Accuracy and reliability

Overall accuracy

Only 2,651 households out of the 32,150 contacted chose to participate in the 2024 Household Budget Survey. For total consumption in the entire population, there is a margin of uncertainty of ±1.4 percent, meaning that the estimated total consumption of 377,034 DKK is expected, with 95 percent confidence, to fall between 366,740 DKK and 387,329 DKK. The uncertainty is greater for specific population groups and for goods that are purchased infrequently. Consumption estimates are therefore more precise for, for example, households with higher incomes or where the main income earner is of Danish origin, is employed, is over 49 years old, or has an education beyond compulsory school, as these groups are relatively more represented among survey participants.

Sampling error

The total sample for the 2024 Household Budget Survey (HBS2024) consisted of 32,150 households selected in 2023 (12,675) and 2024 (19,475). In 2023, 1,148 households participated, while 1,503 participated in 2024, giving a total of 2,651 participants in HBS2024. The participation rate for HBS2024 was therefore 8 percent. Sample uncertainty is calculated for each consumption group based on the households that chose to participate in the survey. For HBS2024, sample uncertainty is based on responses from the 2,651 participating households. Sample uncertainty is quantified using the coefficient of variation (the relative standard error).

The coefficient of variation for total household consumption in HBS2024 was 1.4 percent. This means that the 95‑percent confidence interval for total household consumption is 377,033 DKK ± 10,295 DKK. There is considerable variation in sample uncertainty between total consumption and individual consumption items. Items that are purchased frequently have lower sample uncertainty than those purchased infrequently. Bread, for example, is a product group that most households purchase regularly during the two‑week diary‑keeping period associated with the survey. The coefficient of variation for bread for an average household is 1.8 percent, whereas for a rarely purchased item such as offal, it is 18.6 percent. Similarly, the coefficient of variation for bread is 3.4 percent in the Capital Region (Region Hovedstaden) and 4.6 percent in the North Denmark Region (Region Nordjylland).

Non-sampling error

The sampling frame for the survey consists of occupied private household addresses in Denmark in a given year. The extraction frame, however, includes both private households and collective households. To avoid including collective households in the Household Budget Survey, only households with a maximum of eight residents are selected. This restriction may introduce some bias, as larger private households (with more than eight persons) are not included, but this bias is assessed to have limited impact on the results.

The frame population is based on two calendar years and must represent the target population for a specific year. The target population is therefore represented by combining the frame populations from 2023 and 2024. In general, the frame population from 2023 is considered to be largely representative of 2024, although some households may change characteristics from one year to the next.

Analyses have shown that households typically record fewer purchases in the second week than in the first week—both in terms of number of purchases and total value. The reasons are not entirely clear, but households may forget to register expenses or become more aware of their consumption due to the recording exercise, resulting in fewer purchases. To adjust for this decline, second‑week diary data are upweighted by 4 percent, based partly on previous analyses.

Households may also tend to report consumption that deviates from actual behavior. For example, they may record more organic products than they actually purchase, as organic consumption is viewed positively. Conversely, they may underreport consumption of more negatively perceived items such as wine, spirits, and cigarettes. This can lead to either over‑ or underestimation of consumption.

During the 2024 data collection year, we improved the respondents’ data collection tool. It is now possible for households to scan or photograph receipts, which are then uploaded automatically. Using AI, items and amounts are categorized by purpose.

This improvement has not only reduced respondent burden but also increased data quality, as we now receive households’ actual receipts. In the longer term, we expect that second‑week diary data will no longer need to be upweighted by 4 percent.

Interviewers are instructed on how to guide households in completing the diary forms to minimize errors. We have also developed instructional videos explaining what should be included. Households are asked to record their consumption every day during the 14‑day diary period. Interviewers check in approximately halfway through the period to ensure households have begun recording as agreed.

Of the 32,150 randomly selected households, 29,499 declined to participate in the 2024 Household Budget Survey. This corresponds to an overall non‑response rate of 92 percent. Among the non‑participants, the majority explicitly chose not to participate, while 39 percent were households that could not be reached. The remaining cases were due to other types of non‑response. In total, 2,651 households chose to participate—8 percent—and these form the basis for the survey estimates.

Among the 8 percent who participated, there is an overrepresentation of ethnic Danes, pensioners, employed persons, and employees at basic skill levels compared with the population as a whole. There are also relatively more households with persons aged 60 or older, and fewer young households, as well as fewer households where the main income earner has only compulsory schooling. Weighting procedures aim to correct for these imbalances.

Fewer households with the lowest incomes participate. Although weighting adjusts for this, the higher participation among high‑income households may still lead to overestimation of consumption. The aim is for at least 1,600 households to participate each calendar year to ensure the desired sample uncertainty of 1 percent. The sample size is adjusted accordingly based on expected response rates. Register data—such as income and taxes—are used to supplement the sample.

In recent years, it has become increasingly difficult to obtain household participation. This is partly because interviews have shifted from CAPI (personal, face‑to‑face interviews) to exclusively CATI/CAWI (telephone/web interviews). It is easier for respondents to decline participation when they must answer independently via web or phone compared with when an interviewer is physically present at their doorstep. It has also become harder to reach households due to GDPR rules limiting access to phone numbers, as well as an increasing number of Danes using employer‑provided phones that are not publicly registered.

The Household Budget Survey is based on a two‑year moving average, where data from year 1 are adjusted for prices and volumes to match the level of year 2 (the reference year).

The sampling unit in the survey is the household address, but the target is private households. Because addresses are selected randomly, some collective households may be included, even though addresses with more than eight residents are excluded from the sampling frame. However, only a small number of selected addresses fall into this category and are screened out during data collection.

The number of residents and households at each address is determined by the interviewer and the respondents. The two‑year moving average requires several assumptions about the weighted population. For example, if a household of four (two adults and two children) participated in 2023 and is selected again in 2024 but has dissolved, it is treated as unchanged. Likewise, it is assumed that deaths and emigration are offset by births and immigration.

Another source of uncertainty arises when we use data from external sources that we cannot influence ourselves. The major revision of the National Accounts in June 2025, which was applied retroactively, has likely affected the FU2023 results. When the 2022 figures were converted into price and volume measures, the quarterly National Accounts data used in the calculations had not yet been updated as part of the revision. As a result, the calculated levels for FU2023 may be slightly higher than they otherwise would have been.

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 Household Budget Survey is a sample survey, combined with a wide range of register variables. The survey is based on a simple randomly drawn sample. The survey has over a number of years been conducted by the same procedure.

The study's annual sample is not large enough to only give a reliable estimate of consumption. Therefore included sample data from the previous years also in the total charge for the year. The study has the character of a sort of "moving average".

Through several years the study has been conducted by the same procedure. This has been conducted partly to ensure a better comparison over time and partly to provide faster results and to minimise errors.

The Household Budget Survey was in the period 1994-2013 based on a three-year sample. From 2014 and onwards, the Household Budget Survey is based on a two-year sample. This is an improvement of the Household Budget Survey, since the figures brought forward by 12 months.

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

Since the publication of the Household Budget Survey 2017, there is only an annual publication, therefore no revisions will be made in the future.