Attrition and Non-Response in Longitudinal Surveys: Methodology and Practice

Arising from the work of an expert group established by Longview on the topic of drop-out (sample attrition) in longitudinal surveys, Longview bid successfully under ESRC's Survey and Design Analysis Programme, for a project on "Solving the Problem of Attrition in Longitudinal Surveys". The project runs for three years and comprises three stages:

  1. Secondary analysis of attrition data in the British Household Panel Study (BHPS) 1958  and 1970 cohort studies (NCDS, BCS70), Family and Children Study (FACS).
  2. Consolidation of findings in development of theory for the causes of attrition.
  3. Field experiment to test hypotheses about reducing attrition based on the National Centre for Social Research Omnibus Survey.

The work involves teams in the Institute for Social and Economic Research (University of Essex); the Institute of Education; the National Centre for Social Research; University of Southampton; University of Manchester and University of Bristol.

The programme is managed by Longview, with Harvey Goldstein responsible for statistical coordination.