Accuracy and reliability

Overall accuracy

In general, the quality of the statistics is rated to be acceptable, with a few reservations. It can be problems in comparing figures over time. This is because the data are only collected every five years, so the data only available for 1998, 2002, 2007 and 2012. Moreover, there are break of data from 2012 due to the changes in industry codes (from NACE rev.1 to NACE rev. 1.1 and to Nace rev. 2) and classification of products (from CPA2002 to CAP2008).

The survey is based on a sample and thus some uncertainty is attached to the results. For the whole retail trade sector, however, the uncertainty is considered to be small, since the coverage is extensive and the response rate high. For a few sectors with a relatively higher share of small enterprises, the limit of DKK 2.5 mil. may imply a higher uncertainty due to lower coverage.

Sampling error

The total number of enterprises in the population is estimated at approx. 10,000, and among these the sample is selected. The sample for the reference year 2012 was 2,000 companies and is thus approx. 20 per cent. of the total number of the population. Because of the stratified selection, where large companies are more likely to selection than the smaller, the sample consists approx. 82 per cent. of the total turnover of the population.

The sampling error is not calculated for the reference year 2012.

Non-sampling error

The total uncertainty consists of sampling uncertainty and other systematic uncertainty. In other uncertainties include inter alia measurement error and non response error/lapse.

Coverage error

The population is defined by the industry, the enterprises registered in the Business Register by themselves. Errors of classification of the industries will be expected. This means that there may be enterprises in the population that really should not be classified as an enterprise in the Retail Trade, but also companies in the Retail Trade can be classified with the wrong industry.

In the context of fault tracing and quality assurance of the statistics, number of these errors continuously are found and corrected. Companies that incorrectly classified as Retail Business cause over-coverage while misclassifications within Retail Trade cause over-coverage in one industry but under-coverage in another. These coverage error is detected primarily for companies in the sample, but also in some cases for non-sampled companies. Under- coverage in the form of companies that are classified entirely outside Retail Trade, even though they should properly be classified as Retail Business, are difficult to detect.

The total number of enterprises in the population is estimated at approx. 10,000, and among these the sample is selected. The sample for the reference year 2012 was 2,000 companies and is thus approx. 20 per cent. of the total number of the population. Because of the stratified selection, where large companies are more likely to selection than the smaller, the sample consists approx. 82 per cent. of the total turnover of the population.

Measurement error

Uncertainty due to lapse, incorrectly reported numbers and misunderstandings has sought to minimize by repeated reminders by non-reporting, as well as by substantial verification of the reported figures. Generally the services in the questionnaire are based on Statistical Classification of Products by Activity CPA 2008, with an adjustment to Danish conditions. It can be an uncertainty in the distribution of turnover by products when CPA 2008 are not quite commensurate with the classification of products of the firms.

Non response error

There are 2,000 enterprises in the sample. The usable questionnaires are from 1,777 enterprises, i.e. there is a lapse of 11 per cent.

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

In general, the quality of the statistics is rated to be acceptable, with a few reservations. It can be problems in comparing figures over time. This is because the data are only collected every five years. The survey has been carried out for the reference periods 1998, 2002, 2007 and 2012. Since the 1998 survey, changes have been made in both calculation method and industry and product group nomenclatures.

The 1998 and 2002 surveys use the Danish Industrial Classification 1993 (DB93) and the Statistical Classification of Products by Activity 1996 (CPA 1996). The 2007 surveys uses the Danish Industrial Classification 2003 (DB03) and the Statistical Classification of Products by Activity 2002 (CPA 2002). The 2012 surveys uses the Danish Industrial Classification 2007 (DB07) and the Statistical Classification of Products by Activity 2008 (CPA 2008).

Comparisons across reference years should be done with caution and taking into that statistical nomenclatures and methods have changed.

The survey is based on a sample and thus some uncertainty is attached to the results. For the whole retail trade sector, however, the uncertainty is considered to be small, since the coverage is extensive and the response rate high. For a few sectors with a relatively higher share of small enterprises, the limit of DKK 2.5 mil. may imply a higher uncertainty due to lower coverage.

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 these statistics.