Tourism Research Australia (TRA) runs the International Visitor Survey (IVS) to measure the contribution of international tourism to Australia’s economy and provide input into modelling spend for its regions.

The IVS has imputed estimates from April 2020 to the current release. For further information see the history of changes below.

Survey participants

We sample 40,000 departing, short-term international travellers aged 15 years and over who have been visiting Australia. We use Computer Assisted Personal Interviewing (CAPI) to conduct the survey. This is done in the departure lounges of the 8 major international airports:

  • Sydney
  • Melbourne
  • Brisbane
  • Cairns
  • Perth
  • Adelaide
  • Darwin
  • Gold Coast.

Questions we ask

The IVS contains around 100 questions supported by 'show-cards'. We use these cards to help the respondent answer particular sections including:

  • usual place of residence
  • repeat visitation
  • group tours
  • travel party
  • sources for obtaining information about Australia
  • purpose of visit and places visited
  • transportation and accommodation
  • activities
  • expenditure
  • demographics.

Design of the survey

The National Survey and Data Science section in TRA manages survey design and management. We work closely with consultants, key stakeholders, and industry. This develops and maintains high data quality and relevant outputs.

The IVS has been surveying international visitors in 4 languages since 2004. These are English, Japanese, Mandarin and Korean. We distribute surveys among airports by selecting monthly samples of departing flights and visitors. We do this to achieve acceptable sample sizes in various categories.

How the survey is weighted

The survey results are weighted to Department of Home Affairs’ data on international visitor numbers. The variables used in weighting the data are:

  • country of residence
  • state of arrival
  • main purpose of journey
  • airport of departure
  • age and sex of visitor.

The Australian Bureau of Statistics (ABS) publishes overseas (visitor) Arrivals and Departures (OAD) data (ABS catalogue no. 3401.0) each month.

Data reliability

We base the results in the IVS on a sample, rather than a census, of international visitors to Australia. The results are subject to sampling variability. Note: If all international visitors to Australia were included in the survey, the resulting data might be different.

The relative standard error of the survey and its associated confidence interval indicate the extent to which an estimate might vary by chance from the true number. This is because we only include a sample of the population.

The table below provides the 95% confidence intervals for a range of estimates available in the IVS. There are approximately 19 chances in 20 that the true number is within the range identified by applying the figures in the table.

Size of 95% confidence interval for estimate

 Expressed as a percentage of the estimate.

Estimate

Visits

Nights

Expenditure

2,000

64.40%

#

#

5,000

41.50%

#

#

10,000

29.80%

#

#

20,000

21.40%

#

#

50,000

13.80%

#

#

100,000

9.90%

75.70%

#

200,000

7.10%

55.40%

#

500,000

4.60%

36.70%

#

1,000,000

3.30%

26.90%

#

2,000,000

2.40%

19.70%

#

5,000,000

1.50%

13.10%

97.60%

10,000,000

1.10%

9.60%

72.80%

20,000,000

 

7.00%

54.30%

50,000,000

 

4.60%

36.80%

100,000,000

 

3.40%

27.40%

200,000,000

 

2.50%

20.50%

500,000,000

 

1.70%

13.90%

1,000,000,000

 

1.20%

10.30%

2,000,000,000

 

0.90%

7.70%

5,000,000,000

 

0.60%

5.20%

10,000,000,000

 

0.40%

3.90%

#  95% confidence interval is greater than estimate.

For example, say the estimated number of Chinese visitors who stayed in Queensland was 100,000. Looking at the visits column in the table, an estimate of 100,000 visitors has a 95% confidence interval of 9.9%. We are 95% confident that the true number of Chinese visitors who stayed in Queensland was between 90,100 and 109,900 visitors (100,000 ± 9.9%).

We:

  • base estimates of variation on samplings with replacement principles, and make allowances for the IVS stratification
  • regress the covariance and estimates with a log transformation using Ordinary Least Squares (OLS) regression
  • create 3 independent models for visitors, nights, and expenditure.

The actual relationship modelled was:

  • ln (COV) = a + b * ln (ESTIMATE)
    • where, a = intercept, b = gradient (slope)

The model parameters were approximated as:

Visits

Nights

Expenditure

a. 2.521841

4.218551

5.837435b

b. -0.47831

-0.44906

-0.42362

History of changes

2022

We:

  • based IVS estimates for September quarter 2022 and December quarter 2022 on a combination of interview data and imputation
  • used interview data for airports with sufficient sample sizes
  • added a variable to the dataset to identify interview and imputed records
  • applied imputation at an aggregate level. Granular disaggregations or cross classifications of the imputed data during this period are not available.

We imputed IVS estimates from the March quarter 2020 to June quarter 2022. This was due to COVID-19 related international border closures and insufficient sample size of interviews. Data for country, main reason for visit, and state stopover will be valid at the aggregated level. As imputation was applied at an aggregate level, granular disaggregations or cross classifications of the data during this period are not available.

Due to revisions to OAD data, we revised estimates for:

  • January to March 2022
  • July 2021 to February 2022.

From April 2020 to February 2022 the OAD used a full count (rather than sample) of passenger cards. This was used to measure short-term visitor arrivals to Australia due to the small number of visitor movements occurring. We used more comprehensive OAD data for IVS imputation purposes.

2021

IVS interviewing:

  • recommenced at some airports in November 2021
  • was conducted at some airports from April to July 2021 during the New Zealand travel bubble.

2020

Due to COVID-19 there was no IVS interviewing from 1 April 2020. As a result, we imputed some estimates for the period from 1 April 2020.

TRA used trip characteristics from relevant past IVS records with the same profile to visitors in the OAD. We imputed values for a range of visitor characteristics for the period from 1 April 2020.

Spend values have been sensitivity checked against domestic spend behaviours by item. This is to ensure alignment with current spending patterns.

2018

Due to revisions to OAD data, TRA revised IVS estimates from 2005 to 2018 with the release of year ending December 2018 data. The revisions focus on purpose-of-visit estimates, with minimal change to visitor, nights and spend estimates.

2005

We conducted interviews between 2001 and 2004 with approximately 20,000 international visitors. Since 1 January 2005 we conducted interviews with 40,000 international visitors on an annual basis. We increased the sample size to enhance the estimates for smaller states, territories and regions. Increasing the sample size of the IVS by 100% has improved the reliability of survey estimates.