Accuracy and reliability

Overall accuracy

The survey result is based on submission of data from all relevant Danish parent companies with foreign affiliates. Thus the number of parent-companies in the survey is approx. 5.000 and the number of foreign affiliates is approx. 15.000.

The statistical results are considered highly accurate, however a minor underestimation may occur, e.g. due to missing data submissions and/or missing knowledge of relevant foreign affiliates. This however is appraised to be insignificant for the statistics. The experience is significant unit non-response or lacking inclusion of foreign affiliates only occur very rarely, and underestimation is assessed to be less than 1 per cent in terms of the number of foreign affiliates, and considerably below 1 per cent in terms of the number of employees.

Number of employees should be reported as the average number of employees in the foreign affiliates for the reference period, regardless of these employees work full-time or part-time. Thus meaning that the number reported should be based on head-count. It can be challenging for the reporting unit to deliver this information, hence that the number of employees reported could in fact be number of full-time equivalents (FTE's) and/or reported as end-of-year employees (and not the annual average). This is very difficult for Statistics Denmark to intercept in the data validation.

The foreign affiliates are, due to their geographical localization throughout the world, subject to different national regulations on annual reports, e.g. in regards to when these reports are to be delivered/made public. Therefore the submitted survey data could in fact belong to a previous period, as the reporting units does not have access to any newer data.

The data providers, which are a source for the survey, base their knowledge on foreign affiliates and their association to a Danish company, on world-wide annual reports from the companies and affiliates. This might mean that a company, with a relevant foreign affiliate, first will be a part of the survey a year later than it should have been. This discrepancy is due to the fact that the source material used in population determination is incomplete when it is needed, as e.g. the relevant annual report not has been made public at that time. This, however, is mainly the case for reporting units that are not already a part of the survey, as existing reporting units in the survey that have additional affiliates are required to add these to the form.

Sampling error

Not relevant.

Non-sampling error

The response rate for Danish foreign affiliates is generally on more than 95 per cent. The statistical results are considered highly accurate, however a minor underestimation may occur, e.g. due to missing data submissions and/or missing knowledge of relevant foreign affiliates. This however is appraised to be insignificant for the statistics. The experience is significant unit non-response or lacking inclusion of foreign affiliates only occur very rarely, and underestimation is assessed to be less than 1 per cent in terms of the number of foreign affiliates, and considerably below 1 per cent in terms of the number of employees.

The data providers, which are a source for the survey, base their knowledge on foreign affiliates and their association to a Danish company, on world-wide annual reports from the companies and affiliates. This might mean that a company, with a relevant foreign affiliate, first will be a part of the survey a year later than it should have been. This discrepancy is due to the fact that the source material used in population determination is in-complete when it is needed, as e.g. the relevant annual report not has been made public at that time.

Number of employees should be reported as the average number of employees in the foreign affiliates for the reference period, regardless of these employees work full-time or part-time. However this is very difficult for Statistics Denmark to detect in the data validation.

As no sources are available the survey is not being compensated for unit and item non-responses. The extent of item non-response is not monitored and is therefore not able to be calculated. The impact of unit and item non response is estimated to be of small extent.

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.

The survey result is based on submission of data from all relevant Danish parent-companies with foreign affiliates. Any sample techniques and/or active use of thresholds have not been implemented. Thus the number of parent-companies in the survey is approx. 5.000 and the number of foreign affiliates is approx. 15.000.

With the sources used and available it is appraised that the value and results of the survey presents a true and fair view, however underestimation may occur, e.g. due to missing data submissions and/or missing knowledge of relevant foreign affiliates.

Number of employees should be reported as the average number of employees in the foreign affiliates for the reference period, regardless of these employees work full-time or part-time. Thus meaning that the number reported should be based on head-count. It can be challenging for the reporting unit to deliver this information, hence that the number of employees reported could in fact be number of full-time equivalents (FTE's) and/or reported as end-of-year employees (and not the annual average). This is very difficult for Statistics Denmark to intercept in the data validation.

The foreign affiliates are, due to their geographical localization throughout the world, subject to different national regulations on annual reports, e.g. in regards to when these reports are to be delivered/made public. Therefore the submitted survey data could in fact belong to a previous period, as the reporting units does not have access to any newer data.

The data providers, which are a source for the survey, base their knowledge on foreign affiliates and their association to a Danish company, on world-wide annual reports from the companies and affiliates. This might mean that a company, with a relevant foreign affiliate, first will be a part of the survey a year later than it should have been. This discrepancy is due to the fact that the source material used in population determination is incomplete when it is needed, as e.g. the relevant annual report not has been made public at that time. This, however, is mainly the case for reporting units that are not already a part of the survey, as existing reporting units in the survey that have additional affiliates are required to add these to the form.

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 disseminated. In general there are no regular revisions of the statistics.