided into eligible (i.e. refusals or not contacted) and ineligible. It was estimated that in total 2,785 people in the sample refused or could not be contacted and 1,416 were ineligible. The response was 62%, computed as the proportion of the number of face-to-face interviews conducted from the number of eligible sample members.

Overview of LSN data collections

Notes
1    The respondents in the LSN Network Study are a subsample of the older adults interviewed in 1992 and a selection of the network members identified by these older adults in the interview.
2    The sample was stratified by year of birth and sex.
3    Anchors in the LSN Network Study.
4    Anchors and a selection of their network members are respondents in the LSN Network Study.

LSN Widowhood Adaptation Longitudinal Study (WALS)

Introduction
Objective
Research questions
Societal relevance and policy implications
Current status
Acknowledgement
Datacollection
Data in the Main study
Collaboration
Availability