NCDS6 (2000)


In 1999/2000, NCDS combined with BCS70 to undertake, for the first time, a joint survey. This was in line with the Forward Plan for the cohort studies, developed by the Director of CLS, Professor John Bynner. This sought to integrate the timing, design and analysis of future surveys of NCDS and BCS70 – taking account of the sequencing of Britain’s third birth cohort study, the 1946 cohort (National Survey of Health and Development), housed at University College London. It was envisaged that such a programme would significantly enhance the research potential of the studies, enabling comparisons to be made between cohorts born at different times, or between different age groups at the same point in time.

Funding

Endorsement of the principles of the Forward Plan by the Economic and Social Research Council, and the Government departments, which had provided financial support for previous sweeps, resulted in an ESRC decision to fund the new surveys. Initially, this was restricted to BCS70 because of the need to update the dataset after 13 years without a comprehensive survey, but, subsequently, funding extended to a new survey in NCDS as well. ESRC contributed half the costs, and the rest came from Government departments under the coordination of the Office for National Statistics: Department for Education and Employment (the major funder), Department of Health, Department of Social Security, the Home Office, ONS themselves, the Scottish Office and the Basic Skills Agency.

Life course perspective

Adult identity, comprising an individual’s pre-dispositions, attitudes, behaviour and statuses, needs to be viewed both as a product of past life-course processes and a precursor of later ones. The 'life course' perspective enables us to take on board these different elements of human development, while offering a means of integrating them within a single inter-disciplinary framework. Life courses represent the transitions occurring in the different domains of life through which individuals make adjustments to changes in their social environment and through which their statuses in education, employment, relationships, family life and health are maintained and developed.

Design

The first stage in designing the new surveys was to locate the key issues and questions salient to the stage in adult life that the cohort members in each study had reached. These were formulated in terms of a ‘life course’ theoretical framework that focused on the factors central to the formation and maintenance of adult identity in each of the following main life domains:

  • lifelong learning
  • relationships, parenting and housing
  • employment and income
  • health and health behaviour
  • citizenship and values

As with previous NCDS and BCS70 follow-ups, the surveys were designed in collaboration with advisors drawn from researchers, policy makers and funders. Seven advisory groups were formed in March 1998, one for each of the above major topics, plus the following two additional areas:

  • Child development & education
  • Methodology

Members of each group exchanged ideas via email and/or meetings and, ultimately provided written advice on the content of the survey instrumentation (June-September 1998). Subsequently, members of the advisory groups were updated on the development of the instrumentation, and the progress of the surveys. Later, a number of advisors were involved in the initial assessment of the quality of the information obtained during the survey.

Further details are published in CLS Cohort Studies Working Paper No.1, The design and conduct of the 1999-2000 surveys of the National Child Development Study and the 1970 British Cohort Study. 

Data Specification Parameters

The main focus of the surveys was to elucidate the latest stage of the life course of cohort members, revealing the pathways taken to current statuses, the influences to which they had been subjected and their outcomes in terms of current identities. The parameters used to identify the key variables included:

  • Structural categories: gender, social class, religion, and geographical location
  • Institutions: family, education system, employment system, trade unions, political system, health and welfare services
  • Social relations: assessed in terms of the embeddedness of individuals in social networks, in education, at work, in the family and at leisure.
  • Social roles, statuses and values: embracing roles such as parent, spouse, employee and community member, as well as position in the social hierarchy, membership of associations, etc.
  • Health and lifestyle: these comprise the record of mental and physical illnesses, symptoms, accidents and the treatments received for them and their outcomes. Lifestyle choices such as smoking, alcohol and drug use, diet and exercise all have health-related aspects.

There were also four factors to do with the methodology of collecting such longitudinal data over time, which needed to be taken into account:

Continuity and comparability

By 1999, both cohorts were well into adulthood, with BCS70 subjects aged 29 and NCDS having reached 41. This pointed to a high degree of commonality between the two cohorts in the topics to be covered and the kinds of questions to be asked. It was also desirable for the purpose of inter-cohort comparisons.

On the other hand, each study had specialised somewhat in the past, focusing to a certain extent on different features of childhood and adolescence, and adopting its own distinctive methods of data collection in earlier sweeps (for example, diaries and medical measurements in BCS70, parental and school data and much information on educational performance in NCDS). What was needed was a balance between replication of variables across the two cohorts and repetition of variables within each study from one survey to the next.

Age, cohort and period effects

The data also needed to embrace the common elements of all life histories, such as those determined by biologically-based developmental processes (age effects) and their variability in terms of timing and duration. It was equally important to encompass those life course elements that are culturally determined, and how these differed between different socially defined sub-strata – including socially excluded groups. Such life course elements are subject to two other types of external influence. The effects of social change are embraced by comparisons between different cohorts – cohort effects. Societal influences at the time of data collection, which are held constant for any given survey, reflect period effects. Taken together with geographical location, these constitute the temporal and spatial context (time-space co-cordinates) of the data collection.

Space-time co-ordinates

These comprise information about the year of study (1999-2000), and the geographical location of each individual cohort member. In relation to time of study (period), we needed to consider external events that could influence cohort members, eg Government legislation, health scares and so on. In relation to space, it was necessary to consider external pressures on individuals which are regionally based, eg reductions in employment, and legislation with differential effects across regions, such as devolution. Thus, we needed to take on board distinctions to do with rural/urban, North/South, Scotland/England/ Wales and so on. With respect to age, we needed to separate out the developmental components that are biologically based (eg physical ageing) from those that are more socially determined (eg age at marriage, or at birth of first child). We may nevertheless see the former as defining certain boundaries for, or driving, the latter, eg the age-defined period of the life course when having children is possible.

Harmonisation

A final consideration in setting out the principles guiding the design of the new surveys was the move begun in the 1990s towards the harmonisation or standardisation of the information collected for Government social surveys. In developing the new instrumentation, therefore, reference was also made to ways in which questions were formulated in surveys such as the British Household Panel Study and the General Household Survey, as well as, for maximising the opportunities for inter-cohort comparisons, the National Survey of Health and Development (1946 cohort).