Skip to content

Accuracy and reliability

Contact info

Science, Technology and Culture
Anton Erenbjerg
+45 20 14 57 90

ael@dst.dk

Get as PDF

ICT Use in Enterprises

The results from the survey for enterprises with at least 10 full-time employees are based on data from 4,557 enterprises from a total population of 18,213 enterprises. The survey is sample based and consequently there is some uncertainty in the results in the form of random variation from the branch and enterprise size. Every year uncertainty calculations are produced, and these show that the sampling uncertainty is limited.

Overall accuracy

The total uncertainty concerns primary unit and item non-response - both of which only to a limited extent affect the uncertainty of the survey results. The enterprises' responses were collected from January to July 2023. The response rate for the survey is very high (approx. 98 per cent), but during COVID-19 the response rate was slightly lower (96-97 per cent). The lower response rate only affected the results to a very limited extent.

Sampling error

The results are based on data from 4,557 enterprises from a total population of 18,213 enterprises with 10 or more persons full-time employed. The overall response rate is 98 percent.

Every year uncertainty calculations are produced, and these show that the sampling uncertainty is limited. Examples of variables from the survey 2023, with estimated share of enterprises and associated 95 percent confidence interval, lower and upper:

  • Enterprises with websales (total): 37 pct. (35; 39)
  • Enterprises using cloud computing (total): 69 pct. (67; 71)
  • Enterprises using ERP: 67 pct. (65; 69)

Put differently, the above implies that the share of enterprises with websale, with a probability of 95 percent, is between the lower and upper share in brackets above (35; 39), but that the estimated share in the statistics is the percentage prior to the brackets (37 percent).

For categorical variables in the 2023 survey published results, the absolute sampling error is at most 2%. with a 95 percent. statistical confidence interval for the entire population.

Uncertainty is higher when breaking down by activity or size class. For categorical variables, results for subgroups (main branch crossing with enterprise size) are published in the StatBank if the absolute uncertainty is below 7.5 percent associated by 95 percent confidence interval.

Non-sampling error

Other uncertainty relates to unit and item non-response, and this is in both cases limited in scale.

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 population covers private enterprises with 10 or more persons employed. The population covers private, non-financial enterprises (the financial sector is covered in the reference periods 2005 to 2010).

The net sample in 2022 consists of 4,325 enterprises from a total population of 18,213 formed the sample for the survey of enterprises with 10 or more persons employed. The sample is stratified by activity and size class. The response rate for the survey is very high (approx. 98 per cent), but during COVID-19 the response rate was slightly lower (96-97 per cent). The lower response rate only affected the results to a very limited extent.

The survey is questionnaire based, and data is collected by digital reporting through http://www.virk.dk. Validation comprises macro and micro validation and some extent of recontact with enterprises.

In the survey design, each observation will represent more elements in the population. The weights assigned to each observation are calcuated using generalising regression estimates.

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

Not relevant for this statistics.