Abstract copyright UK Data Service and data collection copyright owner.

The Annual Respondents Database (ARD) is constructed from a compulsory business survey. Until 1997 it was created out of the Annual Censuses of Production and Construction (ACOP and ACOC); these were combined into the Annual Business Inquiry (ABI) in 1998. The ARD is a census of large businesses, and a sample of smaller ones. Smaller firms may receive a "short form". These do not require detailed breakdowns of totals. Hence for certain variables the values may be imputed from third party sources or estimated rather than returned by respondents.

This dataset is created for the Economic Analysis and Satellite Accounts Division for research purposes. To create the ARD, the other surveys are converted into a single consistent format linked by the Inter-Departmental Business Register references over time. Northern Ireland data is held up to 2001. From 2002, the ABI is collected and stored separately in Northern Ireland. Special permission is required to use new NI ABI data.

ABI background
The ABI is the financial information survey conducted by the Office for National Statistics (ONS). This is a statutory survey conducted under the Statistics of Trade Act 1947. Organisations are obliged under this legislation to provide a response. Businesses are sampled from the ONS business register current at the time of drawing the sample: first the CSO Business Register, which ran until 1993; then the Inter-Departmental Business Register, which has run from 1994 onwards. The ONS holds firms' responses to the ABI in the Annual Respondents Database (ARD).

The ABI replaced the following annual survey systems in 1998:

• Annual Employment Survey (AES)
• Annual Censuses of Production and Construction (ACOP/ACOC), which include the Purchases Inquiry (PI)
• The six annual Distribution and Services (DSI) inquiries (Annual Wholesale Inquiry; Annual Retail Inquiry; Annual Motor Trades Inquiry; Annual Catering Inquiry; Annual Property Inquiry; and Annual Service Trades Inquiry

Until 1997 the data were limited to the production and construction industries surveyed by the ACOP and ACOC (construction from 1993 only). The incorporation of the DSI inquiries for six additional sectors is reflected in the number of individual business contributors rising from approximately 15,000 for 1980 to 1996 to approximately 50,000 for 1997/98 and to over 70,000 for 1999.

The ABI is one of the most comprehensive surveys undertaken of business organisations in the UK, covering over 100 key economic variables, and approximately two-thirds of the UK economy. Detailed variables for turnover, employment, costs, capital and the derivation of sales and profits are included. A firm-level measure of Gross Value Added (GVA) is also generated so that the productivity of organisations can be evaluated.

The ABI samples UK businesses and other such establishments according to their employment size and industry sector. It is a census of large businesses, and a stratified sample of small and medium sized enterprises. The stratified sampling framework means that smaller firms move in and out of the survey. The forms are customised for industry sectors and sub-sectors. The statistics produced from the sample data are used primarily to assist in the generation of the National Accounts and the measurement of Gross Domestic Product (GDP).

A number of different form-types are used in the survey. Long form-types are sent to all businesses with an employment of 250 or more and also to a proportion of selected businesses with lower employment. Short form-types are sent to the remaining selected businesses. The forms differ in that long form-types ask for a detailed breakdown of purchases; employment costs; taxes, duties and levies etc, whereas short form-types just ask for the totals of these variables.

The data are collected in two parts: Part 1 is an employment record, collected as soon as possible after 12th December. Part 2 is for financial information, which may be submitted up to twelve months after the financial year end.

Geographical references: postcodes
The postcodes available in these data are pseudo-anonymised postcodes. The real postcodes are not available due to the potential risk of identification of the observations. However, these replacement postcodes retain the inherent nested characteristics of real postcodes, and will allow researchers to aggregate observations to other geographic units, e.g. wards, super output areas, etc. In the dataset, the variable of the replacement postcode is 'new_PC'.

Linking to other business studies
These data contain Inter-Departmental Business Register reference numbers. These are anonymous but unique reference...