Tourist Accommodation, Australia methodology

Latest release
Reference period
2015-16 financial year
Next release Unknown
First release

Explanatory notes


1 This publication presents data from the Survey of Tourist Accommodation (STA). The STA is a census of all in-scope accommodation establishments within Australia. This release includes the four quarters of the 2015-16 financial year, that is September quarter 2015, December quarter 2015, March quarter 2016 and June quarter 2016.


2 Establishments within the scope of the survey provide predominantly short-term non-residential accommodation, i.e. accommodation which is not leased, and which is provided to guests who would generally stay for periods of less than two months. Some of these establishments also provide long-term residential accommodation. The amount of such activity is considered to be insignificant and is included in the data presented in this publication.

3 Establishments in scope of the STA are:

  • hotels and resorts with 15 or more rooms
  • motels, private hotels and guest houses with 15 or more rooms
  • serviced apartments with 15 or more units.


4 For the 2015-16 collection period, the annual frame update process was undertaken using a file provided by STR Global, a company that tracks supply and demand data for the hotel industry. The update process was routine for 2015-16.

5 The 2014-15 frame update process resulted in the identification and subsequent addition of 279 new establishments to the STA beginning with the September quarter 2014. The addition of the 279 tourism establishments resulted in a break in time series between the June and September quarters 2014. The impact of the break in time series is explained in the 2014-15 STA Technical note.

6 During processing of 2015-16 STA data, it became clear that 40 accommodation establishments added during the 2014-15 frame maintenance exercise were duplicates of establishments already on the frame. As a consequence the 2014-15 STA data was overstated. In comparing estimates in original terms, it is recommended that users read the Technical note included in this release for interpreting the movements in the data between 2014-15 and 2015-16.

Accommodation class

7 Data by Accommodation class for states and territories are included in this publication. Accommodation class data has replaced star gradings data (known as star ratings by industry) , however the star grading categories can be fully mapped to the Accommodation class categories. The Accommodation class mappings are:

  • Budget: one and two star rated establishments or equivalent
  • Mid scale: three star rated establishments or equivalent
  • Upscale: four star rated establishments or equivalent
  • Luxury: five star rated establishments or equivalent

Statistical geography

8 Small area statistics for 2015-16 are classified to the Australian Statistical Geography Standard (ASGS): Volume 3 - Non ABS Structures, 2015 Edition (cat. no. 1270.0.55.003) effective from September quarter 2015.

9 Small area data (SA2) are aggregated to tourism regions as defined by relevant state and territory tourism organisations. Tourism regions are reviewed annually and are subject to boundary and name changes. Where changes have occurred, care should be taken when making comparisons with previously published data at this level.

10 Data by tourism regions and small area (SA2) are included in this publication and can be located in the downloads section. Small area data up to and including June quarter 2013 are available in Tourist Accommodation, Small Area Data (cat. no. 8635.0.55.002 for national data and cat. no. 8635.1.55.001 - 8635.8.55.001 for state/territory data).

11 Details of the composition of tourism regions and maps of tourism regions are provided in the ABS publication Tourism Region Maps and Allocation File (cat. no. 9503.0.55.001) available from the ABS web site

Data quality

12 The survey does not have a sample component and the data are not subject to sampling variability. However, other inaccuracies collectively referred to as non-sampling error may affect the data. These non-sampling errors may arise from a number of sources, including:

  • errors in the reporting of data by providers
  • errors in the process of capturing data
  • imputation for missing data
  • definition and classification errors
  • incomplete coverage.

13 Every effort has been made to reduce non-sampling error to a minimum by careful design and testing of questionnaires, and efficient operating procedures and systems used to compile statistics.

Response rates

14 The quality and reliability of survey data can be affected by the degree of response to a survey however, it is rare to achieve a 100% response rate for any survey. The response rates for the Survey of Tourist Accommodation at state level are shown below.

Response rates(a), hotels, motels and serviced apartments


a. Only one response rate is available for the financial year as the collection of data for the four quarters is carried out at one time at the end of the period.

Imputation rates

15 Missing data items are replaced by imputed values based on reported data. Average quarterly movements are applied to previously reported data for each non-responding unit to estimate values for missing data items. Only if previously reported data are not available, will data from a similar unit be used as a 'donor' for the missing data items.

16 The imputation rates for Room nights occupied and Takings from accommodation for the most recent quarters at a national level are shown below.

Imputation rates, room nights occupied

 Sep Qtr 2014Dec Qtr 2014Mar Qtr 2015Jun Qtr 2015Sep Qtr 2015Dec Qtr 2015Mar Qtr 2016Jun Qtr 2016
Licensed hotels and resorts11.311.
Motels, private hotels and guest houses16.716.116.216.414.414.114.013.8
Serviced apartments16.717.317.
Hotels, motels and serviced apartments14.314.314.214.210.410.09.89.9



Imputation rates, takings from accommodation

 Sep Qtr 2014Dec Qtr 2014Mar Qtr 2015Jun Qtr 2015Sep Qtr 2015Dec Qtr 2015Mar Qtr 2016Jun Qtr 2016
Licensed hotels and resorts10.310.19.810.
Motels, private hotels and guest houses15.915.215.415.613.213.012.812.9
Serviced apartments16.517.517.316.
Hotels, motels and serviced apartments13.413.413.



Seasonal adjustment

17 Seasonal adjustment is a means of removing the estimated effects of normal seasonal variation from the original time series so that the effect of other influences on the series may be more clearly recognised. Seasonal adjustment procedures do not aim to remove the irregular or non-seasonal influences which may be present in any particular quarter. Irregular influences that are highly volatile can make it difficult to interpret the movement of the series even after adjustment for seasonal variation, and cannot be assumed to indicate changes in the trend.

18 While the Concurrent method of seasonal adjustment is used, the seasonally adjusted and trend series have been revised following the annual review of the seasonal adjustment on data up to June quarter 2015. Since the collection frequency of the STA moved from quarterly to annual on a financial year basis from 1 July 2013, the annual review was performed to quality assure the seasonal adjustment process. As a result, the seasonally adjusted and trend estimates have been revised.

19 The Survey of Tourist Accommodation collection uses Autoregressive Integrated Moving Average (ARIMA) modelling techniques for the majority of applicable time series. The revision properties of the seasonally adjusted and trend estimates can be improved by the use of ARIMA modelling. ARIMA modelling relies on the characteristics of the series being analysed to project future period data. The projected values are temporary, intermediate values, that are only used internally to improve the estimation of the seasonal factors. The projected data do not affect the original estimates and are discarded at the end of the seasonal adjustment process.

20 For more information on the details of ARIMA modelling see the feature article 'Use of ARIMA modelling to reduce revisions' in the October 2004 issue of Australian Economic Indicators (cat. no. 1350.0). Any queries regarding the ARIMA modelling should be directed to Time Series Analysis on (02) 6252 6345 or email

21 Unreliable seasonal adjustment: In using the seasonally adjusted series, care should be exercised for the following data series: Takings, Australian Capital Territory because of the difficulties associated with reliably estimating the seasonal pattern. This series will be revised during the next annual seasonal review.

Trend estimates

22 Smoothing the seasonally adjusted series reduces the impact of the irregular component of the seasonally adjusted series and creates the trend estimates. The trend estimates are derived by applying a 7-term Henderson moving average to the quarterly seasonally adjusted series. The Henderson moving average used in the middle of the time series is symmetric but, as the end of a time series is approached, asymmetric forms of the symmetric moving average are applied. Unlike the weights of the symmetric 7-term Henderson moving average, the asymmetric weights have been tailored to suit the particular characteristics of individual series.

23 While these techniques enable trend estimates for the latest period to be produced, the process does result in revisions to the trend estimates in recent quarters, particularly as additional original estimates become available. For further information refer to Information Paper: A Guide to Interpreting Time Series - Monitoring Trends (cat. no. 1349.0) available at the ABS web site or contact Time Series Analysis on (02) 6252 6345 or email

Confidentialisation of data

24 Under the Census and Statistics Act, when releasing statistics the ABS is required to do this in a manner that is "not likely" (in a legal sense) to enable the identification of a particular person or organisation. A number of techniques are used to do this, including suppression of information. To ensure provider confidentiality in the Survey of Tourist Accommodation, the ABS uses a computerised process known as Disclosure Avoidance Analysis System (DAAS) to confidentialise the entire tourist accommodation dataset each quarter. This process not only ensures that data are suppressed to ensure individual establishments cannot be identified, but also suppresses data in other (consequential) cells to ensure data cannot be derived through deduction from the information available.

User aggregation of data

25 The aggregation of data by users across time periods should be undertaken with caution, due to the possibility of non-inclusion of confidentialised data (see the above section for more information about confidentialisation). Where one or more cells contributing to a total have been confidentialised (ie, contains the value of n.p.), the resulting aggregated total will be incorrect. However, some broader levels of data may not be affected by confidentialised cells.

26 Where data can be aggregated (ie, no confidentialised cells are included) for calendar and financial year/s purposes, the data items Establishments, Rooms and Bed spaces should not be aggregated. For these items it is recommended that for calendar years, the value of the December quarter is used, and for financial years, the value of the June quarter is used.

27 Any data items that have been derived from other items collected in the survey cannot be aggregated (ie, all those with labels ending in 'rate' or commencing with 'average'). These items must be re-derived based on the aggregation of each of the quarterly items collected in the survey used in the derivation of the rate or average (see Glossary for formulas).

28 Users are cautioned against deriving any non-standard aggregations (eg, aggregation of selected geographical areas such as capital city areas and balance of state; aggregation of selected activities such as hotels and motels combined). This is because data are confidentialised based on the standard data item structure.

Effects of rounding

29 Where figures have been rounded, discrepancies may occur between totals and the sum of the component items.

30 Estimates of movement shown in this publication are obtained by taking the difference of unrounded estimates. The movement is then rounded to one decimal place. Therefore where a discrepancy occurs between the reported movement and the difference of the rounded estimates, the reported movement will be more accurate.

Related publications

31 Other ABS publications and products which may be of interest are outlined below. All publications released from 1998 onwards are available on the ABS web site
Tourist Accommodation, Small Area Data (cat. no. 8635.0.55.002) (data cube for Australia - issued quarterly)
Tourist Accommodation, Small Area Data (cat. no. 8635.1.55.001-8635.8.55.001) (data cubes for each state/territory - issued quarterly)
Tourism Region Maps and Allocation Files, Australia (cat. no. 9503.0.55.001) (annual)
Tourist Accommodation, Australia, Expanded Scope Collection (cat. no. 8635.0.55.001) (irregular)
Short-term Visitor Arrivals Estimates, Australia (cat. no. 3401.0.55.001) (issued monthly)
Overseas Arrivals and Departures, Australia (cat. no. 3401.0) (issued monthly)
Australian National Accounts, Tourism Satellite Account (cat. no. 5249.0) (annual)

32 The catalogue of current publications and other products is available from the ABS web site The ABS also issues release advice on the web site which detail products to be released both in the coming week and the next six months.

ABS data available on request

33 As well as the statistics included in this publication, the ABS has other relevant data available on request. Inquiries should be made to the National Information and Referral Service on 1300 135 070.

Technical note - survey of tourist accommodation


1 The Survey of Tourist Accommodation (STA) is a census of all in-scope accommodation establishments with 15 or more rooms. The STA frame is based on accommodation lists provided by industry sources. Annual frame maintenance of the collection always results in regular "ons" (openings) and "offs" (closures) of accommodation units on the STA frame. For the 2014-15 frame maintenance exercise, a new industry source was used which resulted in a much larger number of units (279) being added to the STA frame than usual. The addition of these establishments resulted in a break in the series. See the Technical note in the 2014-15 release for further information.

2 During processing of 2015-16 STA data, it became clear that 40 accommodation establishments added during the 2014-15 frame maintenance exercise were duplicates of establishments already on the frame. As a consequence the 2014-15 STA data has been overstated. This overstatement is not significant, with the duplicated units representing less than 1% of the total survey frame (4,478 units) for 2014-15. Data for 2015-16 is not affected by the duplicated records.

3 Table 1 below shows the effect of the duplicates on the total establishment count for 2014-15.

Table 1: Number of establishments added as a result of 2014-15 frame maintenance process, state/territory, September quarter 2014

 2014-15 Original2014-15 Duplicates2014-15 Revised 1
New South Wales731360
South Australia1358
Western Australia16313
Northern Territory808
Australian Capital Territory927

1. Duplicate establishments and their associated values have been removed.

4 Whilst the overstatement is small, the change in data as a result of removing the duplicate records is significant enough to impact the direction of the net change in some state level data between 2014-15 and 2015-16 in the original estimates. Therefore users should exercise caution when comparing 2014-15 and 2015-16 original data. Note that the duplicates have no impact on the seasonally adjusted and trend estimates due to the methodology used to calculate these series.

5 To assist users in interpreting movements between the 2014-15 and the 2015-16 series, revised 2014-15 state and national level estimates (in original terms) for key data items (number of establishments, number of rooms, room nights, occupancy rates and takings from accommodation) are included in the Technical note tables in the Data downloads section of this release. These tables show aggregates for 2014-15 as previously published as well as revised 2014-15 data with the duplicates removed. Movements between these respective series and the latest 2015-16 data are also shown. Tables 8 to 11 show revised 2014-15 data including national, state and limited tourism region data. ABS does not generally revise STA data due to confidentiality limitations, and so a complete set of revised data including Tourism Region and Statistical Area 2 for 2014-15 will not be released.

6 Users should contact the National Information and Referral Service on 1300 135 070 for further information on the impacts of the duplicates.


Show all

Quality declaration - summary

Institutional environment








Show all

Back to top of the page