Stats NZ has a new website.

For new releases go to

www.stats.govt.nz

As we transition to our new site, you'll still find some Stats NZ information here on this archive site.

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
Energy Use Survey: Services sector 2010
Embargoed until 10:45am  –  13 October 2011
Data quality

The data quality section provides period-specific and general information about the data.

Period-specific information
This section has information about data that has changed since the last release.

General information
This section contains information about data that does not change between releases.

Period-specific information

Reference period

The survey was posted out in April 2011, with a reference period of the last financial year for which the organisation had results available in May 2011. The majority of respondents had a 31 March 2011 balance date. However a significant number of respondents had either a June 2010 or a December 2010 balance date. A small number stated other dates. No adjustment has been made to produce figures for a single consistent time period.

Effect of Canterbury earthquakes

The Energy Use Survey for 2010 was the first collection since the September 2010 and February 2011 Canterbury earthquakes. A small number of units in the sample were affected by at least one of the earthquakes. These organisations were excluded from the survey under 'compassionate exclusion'. Checks were done to see if certain industries were affected more than the others, however no such impact was found.

Response rate

The survey was sent to just over 5,000 units and the required overall response rate was 80 percent. The response rate achieved was 83 percent. The response rate achieved for key units was 100 percent, which met the target.

Sampling error for 2010 survey

The sample design for the Energy Use Survey: Services sector 2010 aimed to control and reduce the relative sampling errors (RSEs), specifically for key energy types. These energy types were:

  • total energy use in each industry
  • electricity use in each industry
  • combined petrol and diesel use in each industry
  • natural gas for the accommodation and food services, and health care and social assistance industries.

The RSE estimates are larger than the design RSE targets for some energy types. This is due to higher variability in the responses than was expected in the sample design process.

How to read the sampling errors
Sampling errors for this survey were calculated using the RSE measure. RSEs are the sampling error as a percentage of the estimate. The sampling errors are estimates at the 95 percent confidence level. For example, the estimated energy use by the total services sector for 2010 is 64,000TJ. This estimate is subject to an RSE estimate of approximately 5.2 percent. This means that 95 percent of the possible samples of the same size will produce an estimate between 64,000 - 3,328 and 64,000 + 3,328; that is, between 60,672 and 67,328 TJ.

The following table shows the actual RSEs for each industry in the 2010 survey.

Relative sampling errors (RSEs) by industry
Industry Final RSE for total energy use (%)
Accommodation and food services 15.8
Information media and telecommunications 8.1
Financial and insurance services 16.9
Rental, hiring and real estate services 16.8
Professional, scientific and technical services 14.3
Administrative and support services 20.1
Public administration and safety 22.6
Education and training 7.4
Health care and social assistance 10.9
Arts and recreation services 19.9
Other services 6.6
Total for service sector 5.2

Target population

The 2010 survey covered the following services industries of New Zealand's economy:

  • accommodation and food services (H)
  • information media and telecommunications (J)
  • financial and insurance services (K)
  • rental, hiring, and real estate services (L)
  • professional, scientific, and technical services (M)
  • administrative and support services (N)
  • public administration and safety (O)
  • education and training (P)
  • health care and social assistance (Q)
  • arts and recreation services (R)
  • other services (S).

The target population included all economically significant units in the Australian and New Zealand Standard Industrial Classification 2006 (ANZSIC06) categories H and J to S that were live at the time of selection.

Survey population

The collection unit for the survey is the kind-of-activity unit (KAU). Altogether, 263,000 enterprises had KAUs that meet the requirements above for the 2010 collection.

The survey sample was designed to produce results for the industries listed under target population.

General information

Data source

The New Zealand Energy Use Statistics Programme is a product of the Energy Domain Plan that was published in 2006. The Energy Domain Plan was produced by Statistics NZ in collaboration with the Energy Efficiency and Conservation Authority (EECA), and the Ministry of Economic Development (MED). The domain plan identified energy use statistics as a key gap in energy information and prioritised a suite of energy use surveys.

The Energy Use Survey delivers information to help fill gaps in current energy statistics and provides a benchmark of energy use information for New Zealand’s economy, excluding households. Data from the survey will also feed into modelling systems that give current energy-use estimations and future demand forecasts (eg MED’s Energy Data File). Modelling assumptions can then be updated, which will improve the accuracy of modelled information.

Target population for three years of surveys

In the first stage of this three-year cycle (2008), the target population was the primary sector (ANZSIC06 A and B). http://www.stats.govt.nz/browse_for_stats/industry_sectors/Energy/EnergyUseSurvey_HOTP2008.aspx
In the second stage (2009), collection covered the industrial and trade sectors (ANZSIC06 C to G and I). http://www.stats.govt.nz/browse_for_stats/industry_sectors/Energy/EnergyUseSurvey_HOTP2009revised.aspx

Survey population

The KAU was chosen as the selection unit to allow the uses of energy to be associated with an activity as closely as possible. KAU data allows energy use to be separated for larger units with multiple branches involved in different activities (multi-KAU).

Sample design

The sample design was a two-way, one-stage stratified random sample. The stratification and design variables were ANZSIC06, GST sales, and rolling mean employment.

The 2010 survey collected information from New Zealand's services sector on the following commodities (the same was done for the primary, and industrial and trade sectors in previous surveys):

  • electricity – all electricity purchased from the national grid and energy sources used for input into electricity generation and cogeneration
  • electricity generated in the unit – electricity generated within the operations of the unit (this figure is not included in total energy used, to avoid double counting)
  • petroleum products – energy products derived from the refining process of crude oil including:
    • petrol – an aggregated figure of 96 and 91 octane petrol
    • fuel oil – an aggregated figure of the major intermediate products, notably light fuel oil and heavy fuel oil
    • diesel
    • liquefied petroleum gas (LPG)
    • aviation fuel
  • natural gas
  • coal – including all ranks
  • wood and wood waste – used for energy purposes.

The survey also collected information on energy management practices.

Population vs sample estimates

The tables and numbers providing estimates of energy use, in this release, were weighted to account for the whole industry or sector. The sample was designed to be able to provide estimates of the whole sector's energy use, based on the sampled organisations. The tables and percentages providing estimates of energy management initiatives and perceptions were not weighted to account for the whole sector. These figures are calculated only on organisations that actually responded to the survey. While the responded organisations are representative of the whole sector, the data was not designed specifically for the analysis, so weighting was not applied before calculating percentages. When comparing these tables, readers need to remember that the energy-use numbers are estimates of use by the whole sector, while the energy-management numbers are specific responses from representative organisations.

Measurement errors

The survey results are subject to measurement errors, including both non-sample and sample errors. These errors should be considered when analysing results.

Sampling error

Sampling error is a measure of the variability that occurs by chance because a sample, rather than the entire population, is surveyed. The level of sampling error for any given estimate depends on the number of sampled individuals, the variability of the estimate, and the sample size. Due to the random nature of the sample selection the error will differ for different samples.

Non-sampling errors

Given the nature of the data collected, there are limitations on the level of accuracy that can be expected from the survey. Records may not be kept in the form required for the survey, some estimation by the respondent may be required and non-sampling error may occur. Non-sampling errors include mistakes by respondents when completing questionnaires, variation in the respondents’ interpretation of the questions asked, and errors made during the processing of the data. Statistics NZ has extensive procedures to minimise these types of error, but they may still occur and are not quantifiable.

Non-response and imputation

Unit non-response

Unit (or complete) non-response occurs when units in the sample do not return the questionnaire. The initial selection weight of the remaining units in the stratum was adjusted to account for the unit non-response (no item non-response imputation would occur for the units that did not return the questionnaire).

Item non-response

Item (or partial) non-response is when units return the questionnaire but some questions are not answered. Item non-response imputation was carried out for units that answered some but not all of the questions they were required to (based on questionnaire routing rules). Respondents who did not answer any of the questionnaire were classified as unit non-responses and the weights were adjusted accordingly. Item non-response was imputed for and the methods used are as follows:

Imputation of numeric variables

The imputation method for numeric variables was random donor imputation. In this method, the responses of a randomly selected donor from within the same imputation cell as the non-respondent are imputed in the recipient unit. Donor imputation was used so that the distribution was maintained.

Imputation of categoric variables

The imputation method for categoric variables was random donor imputation. The donor supplied responses for all categoric variables requiring imputation. If the donor unit did not respond to any of the variables requiring a response, then the next best donor was selected to supply this information. This was continued until all of the variables had a response.

Energy units standardised

Information on energy usage was collected in the unit that applies to each commodity; for example, litres for petrol and kilowatt hours (kWh) for electricity. These units were converted to a standard unit (joules) for reporting. This conversion enables the energy contained in different forms to be compared directly. The conversions were carried out by applying a calorific value (enthalpy value) to each energy type and form. The calorific values were sourced or derived from MED’s Energy Data File 2010. See the table below for the calorific values used for each energy type.

Energy types and their calorific values
Energy type Details Calorific value
Electricity Electricity's standard universal unit, the watt, is defined as one joule per second 3.6 MJ per kWh
Petrol Two main forms of petrol are in the market, regular and premium, and each has a slightly different conversion factor. The factor used in the Energy Use Survey is a weighted average of the two values, according to their current prevalence in the market 35.1 MJ per litre
Fuel oil There are two types of fuel oil: light fuel oil and heavy fuel oil. The conversion factor used in the Energy Use Survey was derived using a weighted average of the two, according to their current prevalence in the market 40.7 MJ per litre
Diesel The value used is that of regular diesel 37.8 MJ per litre
LPG

Liquid petroleum gas. LPG figures were provided in both litres and kilograms

26.4 MJ per litre 49.5 MJ per kg
Aviation fuel There are two major forms: jet fuel and aviation gasoline 37.3 MJ per litre 33.9 MJ per litre
Natural gas Most natural gas figures were provided in joules, although some were in kilowatt hours 3.6 MJ per kWh
Coal Bituminous 29,240 MJ per tonne
Sub-bituminous 21,720 MJ per tonne
Lignite 15,300 MJ per tonne

Where the type was not known, the conversion factor was an average of the lignite, sub-bituminous and bituminous coal values

22,087 MJ per tonne
Wood and wood waste Hog fuel or bark 7,000 MJ per tonne
Sawmill residues or fuel wood 10,300 MJ per tonne
Black liquor 8,600 MJ per tonne
Joinery, building, or furniture residues 16,300 MJ per tonne
Oven-dried wood 19,200 MJ per tonne
The conversion factor where the wood type was not known was an average of the other types. 12,280 MJ per tonne

More information

More information about the Energy Use Survey is available on our website. http://www.stats.govt.nz/surveys_and_methods/completing-a-survey/faqs-about-our-surveys/nz-energy-use.aspx

Liability

While care has been used in processing, analysing, and extracting information, Statistics NZ gives no warranty that the information supplied is free from error. Statistics NZ shall not be liable for any loss suffered through the use, directly or indirectly, of any information, product, or service.

Timing of published data

Timed statistical releases are delivered using postal and electronic services provided by third parties. Delivery of these releases may be delayed by circumstances outside the control of Statistics NZ. Statistics NZ accepts no responsibility for any such delays.

Crown copyright©

Creative Commons logo.

This work is licensed under the Creative Commons Attribution 3.0 New Zealand licence. You are free to copy, distribute, and adapt the work, as long as you attribute the work to Statistics NZ and abide by the other licence terms. Please note you may not use any departmental or governmental emblem, logo, or coat of arms in any way that infringes any provision of the Flags, Emblems, and Names Protection Act 1981. Use the wording 'Statistics New Zealand' in your attribution, not the Statistics NZ logo.

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
Top
  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+