European Union Statistics on Income and Living Conditions (EU-SILC)

Summary

EU-SILC is the reference source for the study of income distribution, living conditions and social exclusion at European level. The design of EU-SILC involves a common ex ante framework defining the harmonised target variables to be collected or produced by national statistical institutions. As a result, there is a large degree of flexibility in the underlying sources of EU-SILC, and some flexibility in the concepts and definitions used. The flexible survey design also allows embedding EU-SILC into the national systems of social surveys. Therefore, emphasis on output harmonisation, rather than input harmonisation, might lead to diminish the international comparability of the data. Social exclusion and housing condition information is collected at household level (core data, income and tax, housing, material deprivation), while basic demographic data, income, labour, education and health information is obtained at individual level for persons aged 16 and over. Access to personal income data is available only on the subsample of 'selected adult respondents 16+'. The selected respondents are to be analysed at the level of persons only, using special selected weights, without aggregation to household level. The absence of stratum indicators in the dataset should be considered in order to avoid underestimated standard errors, thus leading to erroneous conclusions. In regard of the household and familiar structure, one of the limitations of EU-SILC is that it does not provide a household grid, thus leaving aside the proper identification of the nature of living arrangements. Regarding income variables, some of the aggregations on income or social benefits might result in limitations for the analysis. The 4-year rotational format of the EU-SILC implies that the information on individuals’ history is reduced to four years. The time framework differs across member states, since only 13 countries launched the SILC in 2004. Health and Performance The EU-SILC allows the cross-country analysis on access to health care through its core and longitudinal components. The dataset provides subjective measures, such as general health status, chronic illnesses, or limitations in daily life because of health problems, as well as unmet needs for medical and dental examination or treatment in the last 12 months. Thus, the EU-SILC contributes to the cross-national study of the association of income and health. One of the potential limitations of the variables included is the reporting bias of subjective health measures, as well as the dimensions covered by the health variables depending on the national context.

Type of data

Data Source
Registry + Survey

Type of Study
Survey same
Crosssection regular

Data gathering method
Telephone
Face-to-face
Registries
Self administered questionnaire

Access to data

Conditions of access
EU-SILC micro-data are available to researchers carrying out statistical analyses for scientific purposes. Multi-dimensional datasets and policy indicators are updated on the Eurostat website as soon as the new data become available. Basic indicators are downloadable free of charge as excel files.

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

Formats available
Excel

Coverage

Coverage Years of collection, reference years, and sample sizes
Minimum effective sample sizes 1. Cross-sectional data operation (EU members + Iceland and Norway): about 131,000 households and 273,000 individuals aged 16 or over to be interviewed. 2. Longitudinal data operation (EU members + Iceland and Norway): about 98,000 households and 204,000 individuals aged 16 or over to be interviewed. In most cases participant countries launch EU-SILC from scratch with integrated cross-sectional and longitudinal elements (this is the Eurostat recommendation). Other countries use a combination of registers and interviews. Others seek to adapt existing national sources. Precision requirements are set via the prescription of minimum effective sample sizes, which are specified in the EU-SILC framework regulation 1177/2003. They should be carefully designed to ensure representativeness - and are to be increased by participant countries to the extent that their national sample is not determined on a simple random basis, or to reflect likely levels of non-response, or to reflect any specific national requirements. Separate values are specified for the cross-sectional and longitudinal elements.

First year of collection
2004 (for some countries, EU-SILC was launched in 2003)

Stratification if applicable
Most countries apply stratification on at least one stage, but no stratum indicator is available as part of the EU-SILC dataset. It does provide information on clustering (primary sampling units)

Base used for sampling

Geographical coverage and breakdowns
Given its stepwise implementation, EU-SILC data has been released as following: 2004: Austria, Belgium, Denmark, Estonia, Finland, France, Greece, Iceland, Ireland, Italy, Luxembourg, Norway, Portugal, Spain, and Sweden. 2005: + Cyprus, Czech Republic, Germany, Hungary, Latvia, Lithuania, the Netherlands, Poland, Slovakia, Slovenia, and the United Kingdom. From 2007 onwards: all 27 Member States in addition to Iceland, Norway, Switzerland, and Turkey. For the anonymised microdata, region changed from NUTS2 to NUTS1.

Age range
The EU-SILC target population is 16+ and living in private households. Persons living in collective households and in institutions are generally excluded from the target population.

Statistical representativeness
Population representative

Coverage of main and cross-cutting topics
The EU-SILC collects harmonised information on income distribution, living conditions, and social exclusion. The survey also provides the demographic and educational characteristics of the population. Even when it captures the household characteristics, the information regarding living arrangements and household compositions might be limited. In the temporal dimension, cross-sectional data pertains to a given time or a certain time period with variables on income, poverty, social exclusion and other living conditions. Longitudinal data pertains to individual-level changes over time, observed periodically over, typically, a four year period. The EU-SILC collects information on the general health, chronic conditions, and limitation activities, which are due to health problems. These variables are collected in both the longitudinal and the cross-section component of the database. Since 2004, there have been at least three variables on self-perceived health.

Linkage

Standardisation
– NACE for economic activity – ISCO 88(Com) for occupation – ISCED for education For the anonymised microdata: Most variables related to survey design have been withdrawn or randomised. Time of interview, age, dwelling type, origin, and NACE classification has been recoded. Additionally, there are some country-specific provisions to ensure anonymisation.

Possibility of linkage among databases
EU-SILC microdata do not contain any administrative information such as names or addresses that would allow direct identification.

Data quality

Entry errors if applicable
Some labour variables collected on all adult household members (16+) through the interviews might suffer of proxy answers. The responses given by one member of the household on behalf of the others might also be a source of measurement error.

Breaks
During the transition between the end of ECHP and start of EU-SILC (up to 2007 in some countries), data was provided by NSI's from national sources (with some breaks in series due to lack of information, transition from national data source to EU-SILC, etc.). Regarding sampling methods: Germany, which had previously used quota sampling methods, was granted a transition period until 2008 when it was required to introduce fully representative probability sampling. In most countries, the sample takes the form of a rotational panel by dividing the sample into sub-panels retained for a maximum of four years. Each year, one sub-panel is refreshed by a new replication. Even when the standard number of rotational groups is four, the exceptions are: France (nine-year panel), Norway (eight-year panel), and Luxembourg (pure panel). Member states are allowed to use different survey instruments to collect cross-sectional and longitudinal data. Also, linkages between these datasets are not required.

Consistency of terminology or coding used during collection

Governance

Contact information
Ms. Karien Reinig
European Commission, Eurostat – Unit B1
5, rue Alphonse Weicker
2721 Luxembourg Luxembourg Phone:
Email:
Url:

Timeliness, transparency
Longitudinal and cross-sectional SILC microdata are released 2 years after collection. The extreme deadline currently is October N+1 for most countries for cross-sectional data, with N being the fieldwork year and not the (income) reference year. Following the Framework regulation, Member States shall transmit to Eurostat the cross-sectional data of Year N by 30 November (N+1) and the longitudinal data of Year N by 31 March (N+2). New users’ databases are released in March and August of each year. Most recent longitudinal UDB SILC microdata containing all the trajectories with the most recent wave: Longitudinal UDB SILC 2010 released in August 2012. Annual ad-hoc modules transmitted with and included in the cross-sectional files: Extreme deadline for transmission of micro-data to Eurostat is 30 November (N+1) for Member States where data are collected at the end of year N or through a continuous survey or through registers, and 1 October (N+1) for other Member States. The dissemination by Eurostat of national microdata must be accepted by each national authority.