Content
About the statistics
Definitions
-
Name and topic
-
Name: Energy consumption in service industries (discontinued)
Topic: Energy and manufacturing
-
Responsible division
-
Division for Energy and Environmental Statistics
-
Definitions of the main concepts and variables
-
Industrial buildings: Industrial buildings include private and public buildings which are used for business activity. Public buildings include schools, kindergartens, hospitals, nursing homes and more, with the only exception being private housing. For buildings that are used both for private housing and economic activity, inclusion in the statistics is defined by amount of area that is used for economic activity. If half of the building area is used for economic activity, then the building is defined as an industrial building.
Building type: Describes area of application, which could be a hospital, kindergarten or an office building. In the GAB/Matrikkelen, the buildings within service industries are separated into 100 different building types. However, to avoid over-complex statistics, a choice is made to group the building types into fewer categories. Building types that have similar areas of application or energy use are aggregated into the same groups. The main area of application of the building is what determines which building type a building is categorised in.
Industry: An industry describes the main activity in the building. Often the main activities in a building coincide with building type; a hospital and commodity trade for example. In some buildings it can still be difficult to define main activity based on building type, or the case could be that there are many different activities in the same buildings. Questions on the amount of the area used for different activities are therefore included in the questionnaire.
In the temporary statistics, the main results on building type are published, while in the final publication, the industry will be included.
Energy use: Includes all energy types used in buildings. This includes all energy used for room heating, hot water, cooling, lighting, electrical devices or other purposes. The most common energy sources in industrial buildings are electricity, district heating, heating oil, paraffin, gas and pellets, but other energy types such as wood and local heating from neighbouring buildings should also be included if this is part of the energy use. It is the actual added energy into the building that is reported. This corresponds to the energy that building owners have to pay for. Energy economising, such as installed heating pumps and control systems leads to a reduced need to buy energy, thus capturing energy efficiency.
Area: Energy use are calculated per m 2 heated used area (BRA=bruksareal). This excludes outside walls and basements and car parking not heated. Area that use energy to bee cooled down is included.
Energy use adjusted for temperature: The adjustment for temperature is made using number of degree days. The number of degree days is a measure of the heating requirement. It is assumed that heating is not required when the medium temperature during the day and night is higher than 17 °C. The figure for degree days for one day and night is defined as the number of degrees that the medium temperature is below 17 °C. The energy use is adjusted for temperature by adjusting the part of the total energy use that is dependent on temperature. This is done for each building based on the degree days for the particular year for the municipality of the building and the corresponding normal degree days (1961–1990) for that municipality. For more information, click on the tab “About the statistics”.
-
Standard classifications
-
Building types in these statistics are classified according to the Norwegian standard, NS 3457, for building types (Martikkelen).
Administrative information
-
Regional level
-
National level.
-
Frequency and timeliness
-
The first survey of statistics on energy use in service industries was for the reference year 2008 . The survey is planned to be published every third year. Temporary data are published 10-12 months after the reference year. Final data are published approximately 1.5 years after the reference year.
-
International reporting
-
Not relevant
-
Microdata
-
Micro data are stored in Oracle database.
Background
-
Background and purpose
-
The primary purpose of the statistics is to provide an overview of energy use, heating equipment and energy efficiency in industrial buildings within service industries. StatisticsNorwayusually collects information on energy use in service industries based on data received from producers and distributors of electricity, district heating, heating oil and bioenergy, as well as economic key statistics. The establishment of separate statistics is based on the aspiration of a more detailed and secure energy data for service industries. The survey is based on a co-operative project between among others Statistics Norway and Norwegian Water Resources and Energy Directorate.
-
Users and applications
-
The statistics will be used by private and public institutions that work with energy use in industrial buildings. The statistics will be an important part of the data material that is used as a basis for the calculation of energy accounts, energy balances, and emissions to air, as well as primary material for analysis and research.
-
Coherence with other statistics
-
Statistics on energy use in service industries is also published in the Energy account and energy balance.
-
Legal authority
-
The Statistics Act of 16 June 1989 §§2-1, 2-2 and 2-3
-
EEA reference
-
Not relevant.
Production
-
Population
-
The population covers over 100 000 industrial buildings within service industries. Service industries include both private and public services, limited to industry 45-96 from the standard industrial classification 2007 (SIC2007). The population and sample is drawn from building type 300-800 in central property register (Matrikkelen). Exclusions from the population include particular building types that are assumed to have low energy use or with lacking/incomplete records. For the 2011 survey building type 400 was excluded because of special building types in this group, so we had roughly 64 000 buildings in the population.
-
Data sources and sampling
-
Information on the buildings in the sample is collected from the central register for properties inNorway(Matrikkelen). StatisticsNorwayhas their own copy of this register for statistical purposes - SSB-Matrikkelen.
From the population, a sample consisting of approximately 6 100 buildings was drawn for the 2011 survey. The main goal for the sample is to be as representative as possible and cover a large part of service industries and building types within these industries. The buildings in the sample are supposed to have a geographic spread in order to cover the effect of local climate on energy use. Because of the importance of area, there is a need for the sample to include as many large buildings as possible with over 5 000 m 2 . There is no complete overview of areas for all buildings in Matrikkelen and therefore there is uncertainty concerning how large a share of the total area in service industries is covered by these statistics.
To avoid double reporting and also to strengthen the data foundation, building information from Enova’s building network is incorporated into the data basis.
-
Collection of data, editing and estimations
-
The survey is based on data collected by questionnaire. The letters for the 2011 survey was distributed in August 2012 and the questionnaires submitted electronically via Idun. It was also possible to call Statistics Norway to receive the paper questionnaires to fill out, but most of the respondents used Idun. The questionnaires are sent to landed property because of a lack of information on building owners. Because the Statistics Act is applied in these statistics, the owner of the landed property has a filing requirement. The deadline for returning the questionnaire was in September. There were two reminders, with a final deadline in November.
In the electronic questionnaire there are several logical controls and controls of high and low energy use per m 2 area. Buildings with unusually large or low energy intensity were closely examined. This includes buildings outside the range 50-600 kWh/m 2 . Within most building types, energy use above 600 kWh/m 2 or below 50 kWh/m 2 is unusual if the building is in normal use. In these cases, therefore, either energy use or area is probably incorrectly reported. But because some exceptions exist, a closer examination on factors that lead to either high or low energy use is done before these numbers are corrected.
A comparison with other sources that have statistics on energy use in buildings is also done. Some of the most comprehensive sources are Enova’s building network and the energy report by Statsbygg. This way, we are able to control the reasonableness of the data collected by Statistics Norway.
All energy commodities are measured in kWh based on their theoretical energy contents. For conversion factors see “About the statistics” in the Energy account and energy balance .
The revision is done in ISEE (Dynarev) revision data base. This data base was made for the 2011 survey and has contributed to much more effective and reliable revision.
-
Confidentiality
-
All the questionnaires are subject to confidentiality considerations by the Statistics Act, so that no data should be published if they can be traced back to the respondent.
-
Comparability over time and space
-
The statistics has been published for the reference year 2008 and 2011 and are planned to be published every third year.
Accuracy and reliability
-
Sources of error and uncertainty
-
In statistical surveys, there will always be some measurement errors. Measurement errors are caused by the questionnaire or the respondent’s internal system for obtaining the data. Since this survey is relatively comprehensive, there is a probability for measurement errors in the data. One specific cause could be that the respondents report only part of their energy use, so that energy use per m 2 becomes too low, or that the respondents lack accurate information on total heated area.
Processing errors can occur when Statistics Norway processes the data. Typical examples are misinterpretations of answers or correct answers that for some reason are assumed to be false and corrected. Paper questionnaires are optically read with automatic verification and transmission to an electronic medium. The current techniques for optical reading are of a high quality, and few errors are found in this phase of the production. The introduction of Idun has also helped to reduce such errors, as data from electronic questionnaires are loaded directly into the system.
The consequences of measurement and processing errors in the data are difficult to map. Removal of all obvious mistakes implies that this will not affect the published data to a large degree. Buildings with extremely high or low energy use are excluded from the published data, unless there is a justified reason for the abnormal energy use.
A total of 6 100 questionnaires were distributed, and at the end of the final deadline about 5 200 were either received or notification had been received that they were not able to respond. Many of the questionnaires were of poor quality so that we were left with around 2 800 questionnaires. This gives an answering share of 46 per cent. Some respondents reported information for several buildings, while others reported via mail so that the response rate is higher than 2 800.
A lack of information about areas in buildings implies that we are not able to say exactly what the response rate for the total area is.
The most important reason for non- response was that the owner did not use the building and therefore did not pay for energy. The questionnaire was then also sent to the tenant, but the response rate here was generally low. Other reasons for not filling in the complete questionnaire were that the building was reported to be smaller than 200 m 2 or not heated in more than 90 days during the year.
Coverage error refers to uncertainties that occur in sample surveys as opposed to full counts. There is a clear connection between area and energy use for all building types, where the variation increases with area. Large buildings are over represented in the survey.
Under data collection process and controls, we detected that there are some buildings that were incorrectly classified. This can lead to skewed measurement of energy use within some building types. We have corrected these mistakes as they have been detected.