Retail Trade Survey: September 2011 quarter

Data quality

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

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

Period-specific information

Measurement errors

All statistical estimates are subject to measurement errors. These include both sample errors and non-sample errors. In addition, the survey applies imputation methodologies to cope with small firms and non-response. These measurement errors should be considered when analysing the results from the survey. For more information on measurement errors, please refer to 'General information' in the Data quality section.

Sample errors

The postal survey was designed to give statistics at the following levels of accuracy (at the 95 percent confidence interval limit):

  • 3 percent for sales at the total national retail trade level
  • 10 percent for sales at the published national retail industry level.

This means, for example, that there is a 95 percent chance that the true value of total retail trade sales lies within 3 percent of the published estimate.

Retail Trade Survey: September 2011 quarter sample errors by industry
At the 95 percent confidence interval limit
Retail industry Level
(relative percent)
(absolute percent)
Motor vehicle and parts 7.7 4.2
Fuel 2.4 1.4
Supermarket and grocery stores 5.8 4.2
Specialised food 6.1 4.7
Liquor 6.7 1.5
Furniture, floor coverings, houseware, textiles 6.9 3.0
Electrical and electronic goods 2.1 1.5
Hardware, building, and garden supplies 4.3 2.0
Recreational goods 17.8 6.3
Clothing, footwear, and accessories 14.5 3.3
Department stores 0.0 0.0
Pharmaceutical and other store-based retailing 7.8 2.7
Non-store and commission-based retailing 10.6 7.2
Accommodation 4.0 3.1
Food and beverage services 3.9 2.2
Total retail trade 2.1 2.4

Industries with zero sample error are full-coverage industries. In these industries, all large firms are surveyed and all small to medium-sized firms are modelled using administrative data sourced from Inland Revenue.

Retail Trade Survey: September 2011 quarter sample errors by region
At the 95 percent confidence interval limit
Region Level
(relative percent)
(absolute percent)
Auckland 6.4 0.9
Waikato 18.2 3.7
Wellington 12.7 2.4
Remainder of the North Island 10.9 6.1
Canterbury 13.8 1.4
Remainder of the South Island 12.2 4.8


Small firms

Small to medium-sized firms are generally not surveyed. Their variables are instead modelled from administrative data (GST) sourced from Inland Revenue. Ratios calculated from the postal sample units are applied to the administrative data to provide an estimate of their variables.

Non-response imputation

Although every attempt is made to achieve a 100 percent response rate, in practice this does not occur. Values for non-responding businesses are estimated by a number of methods, including:

  • regression imputation
  • historic imputation
  • mean imputation.

Regression imputation involves estimating sales from the unit's administrative data (GST sales) based on the relationship shown by similar businesses. Historic imputation involves multiplying their response in the previous period by a non-response factor. The non-response factor is the average movement of similar businesses over the month. Mean imputation involves estimating a value for a unit by using the average value for a set of similar businesses.

Sales imputed in the September 2011 quarter
Retail industry Tax modelled Non-response
Percentage of sales
Motor vehicle and parts 8.1 13.6
Fuel 1.8 12.3
Supermarket and grocery stores 4.8 10.7
Specialised food 12.0 16.8
Liquor 12.0 12.9
Furniture, floor coverings, houseware, textiles 12.2 11.1
Electrical and electronic goods 10.1 11.2
Hardware, building, and garden supplies 11.3 11.0
Recreational goods 9.1 18.1
Clothing, footwear, and accessories 8.0 14.1
Department stores 0.0 0.7
Pharmaceutical and other store-based retailing 10.2 16.9
Non-store and commission-based retailing 11.3 12.9
Accommodation 9.5 17.2
Food and beverage services 11.0 20.0
Total retail trade 7.2 12.9

Postal response rate

The response rate describes the proportion of geographic units (GEOs) that provided survey responses. Note that the calculation of this response rate relates only to data for the postal sample. The Retail Trade Survey has a target response rate of 85 percent. The response rate achieved for the September 2011 quarter was 86.1 percent.

General information


The target population for this survey is all GEOs operating in New Zealand that are classified on Statistics New Zealand's Business Frame to the Australian and New Zealand Standard Industrial Classification 2006 (ANZSIC06) below:

  • retail trade (ANZSIC division G)
  • accommodation and food services (ANZSIC division H).

Industry descriptions

A GEO is included in an industry based on its predominant activity in terms of sales. For example, a petrol station will sell petrol and diesel, but it may also sell car parts and grocery items. The store will be classified to the fuel retailing industry if most of its sales income comes from the sale of fuel. Data are published for 15 industries, which are defined as follows:

ANZSIC06 industries, class codes, and descriptions for the Retail Trade Survey (RTS)
RTS industry and description used in published tables ANZSIC06 class and description
G1110 Motor vehicle and parts G391100 Car retailing
G391200 Motor cycle retailing
G391300 Trailer and other motor vehicle retailing
G392100 Motor vehicle parts retailing
G392200 Tyre retailing
G1120 Fuel G400000 Fuel retailing
G1210 Supermarket and grocery stores G411000 Supermarkets and grocery stores
G1221 Specialised food G412100 Fresh meat, fish, and poultry retailing
G412200 Fruit and vegetable retailing
G412900 Other specialised food retailing
G1222 Liquor G412300 Liquor retailing
G1311 Furniture, floor coverings, houseware, textiles G421100 Furniture retailing
G421200 Floor coverings retailing
G421300 Houseware retailing
G421400 Manchester and other textile goods retailing
G1312 Electrical and electronic goods G422100 Electrical, electronic, and gas appliance retailing
G422200 Computer and computer peripheral retailing
G422900 Other electrical and electronic goods retailing
G1313 Hardware, building, and garden supplies G423100 Hardware and building supplies retailing
G423200 Garden supplies retailing
G1321 Recreational goods G424100 Sport and camping equipment retailing
G424200 Entertainment media retailing
G424300 Toy and game retailing
G424400 Newspaper and book retailing
G424500 Marine equipment retailing
G1322 Clothing, footwear, and accessories G425100 Clothing retailing
G425200 Footwear retailing
G425300 Watch and jewellery retailing
G425900 Other personal accessory retailing
G1330 Department stores G426000 Department stores
G1340 Pharmaceutical and other store-based retailing G427100 Pharmaceutical, cosmetic, and toiletry retailing
G427200 Stationery goods retailing
G427300 Antique and used goods retailing
G427400 Flower retailing
G427900 Other store-based retailing nec
G1350 Non-store and commission-based retailing G431000 Non-store retailing
G432000 Retail commission-based buying/selling
H2110 Accommodation H440000 Accommodation
H2120 Food and beverage services H451100 Cafes and restaurants
H451200 Takeaway food services
H451300 Catering services
H452000 Pubs, taverns, and bars
H453000 Clubs (hospitality)
 Note: nec = not elsewhere classified 

Sample design

The survey population is stratified according to:

  • industries defined by the ANZSIC-based ANZIND classification at the inter-industry level
  • size (in terms of rolling-mean employment)
  • turnover (annualised GST sales).

Each ANZIND inter-industry contains between two and four substrata. Because of the contribution that large units make to the economic activity within each industry, they are all included in the sample. A portion of the remaining medium to large units is also included in the sample. In addition, small to medium-sized businesses have their data modelled from administrative data (GST) sourced from Inland Revenue. The Inland Revenue data have been forecast two months ahead. All retailing GEOs belonging to a selected 'enterprise' are included.

The sample is based on approximately 52,000 retail outlets in New Zealand. Around 2,500 enterprises (between 8,000 and 8,500 GEOs) have been selected in the Retail Trade Survey (RTS) postal sample. The postal sample is supplemented by GST data representing smaller retailers, approximately 26,400 enterprises (26,500 GEOs).

Sample maintenance

Sample maintenance is the process that maintains the sample over time, to reflect 'births', 'deaths' and other structural changes identified on the Business Frame. The information for Business Frame changes can be from a variety of sources, including GST registrations and respondent contact.

New enterprises are identified when they register for GST. Once a quarter, the new enterprises are selected into the sample using the same criteria as for the original sample. These are referred to as births. When an enterprise ceases trading, its retailing GEOs are removed from the survey. These are referred to as deaths.

Enterprises can also enter or leave the survey sample if they are reclassified to a different industry. Reclassifications occur when an enterprise changes its main form of activity (eg from wholesale trade to retailing). These are usually identified in the Annual Frame Update Survey conducted in February of each year.

Sample reselection

The sample for the RTS is reselected each quarter to ensure the sample reflects changes occurring in the retailing population.

Measurement errors

Errors in the survey are divided into two classes:

Non-sampling error

Non-sampling error includes errors arising from biases in the patterns of response and non-response, inaccuracies in reporting by respondents, and errors in recording and coding data. The size of these errors is difficult to quantify. Statistics may be revised if significant errors are detected in subsequent quarters.

Sampling error

Sampling error is a measure of the variability that occurs by chance because a sample, rather than an entire population, is surveyed.

Use of retail trade data in quarterly national accounts

A key use of the RTS is in calculating retail trade value added for compiling quarterly gross domestic product (GDP).

The quarterly GDP retail trade indicator uses retail sales volumes expressed in September 1995 quarter prices, by industry, series from the RTS. These series are chain-linked to give constant price sales at the ANZSIC96 working-industry level. The chain-linking weights are calculated using annualised quarterly current price sales by RTS industry.

Seasonally adjusted series

The X-12-ARIMA package has been used to produce the seasonally adjusted estimates and trend estimates for sales in all subdivisions. Seasonal adjustment aims to eliminate the impact of regular seasonal events (such as annual cycles in agricultural production, winter, or annual holidays) on time series. This makes the data for adjacent quarters more comparable.

All seasonally adjusted figures are subject to revision each quarter. This enables the seasonal component to be better estimated and removed from the series.

Estimated trend

For any series, the survey estimates can be broken down into three components: trend, seasonal, and irregular. While seasonally adjusted series have had the seasonal component removed, trend series have had both the seasonal and the irregular components removed. Trend estimates reveal the underlying direction of movement in a series, and are likely to indicate turning points more accurately than are seasonally adjusted estimates.

The trend series are calculated using the X-12-ARIMA seasonal adjustment package. They are based on a five-term or seven-term moving average of the quarterly seasonally adjusted series, with an adjustment for outlying values.

Trend estimates towards the end of the series incorporate new data as they become available and can therefore change as more observations are added to the series. Revisions can be particularly large if an observation is treated as an outlier in one quarter, but is found to be part of the underlying trend as further observations are added to the series. Typically, only the estimates for the most recent quarter will be subject to substantial revisions.

Changes to the Retail Trade Survey deflators

The RTS deflators that appear in tables 13 and 14 measure change in the prices of goods and services sold by businesses in the 15 retail industries. Movements in actual retail sales values can be explained by changes in price, and by changes in volume. The deflators are used to remove the effect of price change, which allows change in the volume of retail sales to be estimated.

The deflator for each industry consists of a 'basket' of indexes, drawn mainly from the consumers price index (CPI). The CPI indexes and other indicators in each deflator's basket represent the goods and services sold by the industry. Each good or service is weighted to reflect the relative importance of the mix of goods and services sold by the industry.

In 2010, the RTS was redesigned to reflect the updated industrial classification, the Australian and New Zealand Standard Industrial Classification 2006 (ANZSIC06). The RTS deflators were developed to deflate estimates of retail sales in current prices from the redesigned RTS.

The weights of the new deflators are based mainly on information used in the 2011 Consumers Price Index Review, including household spending patterns, by industry, that were reported in the 2009/10 Household Economic Survey.

The ANZSIC06 deflator weights were calculated for the 41 lower-level industries that feed into the 15 published retail industries. These weights are expressed in the prices of the June 2011 quarter. The 41 industries are aggregated to the 15 published retail industries using information on their relative importance, based on RTS results in the year to the June 2011 quarter. The relative importance of the 41 industries will be updated annually, based on sales for the latest year to June.

The ANZSIC06 deflators were directly calculated to measure price change from the June 2010 quarter onwards. Before the June 2010 quarter, the deflators were derived from backcast estimates of retail sales in current and constant prices.

Regional estimates

In the October 2003 month, the RTS sample of GEOs changed. ANZSIC06-based regional data is not available prior to the December 2003 quarter.


While all care and diligence has been used in processing, analysing, and extracting data and information in this publication, Statistics NZ gives no warranty it is error-free and will not be liable for any loss or damage suffered by the use directly, or indirectly, of the information in this publication.


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.

Page updated: 18 November 2011