Accuracy and reliability

Overall accuracy

Calculation of the Construction cost index for residential buildings is based on the assumption that three chosen housing constructions are valid representations of the actual costs of constructing a house in Denmark. The three houses were chosen on the basis of an analysis of actual constructions of residential buildings, using information obtained primarily from the Dwelling- and Housing Register (BBR). In the case of significant changes to how houses are typically constructed the accuracy of the index could be affected.

The index is based on the development of about 200 representative goods, which cover about 20 work functions, about 52 material groups and about 10 equipment groups. Each representative good is linked to a number of prices. It is not possible to determine sampling accuracy, as prices are not collected randomly. However, representative goods are chosen on the basis of the costs of three actual, typical housing construction projects, using principles of importance and representation. It is therefore assumed that price developments of the collected sample prices are representative of price development in actual housing constructions.

Continuous monitoring and improvement of the quality of pricing data and weights means that the index is evaluated as being of high quality and representative of the general development of house building costs in Denmark.

Sampling error

The index is based on the development of about 200 representative goods which cover about 20 work functions, about 52 material groups and about 10 equipment groups. To each representative good a number of prices/goods are attached. It is not possible to determine sampling accuracy, as prices are not collected randomly. However, representative goods are chosen on the basis of the costs of three actual, typical housing construction projects, using principles of importance and representation. It is therefore assumed that price developments of the collected prices are representative of price development in actual housing constructions.

The Price index for Domestic Supply, from which most of the prices used in the Construction cost index for residential buildings are gathered, uses a top-down principle by which a minimum of 70 pct. of the Danish production and import is covered. The used samples are therefore considered to be representative of actual price developments. Because reporting companies are not selected randomly it is not possible to measure the sample error. A source of error may be if prices of relevant goods that are not included in the index differ significantly from prices of included goods.

The sample used in the implicit indices of average earnings includes a large proportion of the total population, i.e. by inclusion of all companies with more than 100 employees. Therefore, the statistical error may be higher for industries that consist of many smaller businesses.

Non-sampling error

The three house constructions that the Construction cost index for residential buildings is based were chosen on the basis of an analysis of actual housing constructions in 2015, using information obtained primarily from the Dwelling- and Housing Register (BBR). They were chosen to ensure a representative basis for calculations of the index. Therefore, the Construction cost index for residential buildings is based on the assumption, that the chosen buildings represent typical constructions of single- and multifamily buildings in Denmark. This also means, that the index may be less representative of housing projects that are distinct, architecturally or otherwise, from typical housing projects.

In the case of significant changes to how houses are typically constructed the accuracy of the index could be affected. This is addressed by updating the base weights approximately every five years.

The construction cost index for residential buildings is based on a fixed basket of products. However, in real life companies may substitute what products they use, e.g. due to pricing differences for similar products, due to company preferences, or because of technological advances. This may lead to the index diverting from actual cost developments. To accommodate this, Statistics Denmark is in running contact with reporting companies about the relevancy of their reported products. Substitution of the products used in the index may lead to changes in product quality, which is addressed by continuous quality corrections.

Errors may occur during registration of prices and labor costs by reporting companies or at Statistics Denmark. However, both manual and computerized inspections of such errors are performed, and is therefore not considered to be a significant source of error.

Missing price reports are found to be less than 1 pct. per index period. Missing prices are mostly estimated by a continuation of the last reported price. In the case of essential products, prices may be imputed. In cases where the price is unchanged over a period of 13 months, the sources are contacted to investigate if their reporting is erroneous.

Calculation of the Construction cost index for residential buildings is done by computer in a dedicated price index system. The likelihood of calculation errors is therefore inconsiderable.

It is assumed that developments of the prices used in the Construction cost index for residential buildings are accurate and representative of actual price developments for relevant product groups. The products used in the index have not been chosen specifically for use in the index, but it is assumed that these products are representative of materials used in the construction of dwellings.

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

Overall the Construction Cost index for residential buildings is assessed to be of high quality and representative of the construction of residential housing in Denmark. The weight basis for the index is based three housing constructions: a one-family house, a terraced house and a multi-family house. They were chosen through a thorough analysis of typical residential, housing constructions in Denmark in the weight reference period (2015). Every five years the weight basis is updated in order to reflect the contemporary state of housing construction.

It is not possible to specify the total error related to the of the index. The gathered price data is continuously monitored in order to raise the data quality. This may include substitution or addition of reporting companies, changes in the reported products, or changes in applied price definitions and calculations. Running dialogue is carried out with involved companies to prevent misunderstandings and reporting errors.

In order to minimize errors, the index is calculated in a dedicated computerized pricing index system.

In general, it is assessed that the error margin is highest at the most detailed, least aggregated index levels, and drops at less detailed, more aggregated levels. The on-going monitoring and improvement of the quality of both incoming data and the weight basis means that the Construction Cost index for residential buildings is assessed to be of high quality and representative of the construction of residential housing in Denmark

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 statistics are calculated and published. Revisions are only performed if errors are detected in already published values.