Registers and reference types
Statistics Denmark has gathered a vast series of historical register data in our databank of basic data, which users can access via the platform Denmark’s Data Portal. Research Services manages the databank of basic data and handles access to the platform, support, etc. Most registers in the databank are updated at least once a year in connection with release of the register-based statistics (, see Scheduled releases, ). , The data safari and the List of registers and variables (below) both show the registers in Denmark’s Data Portal, and here you can see variables for the individual registers. The documentation of variables is available in Statistics Denmark’s , documentation system, ., Go to Data safari , Go to List of registers and variables (in Danish), , Overview of rerun registers (in Danish), Genkørte registre 2025-1. kvt (pdf), Genkørte registre 2024-4. kvt (pdf), Genkørte registre 2024-3. kvt (pdf), Genkørte registre 2024-2. kvt (pdf), Genkørte registre 2024 - 1. kvt (pdf), Genkørte registre 2023 - 4. kvt (pdf), Genkørte registre 2023 - 3. kvt (pdf), Genkørte registre 2023 - 2. kvt (pdf) , Genkørte registre 2023 - 1. kvt (pdf), Genkørte registre 2022 (pdf), , Reference types,
Registers in the basic data overview are compiled by means of different reference types. Next to each register in the basic data overview, you can see which reference type a register has: ’Status’, ’Statusperiode’ (status period), ’Forløb’ (longitudinal) or ’Hændelse’ (incident).,
Status, The reference type shows the status for a given date. For example, LONN (structure of earnings), which shows what a citizen earns as of the register date (e.g. 31 December 2021). Or BEF, which shows the population as of the quarter date (including status of residence, age, family, etc.)., Data definition: Clear status as of a given date. The population delimitation and all data content is focused on the date., Status period, This reference type shows the period status, where the population is delimited as of a given date, but the variables contain summed up data for a specific period. For example, IND, which contains the labour income for a year (the period appears from ’Opdateringsfrekvens’ (update frequency) in the basic data overview). Other examples of status period registers: PERSBEST (board members and managers), MFR (medical birth register), HANDICB (financial support for disability cars), DMRB (motor vehicles). It is not always easy to see what is being summed up., Data definition: The population delimitation is made as of a given date, but the content of the variables is accumulated over a given period. The period cannot be deduced from dates in microdata, but from the indicated period (shown under ‘Opdateringsfrekvens’ (update frequency)) – meaning that content in for example amounts, volumes, quantities etc. is aggregated over the indicated period (e.g. a quarter, a year)., Longitudinal,
Here, data covers a longitudinal study. There will always be just one version of the register available. For example, UDD, which contains Highest educational attainment. Or BEFADR, which is an address key register (where e.g. 1.4m addresses changed key on 1 January 2007 in connection with the local government reform). When a longitudinal register is updated, the individual dataset is updated. This is why there is always only one dataset for a longitudinal register.,
Data definition: The definition of longitudinal data is that data contains a start date and an end date., Incident,
Here, data covers an incident. For example, UDFK, which contains primary and lower secondary school marks (does not include a date but a school year), or OPHGIN (basis of right of residence for immigrants). When a longitudinal register is updated, the individual dataset is updated with new incidents. This is why there is always only one dataset., Data definition: The definition of incidents is first and foremost that data contains a date - only one date - for the occurrence of the incident, and will usually also have one incident type attached.,
Documentation for the use of registers, Statistics Denmark has prepared a memo describing the coherence between several of the most used registers in Statistics Denmark’s microdata scheme and their connection with the published statistics., The social statistics registers in Statistics Denmark consist of comprehensive data collections, which have been built and extended since the early 1980s. Data is of high quality and comprises the whole population. This gives the users of data unique possibilities of analysis, allowing them to analyse both status at a given point in time and the development over time., The memo is primarily intended for researchers, analysts and other users of microdata who want to obtain deeper insight into the quality of the coherence between the different registers. , Read more on Documentation for the use of registers (in Danish),
Especially on the Data Warehouse for Business Statistics, In January 2024, Statistics Denmark launched the new Data Warehouse for Business Statistics – a significant extension and improvement of the existing business registers. , The new warehouse ensures wider and better access to anonymised data on enterprises and facilitates extraction of unique data by linking data across more statistical registers. The data warehouse also facilitates linking of business statistics and social statistics at micro level, the so-called ‘Linked Employer-Employee Data’ (LEED). , Read more in , this brochure (pdf), or see , the presentationen of The Data Warehouse for Business Statistics on 30 November 2023 (pdf), .
https://www.dst.dk/en/TilSalg/data-til-forskning/generelt-om-data/registre-og-referencetyper
