Centre for Longitudinal Studies

In 2004, CLS was granted long-term funding as an ESRC Resource Centre to establish a stable infrastructure for the study and ensure that fieldwork is conducted at regular intervals.

The Centre for Longitudinal Studies led the development of the 2004/5 survey and commissioned the fieldwork.   Data collection took place in a face-to-face interview when the cohort members were aged 34/35.

Funding

Fieldwork was funded by the Economic and Social Research Council (ESRC).

Design

The survey was designed in consultation with those who have been involved with the design and analysis of earlier BCS70 (and NCDS) surveys: other research advisors, funders and policy-makers, in accordance with the following principles:

• Relevance to the stage of life reached
• Continuity with previous surveys
• Comparability across NCDS and BCS70
• Compatibility with other surveys (eg: BHPS, the General Household Survey and the (US) National Longitudinal Survey of Youth)

The main aim was to update information gathered previously, in order to explore factors central to the formation and maintenance of adult identity in each of the following domains:

• Lifelong learning
• Relationships, parenting and housing
• Employment and income
• Health and health behaviour
• Citizenship and values

Survey Instruments

The core of the BCS70 2004/5 follow-up was an interview and self-completion.  As in the 2000 survey, this was done by CAPI and CASI instruments (Computer-Assisted Personal Interview/Self Interview).

For a one-in-two sample of BCS70 cohort members, information was also gathered from and about all natural or adopted children living with them. 

Information about children was gathered from the cohort member by interview and self-completion, and from the children using child assessments for all those aged 3<17 years, and by self-completion for all aged 10<17 years. 

This Parent & Child survey provides for both inter-generational comparisons, using earlier BCS70 data, and cross-cohort comparisons, using data from the very similar Mother & Child survey that was based on a one-in-three sample of NCDS cohort members at the time of the fifth follow-up in 1991.

Fieldwork

Following competitive tendering, the National Centre for Social Research (NatCen) was commissioned by CLS to carry out both the BCS70 2004/5 face-to-face survey and the 2004/2005 NCDS telephone survey.  NatCen’s responsibilities were to work in collaboration with CLS in the development of instrumentation, carrying out the data collection and initial data preparation, as well as supplementing CLS tracing activities, and making contact with cohort members.

The main data collection took place between February 2004 and June 2005. It had originally been scheduled to be completed within nine months but was extended when early fieldwork showed that a relatively high proportion of cohort members had moved and needed to be traced to new addresses.

Data coding and editing

Data were returned from the field (via modem) and coding and residual editing undertaken.

A major advantage of the use of CAPI and CASI is the reduced need for post-fieldwork editing – the majority of checks for validity, range and consistency can be incorporated into the CAPI/CASI program.  Inevitably, however, there were checks, which were overlooked, or not initially thought necessary.  These checks were incorporated into the data processing activities undertaken by NatCen after the survey.

The BCS70 interview and self-completion include a number of open-ended questions where the verbatim answers of cohort members are keyed by interviewers, and a rather larger number of questions where precodes are provided for answers but provision is also made to record additional information where then precode ‘other’ is used.  Following the start of the surveys, these questions were reviewed by the CLS team to determine the priorities for coding, and to identify the appropriate coding frames.  Where possible, coding frames that had been employed for earlier BCS70 surveys were adopted, although it was usually necessary to include additional codes.  In other instances, it was necessary to develop a coding frame from scratch.

Coding was undertaken by NatCen.