Data are collected by web from 30 enterprises for the sector Market Research and Public Opinion Polling. In the questionnaire the enterprises are asked to breakdown their turnover by a number of products, services and client categories. The reported data are checked by comparing the distribution of turnover among enterprises that are similar. Survey data is grossed up in part by including information from administrative sources on the VAT turnover in the reference year for the entire population.
The statistics is based on a sample of enterprises in the sector for Market research and public opinion polling. The enterprises are selected based on their number of full-time employees. The number of full-time employees is calculated based on the average for full-time employees in each of the four quarters of the year.
The sample is stratified random sample consisting of all enterprises with 50 employees or more and a part of enterprises with 5-49 employees. The sample is divided into 5 size groups: 5-9, 10-19, 20-49, 50-100 and more than 100 full-time employees. Each year a new sample is drawn and the population consists of enterprises in the Business Register System, which are active in the drawing time.
Data are collected every second year.
Data are collected by web questionnaires. In the questionnaire the enterprises are asked to distribute their turnover in DKK 1,000 by a number of products, services and client categories.
The reporting is mandatory, i.e. the enterprises are required to report accurate data on time. If the data has not been received , the companies will be reminded 3 times by mail or telephone. After that a registered letter will be sent to inform the enterprises that if the information are not submitted within one week, Statistics Denmark will refer the matter to the police for requesting the imposition of a fine. And the enterprises still have to give us the data we have asked for.
The submitted data are checked for errors, ex. data of the individual reporter are compared with prior periods and similar enterprises reporting in the current period. If the data are significantly different than expected, the enterprise will be contacted. It must be assumed that not all errors in the data are found. That is why the statistics may be subject to some uncertainty associated with error reporting.
The data from the questionnaires received are grossed up to the provisional total turnover for the statistics Purchases and sales by enterprises.
The data are corrected for bias in the sample and response pattern. A so-called ratio estimate is used, which includes information on the taxable turnover of the year concerned for the entire population of the sector for Market research and public opinion polling.
Not relevant for these statistics.