Skip to content

Accuracy and reliability

Contact info

Science, technology and culture
Anne Vibeke Jacobsen
+45 20 14 84 28

avj@dst.dk

Get as PDF

ICT Usage in Households and by Individuals

The survey is based upon simple random selection samples, and the results are therefore subject to statistical uncertainty. For the entire populations access to Internet the uncertainty is limited, since the degree of coverage is high. Uncertainty reflects variations in the collected data in comparison to the size of the sample. If all persons/families were identical with respect to access to and use of the Internet, then a sample size of 1 would in principle be adequate. The larger differences there are, the larger a sample is necessary to attain limited statistical uncertainty on the figures.

Overall accuracy

Coverage: Up to the 2006 survey results were based upon approximately 1,000 interviews each month. The coverage of the sample is therefore acceptable and very high for the annual publications 2001-2003 where interviews were held in several months. From 2007 and onwards, the survey is based upon approximately 4,300 completed interviews (16-74 years). Data collection takes place in April every year.

Sample/section: The sample is made by simple random selection from CPR. Up until the 2006 survey, a new sample was drawn every month among a representative section of the Danish population between 16-74 years. 1,000 interviews were carried out each month, from a sample of approximately 1,460 persons. Types of unit non-response are no telephone number for the dwelling unit, refusal and non-contact. In 2007 approximately 3,000 interviews were held in April, from a sample of approximately 4,500 persons. Types of unit non-response are the same as before. The 2009 survey is based on 4.303 completed interviews. The number of completed interviews was 4.600 in 2010, when coverage was extended from 16-74 years to 16-89 years. In 2013 the survey is based un 5.696 interviews (16-89 years).

Collection/measuring: Information is collected by telephone interviews. From 2008 it is possible for respondents to choose to supply information via a web-based questionnaire. Background information about the respondents is collected from register information form CPR.

Calculation method: After collecting the data the sample is grouped after for instance sex, age, civil status, family status and gross income. Within each group individual results are multiplied with an individual weight. The weight of person is equal to the number of persons in the CPR and the number of persons in the group. This weight is afterwards further adjusted up and/or down depending on the composition in the enumeration population with a regression estimate. The use of this method adjusts for an uneven loss in the sample.

There are calculations of sampling error for chosen questions and chosen groups.

Sampling error

Sampling error 2022 for the EU-questionnaire (Standard error in percentage points)

  • Proportion of individuals having ordered goods or services for private use over the internet : 0,52

The statistical uncertainty of the shares of individuals that have ordered goods or services for private use is calculated to ±0,52 per cent in 2022 with an interval of confidence of 95 per cent. Uncertainty increases in relation to background variables, e.g. type of occupation, since these shares are based on a smaller number of interviews. The uncertainty was for instance estimated to ±1,57 per cent for 2022 for 65-74 year olds who have ordered goods and services for private use.

Non-sampling error

There aren’t any questionnaires with errors, due to built-in filters that help the respondents. E.g. a respondent that has answered ‘no’ to Internet use, will not get questions on Internet activities.

Quality management

Statistics Denmark follows the recommendations on organisation and management of quality given in the Code of Practice for European Statistics (CoP) and the implementation guidelines given in the Quality Assurance Framework of the European Statistical System (QAF). A Working Group on Quality and a central quality assurance function have been established to continuously carry through control of products and processes.

Quality assurance

Statistics Denmark follows the principles in the Code of Practice for European Statistics (CoP) and uses the Quality Assurance Framework of the European Statistical System (QAF) for the implementation of the principles. This involves continuous decentralized and central control of products and processes based on documentation following international standards. The central quality assurance function reports to the Working Group on Quality. Reports include suggestions for improvement that are assessed, decided and subsequently implemented.

Quality assessment

The quality of the survey in general is appraised to be high.

In 2023 the EU-survey result are based on responses from 4080 respondents aged 16-74 years.

The survey is based on simple random samples and therefore the results are subject to a sampling error.

Data revision - policy

Statistics Denmark revises published figures in accordance with the Revision Policy for Statistics Denmark. The common procedures and principles of the Revision Policy are for some statistics supplemented by a specific revision practice.

Data revision practice

Only final indicators are published.