Accuracy and reliability

Overall accuracy

The weight bases of the indices for earthwork, asphalt work, concrete structures, iron and steel structures are based on an analysis of actual completed and ongoing civil engineering projects. The weights were established in 1976 in agreement with the Danish Roads Directorate, State Railways and the Danish Association of Builders. The weight basis for the construction cost index for roads was prepared in 1995 in collaboration with the Danish Roads Directorate on the basis of an analysis of various completed motorway and highway projects. The weights express how large a share labor costs and different materials and services constitute of the total cost of carrying out civil engineering work. The index is therefore based on the assumption, that the chosen civil engineering projects are representative of actual civil engineering work. Therefore, in the case of significant changes to typical civil engineering work, the accuracy of the index could be affected. The weight basis is reevaluated approximately every 5 years, lastly in 2015, where the original weight basis was found to still be adequate.

The statistics are primarily compiled on the basis of data from the Price index for Domestic Supply and wage rates from collective agreements between the Danish Association of Builders and the United Federation of Workers in Denmark (3F). 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 actual civil engineering 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 construction work. Users are referred to the documentation of statistics for the Producer and Import Price Index for Commodities.

Sampling error

Sample uncertainties are not calculated as data sources are not randomly selected.

The price index for domestic supply, from which most of the prices used in Construction cost indices for civil engineering projects 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. 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. For more information see the documentation of statistics for Producer and Import Price Index for Commodities.

Non-sampling error

Product substitutions: The Construction cost indices for civil engineering projects is calculated as a Laspeyres-type index with 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, as companies may substitute products with lower cost versions. To accommodate this, Statistics Denmark is in running contact with reporting companies. They are requested to review their reported suite of products every two years in order to assess if the products are still representative of general sales or import. Companies are then requested to replace outdated products.

Substitution of the products used in the index may lead to changes in product quality, which is addressed by continuous quality corrections. Newly added products are included in calculations after their prices are registered for two subsequent periods.

Reporting errors: Errors may occur during registration of prices by reporting companies, either because wrong prices are given for a products, or because prices are given for the wrong product. These errors are usually due to misunderstandings and are accommodated by running dialogue with reporting companies. Errors may also occur during registration of prices 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.

Index calculation: Calculation of the Construction cost indices for civil engineering projects is done by computer in a dedicated price index system. The likelihood of calculation errors is therefore inconsiderable.

The basis for the weights for earth work, asphalt work, concrete structures, and iron structures were developed in cooperation with the Danish Road Directorate, DSB and the Danish Construction Federation in 1976 based on an analysis of a ongoing and completed civil engineering project. The weights reflect the shares of labor costs, material costs and equipment costs of the total costs of performing civil-engineering projects. The statistics are considered representative of the actual cost development for typical civil engineering projects in Denmark. Thereby the indices may be less representative for projects that differ from typical civil engineering projects. Also, the representativeness of the indices may decline further from the weight reference period, or if there is a change in the involved goods. Therefore, the weight basis is reevaluated approximately every five years, lastly in 2015, where the original weight basis was found to still be adequate.

It is assumed that developments of the prices used in the Construction cost indices for civil engineering projects 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 civil engineering projects.

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 indices for civil engineering project is assessed to be of high quality and representative for the general trends in costs of civil engineering projects in Denmark. The index is based primarily on existing data that has already been checked for errors.

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.

The indices are calculated in a dedicated, computerized price index system to prevent manual errors. Errors may occur during the price reporting process, either from involved data sources or by Statistics Denmark, but this is accommodated by continuous monitoring of prices.

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 indices for civil engineering projects are assessed to be of high quality and representative of actual construction of civil engineering projects in Denmark. Weight bases are reevaluated approximately every five years to assess the actuality of the indices compared to actual civil engineering work.

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 figures are compiled. Revisions are not performed, unless errors are found in already published numbers.