Statistical processing

Source data

The Cultural Habits Survey data is compiled from questionnaire surveys among a random sample and from demographic information drawn from Statistics Denmark's population register. The random sample is drawn randomly with equal likelihood for each person in the population. Persons without a registered address in the Civil Registration System, with a town hall address or with a non-public address are not eligible to be drawn. Every quarter a sample of 15,100 persons is drawn. This sample is divided into four groups which each participant is assigned to with equal likeliness. The groups are used used to divide respondents across four different versions of the survey questionnaire.

Survey questionnaire for the Cultural Habits Survey 2024 in Danish.

The statistic on membership for youth and outdoor activities organisations is compiled on the basis of reports from the following organisations:

• The Danish Scout Association
• FDF - a Christian organisation for children and youth in Denmark
• The Green Girl Guides in Denmark
• YMCA Scouts in Denmark
• Danish Baptist Scout Association
• DUI-LEG og VIRKE - a children's association
• DCU (Danish Camping Association)
• The Danish Society for Nature Conservation
• Danish Association of Hunters
• The Danish Horticultural Society
• The Danish Ramblers' Association
• DOF - The Danish Ornithological Society

Frequency of data collection

Data for the Cultural Habits Survey is collected quarterly.

Data for the membership for youth and outdoor activities organisations is collected yearly.

Data collection

Data for the Cultural Habits Survey is collected by the in-house DST Survey in cooperation with Norstat through a combination of computer-assisted telephone interviews (CATI) and computer-assisted web interviews (CAWI). In the first quarter of 2024, the interview type was distributed across 20 pct. CATI and 80 pct. CAWI. The questionnaire consists of a shared part answered by all respondents, as well as four sub-parts answered by approximately a fourth of respondents interviewed by CAWI. Respondents interview by CATI is only posed the questions in the shared part.

The statistic on the membership for youth and outdoor activities organisations is collected through written reports attained through direct communication with each youth and outdoors organisation.

Data validation

Data is validated primarily during data collection. Answers to questions earlier in the questionnaire are used later to avoid posing questions that would results in non-consistent or logically impossible answers. E.g. if a respondent says that they have attended a performing arts performance, they are not asked about the reasons why they have not attended a performing arts performance. Data is also compared to results from previous surveys and large deviations are investigated.

Data compilation

Calculation of weights to enumerate the entire population from the information in the sample is a crucial part of the data processing of the survey. The purpose of the weighting is to partially compensate for the systematic non-response of the sample. Each respondent is assigned a weight such that the respondents as a weighted group resemble the target population as measured on a series of demographic variables. These variables are drawn from Statistics Denmark's population register and consists of gender, whether the respondent is of Danish origin or not, age in seven groups, education in three groups and family income in three groups.

Five weights have been calculated: one for the shared part and one for each of the four sub-parts. The more systematic the non-response, the greater the variation of the sizes of the weights. If the weights are standardised such that the average across all respondents is 1, the weights for the shared part in the first quarter of 2024 have a standard deviation of 0,45 and the weights for the sub-parts have standard deviations of between 0.48 and 0.64.

Data is also processed by creating derived variables. E.g. variables indicating participation in each cultural activities within the last 12 months are derived from variables indicating frequency of participation.

Yearly datasets are created by compiling data for the each years four quarters into one dataset. New weights are calculated by dividing the quarterly weights by four.

Adjustment

Not relevant for these statistics.