ONS (Office for National Statistics) Longitudinal Study (LS)

Summary

Strengths • Size: The Longitudinal Study (LS) contains records on over 500,000 people usually resident in England and Wales at each point in time. This approximates to a 1 per cent sample of the population. At any point in time it is largely representative of the population as a whole. The LS is the largest longitudinal data resource in England and Wales. Users can find an indication of the sizes of various groups in the LS sample by looking at tables describing the quality of LS data. • Longitudinal design: The LS has linked records at each census since the 1971 Census, as well as incorporating registration events. This allows, for example, analyses of mortality and cancer incidence based on occupation at one or more points in time, and analyses of occupational and geographic mobility. • Quality of linkage: The LS links eligible records from different censuses very effectively. Based on vital registration systems and the census, the LS has low 'attrition' when compared to other longitudinal studies, such as cohort or panel studies. • Information on co-residents: The LS contains detailed census information for people who are enumerated with LS members living in private households. This allows the analysis of, for example, household formation and dissolution, or the social class of LS members' parents and members' own social class 20 or 30 years later. • Users can add area-based information to individual records using a geographic identifier. • Support is provided for both academic and non-academic users. Weaknesses • The information contained in the LS is limited to what was collected at census or event registration. To date, censuses have not included the information on financial circumstances or behaviours and attitudes that is available from some surveys. • Census information is collected once every ten years and mostly relates to people's circumstances at the time. For example, census information on occupation usually relates to a person's activity 'last week', that is in the week before the census. • Information about people enumerated with LS members is not linked over time. • The confidential nature of the data in the LS means that it must be accessed from within a safe setting - that is at the Office for National Statistics.

Type of data

Data Source
Other, please specify

Type of Study
Survey same

Data gathering method
Registries
Self administered questionnaire
Other: ONSLS is a linked 1% sample of the population, based on 1971, 1981, 1991 and 2001 census data. The individual records also contain administrative data for events such as deaths, births to sample mothers, emigrations and cancer registrations.

Access to data

Conditions of access
Data are not available for download. Because of the confidentiality restrictions on the LS data set, it is not possible for users to have uncontrolled access to the data. Users are required to sign a Data Access Agreement before LS data can be released to them. Users can then receive data in a number of ways. They can receive cross-tabulations, frequencies and regression models for their chosen study population, which have been produced by a CeLSIUS support officer. They can also receive aggregated data sets, which they can use at their own place of work. Aggregated data sets contain counts of individuals for combinations of different variables. To maintain the security of the LS, all data are encrypted before being sent out to users and the approved encryption software is Safeguard PrivateCrypto. Users can download this software for free from the PrivateCrypto Web site. It may also be possible to go to the Office for National Statistics to work on an extract of the data set prepared by a CeLSIUS support officer. All outputs will be checked by CeLSIUS staff before they are released to users, so that they meet the confidentiality requirements. Unless you are based outside the UK or in the private sector, you should be eligible for support from CeLSIUS (the service is free to the user, being funded by the Economic and Social Research Council). You should contact the CeLSIUS Administrative Officer: celsius@ucl.ac.uk, +44 (0)20 7679 1995 The Administrative Officer will discuss your research question and your timescale, and pass your request on to a Support Officer who will guide you through the subsequent processes and make sure you get the best data for your research question. If you require one or two straightforward tables from the LS, either to establish whether the LS is likely to be useful, or to provide some background information, we can arrange this with a minimum of formality. If you are based outside the UK or in the private sector, please contact the LS Unit at the Office for National Statistics (ONS) directly: maus@ons.gov.uk, +44 (0)1329 447 871, MAUS, Room 1400, Office for National Statistics, Segensworth Road, Titchfield, Fareham, Hants. PO15 5RR Information is also available from the UK Data Service (previously the Economic and Social Data Service, ESDS): http://ukdataservice.ac.uk/

Type of available data (e.g. anonymised microdata, aggregated tables, etc.)
Anonymised microdata

Formats available
Data are not available to download

Coverage

Coverage Years of collection, reference years, and sample sizes
The ONS Longitudinal Study (LS) links information from the Census along with information from the registration of births, cancers and deaths. It contains linked census and event data for one per cent of the population of England and Wales. Data from the original 1971 Census sample have been linked to information from the 1981, 1991 and 2001 Censuses and with information on events such as births, deaths and cancer registrations. The original LS sample included 1971 Census of Population information for people born on one of four selected dates in a calendar year. These four dates were used to update the sample at the 1981, 1991 and 2001 censuses and to add new members between censuses. Data on approximately one million sample members have been collected over the 30 years of the study. At each census, data on slightly more than 500,000 sample members are collected. Routine event registrations are also linked to members of the LS. These events include births to women in the sample, cancer registrations, deaths and deaths of spouses. New LS members enter the study through birth and immigration. Data are not usually linked to a member after their death or after de-registration from the NHS Central Register but these members' records remain available for analysis. Cancer registrations take four years to be processed and are added to the LS database retrospectively. Census information is also included for all people enumerated in the same household as an LS member, but only information on LS members is linked over time. `Note: Older people are represented in this data source (approximately) according to their proportion in the population. In 2001, approximately one third of the total UK population was aged 50 and over.`

First year of collection
1971

Stratification if applicable
The data includes a variety of demographic variables, including age and sex.

Base used for sampling

Geographical coverage and breakdowns
Countries (England, Wales) Government Office Regions (NUTS1) Counties and grouped unitary authorities (NUTS2) Local authority districts and unitary authorities (NUTS3) More detailed spatial data are available, depending upon the analysis that is carried out. Note: • A similar study with complete coverage of Scotland is the Scottish Longitudinal Study (SLS), see: http://www.lscs.ac.uk/sls/ • A similar study with complete coverage of Northern Ireland is the Northern Ireland Longitudinal Study (NILS), see: http://www.qub.ac.uk/research-centres/NILSResearchSupportUnit/

Age range
All ages

Statistical representativeness
Population representative

Coverage of main and cross-cutting topics
Users can combine census data with information on events for the individuals in the LS. The LS documents periods of major change in all aspects of life in England and Wales. It allows users to extract information on changes over time on topics such as: • health/mortality • fertility • ageing • family formation • ethnicity • migration Users can derive new information from existing data, such as a measure of social-class mobility over ten, twenty or thirty years. Users are also able to derive their own variables from the detailed fields in the LS, such as socio-economic classifications created using employment information. Users may also merge data, such as area-based statistics, with the LS using geographic identifiers. Topics cover: • all information from census returns since 1971, including data on occupation, economic activity, housing, ethnicity, age, sex, marital status and education • event data on births and deaths, fertility, mortality, morbidity, migration and cancer registrations • ecological data: population density, urban/rural indicator and deprivation indicator

Linkage

Standardisation
There is an ongoing cross-governmental programme of work in the UK which aims to develop and improve standardised inputs and outputs for use in official statistics. This is known as harmonisation, and is led by the Office for National Statistics (ONS). While this work primarily affects government-run surveys, the results have an impact on most national UK data sources. Furthermore, harmonisation has important benefits for all researchers using these surveys, and not just government statisticians. For more information, see: http://www.ons.gov.uk/ons/guide-method/harmonisation/harmonisation-index-page/index.html Additionally, the LS uses definitions that are derived from the respective censuses (e.g. the Standard Occupational Classification). More information on census definitions is available from Celsius and ONS, for example see: http://www.ons.gov.uk/ons/about-ons/who-we-are/services/longitudinal-study/resources-and-definitions/index.html

Possibility of linkage among databases
Data are anonymised, and individual linkage is carried out by Celsius and ONS (i.e. confidential).

Data quality

Entry errors if applicable
Tracing rates indicate the likelihood of data linkage for groups within the Longitudinal Study (LS). LS members are 'traced' when their record has been found on the National Health Service Central Register (NHSCR). The NHSCR enables records for LS sample members to be linked to various life events for these individuals. It also facilitates matching of records collected at different points. LS members are 'untraced' if they have not been found, either because they have not been registered with a doctor, or inconsistent names or dates of birth have been used. Linkage of event data through the NHSCR is unlikely for untraced LS members. A number of factors have been identified that are related to LS members not being traced. These include: • being a young adult • being divorced • being born outside of the UK • living in London • living in certain types of communal establishment, such as psychiatric hospitals • serving in the armed forces • being a full-time student or in the economic position 'other inactive' • being a member of an ethnic minority • being long-term unemployed or having never worked For more information on data quality, contact Celsius.

Breaks
There are no major breaks for this data source. However, potential users are advised to consider the design of the study. For example, some data are collected only every 10 years (because of the period between censuses) and not all data are collected at every census.

Consistency of terminology or coding used during collection
The consistency of this data source is very good. For more information on data quality, contact Celsius.

Governance

Contact information
Centre for Longitudinal Study Information and User Support (CeLSIUS), Department of Epidemiology and Public Health
CeLSIUS
Gower Street
WC1E 6BT London United Kingdom Phone: +44 (0)20 7679 1995
Email: celsius(at)ucl.ac.uk.
Url: http://www.ucl.ac.uk/celsius

Timeliness, transparency
Administrative data are added to the study according to an annual schedule, with delays dependent upon the administrative source. Census data are added according to the individual timetable of successive censuses. Contingent upon the results of data quality tests that are underway, a database that includes 2011 Census data is planned for release in November 2013. For more information, see: http://www.ons.gov.uk/ons/about-ons/who-we-are/services/longitudinal-study/2011-census-link-beta-test/index.html