The ONS longitudinal study: Quality issues from 30 years of data linkage

Abstract

The Office for National Statistics Longitudinal Study (LS) contains anonymised linked census and life event data for one per cent of the population of England and Wales. This dataset has been prepared, maintained and developed for the last thirty years to a high standard of quality. Data quality is maintained through an optimal balance between user needs and processing constraints, including costs. The current linkage of 2001 Census data is described with reference to the quality dimensions of relevance, accuracy, timeliness, accessibility and clarity, comparability, coherence and completeness. Some of the processes described are context-specific, like the availability of expert tracing staff at the National Health Service Central Register and the use of ONS administrative data. Others are good practices that can be shared by other Statistical Agencies, including effective communications, comprehensive metadata, a safe setting environment for authorised users to access the data, user support and training. Quality management in the LS involves the monitoring of processes and products in a cycle of continuous improvement. This process of proactive, continuous change is demanding and relies crucially upon the contribution of a dedicated and expert staff.