Accuracy and reliability
Contact infoShort Term Statistics, Business Statistics
Simon B. Halifax
+45 39 17 30 88
Get as PDF
Sampling error is estimated to plus/minus 1-2 percentage points for the confidence indicators. Non-sampling error occurs in connection to non-responses and drop out of the sample. Especially big companies' non-responses a certain month may influence the development from month to month. Finally, there is measurement error connected to the statistics associated with the respondents level of information and knowledge about the company's current state.
The statistics covers a substantial part of enterprises in the selected industries and is considered as statistically representative. In general, sampling error at plus/minus 1-2 percentage point is expected for the indicators. Non-sampling errors are caused by non-response errors (about 23 pct. non-responses), and especially missing responses from large companies may cause fluctuations from month to month.
Also measurement errors influence the statistics. A senior person in the company is asked to check mark his perception of the developments over the past three months and his expectations for the next three months. Those assessments are associated uncertainty, depending on the respondent's current level of information and knowledge about the current state of the company.
The survey variables are mostly categorical - with answer options of the type increase, unchanged or decrease. For each month' statistics is calculated confidence interval with upper and lower values for the three variables constituting the Composite Confidence Indicator. The confidence interval varies from month to month, and depends on the actual number of responses in each answer category. In general, a sampling error at plus/minus 1-2 percentage point is associated the indicators.
There is no systematically checking for errors. By sample update, new enterprises may inform they are not registered in correct business sector and therefore not relevant. Between lines of businesses there may be some over-coverage because of wrong or changed registration.
The tendency surveys are subjects to measurement errors: A senior person in the company is asked to check mark his perception of the developments over the past three months and his expectations for the next three months. The expectations are inherently uncertain. The perception of the development over the past may also be associated uncertainty, depending on the respondent's current level of information.
Several types of non-responses are associated the survey.
- There are non-responses because of liquidation or because an active statement about not to participate in the survey.
- Then there are unit non-responses, about 23 pct. The respondent stays in the sample but has not responded the current month or responds after ultimate deadline.
- Item non-responses are few, and is estimated not to exceed the unit-non-response level with more than a couple of percent.
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.
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.
The statistics covers a substantial part of enterprises in the selected industries and is considered as statistically representative. The strength of the statistics is that results are published in same months as data is collected and related to. Variations in figures from month to month should be interpreted with caution, since it is not necessarily a development in trend, but rather attributed to the uncertainty of the statistics.
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 figures are published in the survey results.
The published figures are not subject to revision but for investments there are four measurements for each year (October the preceding year, April same year, October same year and the final numbers in April the following year). In IFI01 and IFI02 only the latest version is shown. In IFI01R and IFI02R the revisions are shown so you can see all four estimates of a given year come April the following year.