Road traffic volumes

Name: Road traffic volumes.
Topic: Transport and tourism.

Division for Energy, Environmental and Transport Statistics.

Municipality.

Frequency: Annual.

Timeliness: The statistics are published within six months of the end of the statistical year.

The statistics are used for reporting road traffic volumes to Eurostat.

The microdata of the statistics are stored as text files.

The statistics describe the total and average distances covered by Norwegian passenger cars, goods road vehicles, buses and coaches by type of vehicle, type of fuel, age of vehicle and the home county and municipality of the owners.

The statistics on road traffic volumes were established in 2009, with figures for the statistical years 2005-2008. The statistics have since been disseminated annually. The development of the statistics was partly financed by the Norwegian Ministry of Transport and Communications.

A more detailed model for calculating road traffic volumes was developed in 2018 within the Ministry of Transport and Communications' framework for statistics for transport and the environment. The new model makes it possible to release statistics at a more detailed level for different fuel types within each vehicle type.

The statistics are primarily used in transport planning and transport models by public authorities and research institutes (the Ministry of Transport and Communications, the Public Roads Administration, the Institute of Transport Economics, Statistics Norway). Environmental authorities and organisations, transport organisations and transport companies are also main users. In Statistics Norway the figures are included in energy and emission analysis of the Norwegian transport sector. The statistics are also used for Eurostat’s collection of road traffic volumes.

No external users have access to the statistics and analyses before they are published for all users on ssb.no at 8 am. Prior to this, a minimum of three months' advance notice is given in the Statistics Release Calendar. This is one of Statistics Norway’s key principles for ensuring that all users are treated equally.

The Road Traffic Index of the Norwegian Public Roads Administration measures the annual growth in road traffic volumes based on data from counters located on Norwegian roads.

Statistics Norway also compiles annual statistics on the stock of registered vehicles and the number of cars that have been scrapped through the national refund programme.

Eurostat disseminates statistics on road traffic volumes for various countries in Europe.

The Statistics Act (in Norwegian)

The statistics cover passenger cars, goods road vehicles, buses and coaches which are registered in the Norwegian Register of Vehicles throughout or in parts of the statistical year. Vehicles with special licence plates and vehicles registered prior to 1 January 1960 are not included in the statistics, because these vehicles are not included in the regulation on periodical roadworthiness tests. The statistics are entirely based on data collected from registers administered by the Norwegian Public Roads Administration. The registered vehicle is the observation unit and the statistical unit.

The statistics are compiled by combining information from the Norwegian Register of Vehicles with data from the odometer readings collected by the Norwegian Public Roads Administration from the inspection bodies that conduct the periodical roadworthiness tests on the vehicles. The Norwegian Register of Vehicles contains data on ownership, registration status and technical characteristics of the vehicles. The odometer readings provide information on date of inspection and the odometer reading of the vehicle at the time of inspection.

The regulation on periodical roadworthiness tests requires passenger cars to be inspected for the first time in the fourth year after being registered as new, while utility vehicles for the most part are required to be inspected in the second year after registration. Passenger cars are then required to be inspected again every other year, while utility vehicles are required to be inspected every subsequent year.

Approximately 75 per cent of the vehicles in the population for each statistical year have odometer readings which can be used to calculate annual driving distances.

Odometer readings for newly registered vehicles will not be available when compiling the statistics for each statistical year. In addition to this, the odometer readings of some vehicles may have been be rejected as a result of being identified as incorrect in the logical or statistical controls.

Annual driving distances are estimated for the approximately 25 per cent of the vehicles which lack odometer readings by using the average mileage for similar vehicles with valid odometer readings. In contrast to what happens in most sample surveys, values are estimated only for a minor part of the population in the statistics on road traffic volumes.

A more detailed model for calculating road traffic volumes was developed in 2018. The purpose of the adjustment was to make it possible to release statistics at a more detailed level for different fuel types within each vehicle type.

In the new model, the odometer data is controlled mechanically by means of logical and statistical controls. Meter readings that are identified as errors by the controls are removed.

Average daily vehicle-kilometres are calculated for all vehicles with valid odometer readings based on the two last valid readings for the vehicle.

Both the vehicles with valid odometer readings and the vehicles without valid odometer readings are then divided into 145 strata following these characteristics:

• Age groups
• Vehicle groups
• Type of fuel
• Total permissible weight classes
• Home county and home municipality of the owner

Aggregated average daily vehicle-kilometres are calculated for each of the strata based on the vehicles that have valid odometer readings. The aggregated average daily vehicle-kilometres of each stratum are then imputed on the vehicles in a similar stratum which are missing valid odometer readings. Finally, the average yearly vehicle-kilometres of each vehicle is calculated based on the number of days the vehicle has been registered in the Norwegian Register of Vehicles during the statistical year.

To ensure that as much as possible of the changes in driving patterns caused by the Corona situation are reflected in the statistics, only odometer readings from cars checked in the second half of 2020 were used as a basis for calculating the vehicle-kilometres for 2020.

Figures are not disseminated for table cells with less than three observations.

The statistics on road traffic volumes were established in 2009, with figures for the statistical years 2005-2008. The statistics have been disseminated annually from 2009. Figures on the municipality level were disseminated for the first time for the statistical year 2010.

Starting from the statistical year 2010, the stratification of the population was changed to include the home county and home municipality of the owners of the vehicles. At the same time, the age of the vehicles were made the primary stratification criteria. The change in stratification increased the precision in the estimates of road traffic volumes for age groups, counties and municipalities on detailed levels. On the more aggregated levels, however, the modification did not cause considerable changes to the estimates of road traffic volumes in the statistics.

A more detailed model for calculating road traffic volumes was developed in 2018. The purpose of the adjustment was to make it possible to release statistics at a more detailed level for different fuel types within each vehicle type. On 21 March 2019, new figures for 2018 and adjusted figures for 2016 and 2017 were published based on the new model. Although the change in method generally has limited effect on the statistics, there are some deviations from previously published figures for 2016 and 2017.

When developing the detailed model in 2018, an error was corrected in the grouping of motor homes and camper vans. Due to an unfortunate error, a new code for camper vans registered after 1 January 2009 was not correctly updated in the original mileage model. A considerable number of new camper vans were thus not included in the data basis for the statistics in the years 2009-2015, resulting in the total mileage for camper vans being underestimated for this period.

When using the fuel type figures, it is important to be aware that "other fuels" included paraffin, gas and hydrogen up to and including 2015. It should also be noted that petrol and diesel hybrids were generally registered as regular petrol and diesel cars in the motor vehicle register up to and including 2015. The transition to the use of specific fuel codes for petrol and diesel hybrids was gradual, with a major re-encoding implemented at the beginning of 2016.

In March 2020, the 40-year-old Motor Vehicle Register of the Norwegian Public Roads Administration was replaced by a new Vehicle Register based on new technology and partly new content. Changes in the use of variables and code lists in the new register have resulted in some discrepancies between 2019 and 2020 at the most detailed levels of the statistics.

Prior to 2020, vehicles on Svalbard not registered for use on public roads were not included in the statistics.

Measurement errors are errors that that occur during data collection and are caused by the survey instrument, in this case errors in the registered information on the vehicles. A typical error may be use of the wrong scale, for instance Norwegian miles (10 kilometres) instead of kilometres. Manual adjustments of odometers may also be a source of measurement errors. Processing errors are errors introduced during data processing in Statistics Norway.

In the statistics on road traffic volumes, there will also be measurement errors due to the fact that the last available odometer reading is used in order to estimate the average daily driving distances for the vehicles with odometer readings. As a result of the regulation on periodical roadworthiness tests, some of the odometer readings for passenger cars and vans may be up to two years old when road traffic volumes for the statistical year are estimated. Changes in driving patterns for passenger cars and vans caused by economic or social trends may thus take some time to register in the statistics. For most other utility vehicles, odometer readings are collected annually.

The scope of this measurement error is hard to quantify. Analyses show that the average driving distances decrease gradually with the age of the vehicle. However, the differences are not too large from one year to the next. Within a shorter period of time, like a few years, the average driving distances of a vehicle will mainly be determined by the area of use. Thus, the calculation of average daily driving distances based on the last available odometer readings for the vehicles should provide a fairly accurate and robust method of estimating road traffic volumes.

The scope of the remaining measurement errors and processing errors in the survey is also unknown, but is not believed to introduce systematic errors in the results.

Approximately 2 per cent of the vehicles in the population have odometer reading that are deleted, either due to uncorrected logical errors in one or more of the readings or because the last valid reading is too old to comply with the control intervals prescribed by the regulation on roadworthiness tests for the vehicle. Unit non-response such as these contributes to the skewness and an uncertainty added to the uncertainty due to sampling errors.

Sampling errors arise from the fact that the estimates are based on a sample and not a census of the entire population. The sampling error is measured by the standard error which estimates the expected deviation between the survey estimate and the estimate that would have been obtained if a complete enumeration had been carried out.

Approximately 25 per cent of the vehicles in the population for each statistical year are missing odometer readings which can be used to calculate vehicle-kilometres. The majority of these vehicles are newly registered vehicles which have not yet been required to undergo roadworthiness tests. In contrast to what happens in most other sample surveys, values are estimated only for a minor part of the population in the statistics on road traffic volumes.

In the statistics on road traffic volumes there may be coverage errors due to time gaps in registration. For vehicles that are deregistered or scrapped during the statistical year, the information in the Norwegian Register of Vehicles may not be accurate at the time when the statistics are being produced. Coverage errors may also occur when vehicles are registered as normal, but no longer in use. Changes in the type of use of vehicles between two readings may also be a source of coverage errors, for instance when used taxis are sold as private cars. The scope of such errors in the data is not known.