The National Visitor Survey commenced in 1998 to provide an official measure of travel by Australian residents.
Each year, interviews are conducted with 120,000 residents aged 15 years and over. COVID-19 has impacted the target sample size. See history of changes below for more information.
Respondents are interviewed through a Computer Assisted Telephone Interviewing (CATI) system with phone numbers selected using random digit dialling.
The survey runs continuously with interviews taking place on each weekday and on weekends (excluding national public holidays). Residents classified as ‘in scope’ for interviewing include those who:
Respondents interviewed in the NVS are randomly sampled to be representative of the Australian population.
The survey uses specific recall periods to collect information on recent travel experiences. This includes details on:
Overnight trips must include at least one night away from home and be a minimum of 40 kilometres from the respondent’s usual place of residence.
Day trips must have a round trip distance of at least 50 kilometres from the respondent’s usual place of residence and a minimum duration of four hours. Day trips taken as part of an overnight trip, or those that are routine (for example, from home to work/school, or an intrinsic part of a person’s job), are not collected.
The survey contains over 70 questions related to:
NVS definitions are based on those provided by the United Nations World Tourism Organisation (UNWTO).
Interviews are conducted with people who have travelled for purposes including holiday, visiting friends and relatives (VFR), business, education and employment. To be included, travellers must not have been away from home continuously for more than 364 days, or 365 days in a leap year.
The NVS is co-funded by the Commonwealth, state and territory governments under the auspices of the Australian Standing Committee on Tourism (ASCOT).
Results from the NVS are published quarterly.
Weights for the NVS are calculated on an individual trip basis. They take into account:
The NVS is benchmarked to population estimates of those aged 15 years and over.
The NVS provides information on travel activity and spend by domestic visitors. Spend is only collected for the respondent’s entire journey, not for individual stops.
In order to determine the impact that the visitor activity is having on a particular region, Tourism Research Australia uses a model-based approach to allocate visitors’ spend to Australia’s tourism regions.
We revised estimates for August to November 2022. This was due to the Integrated Public Number Data (IPND) frame being unavailable in the period.
The NVS Computer Assisted Telephone Interviewing (CATI) approach has evolved in response to changes in the use of telecommunications and respondent behaviour.
A history of changes and impacts to be aware of when using the NVS data is outlined below.
COVID-19 has had a wide range of impacts across the Australian economy, increasing labour market competition, changing respondent behaviour, and increasing absenteeism within the workforce. The collection of National Visitor Survey data is being impacted by these unprecedented changes in the economy. Under normal circumstances, the National Visitor Survey targets a sample size of 120,000 interviews annually. Current estimates indicate that 90,000 interviews will be collected in the 2022 calendar year. This sample size still ensures reliable tourism estimates will be produced at the National, State and tourism region levels.
LGA area profiles are currently based on an average of four years of data to reduce variability and improve reliability. As such, LGA profiles are not expected to be significantly impacted. Users of more granular spatial data may see impacts in the reportability of data due to smaller sample sizes. Austrade recommends using caution when reporting estimates at this level and seeking guidance if required.
The NVS moved to a 100% mobile phone sample. The review conducted by TRA showed 97% of the Australian population aged 15 or more owned a mobile phone. The change improved the accuracy of national, state and territory estimates. However, the change in collection resulted in a series break between 2018 and 2019. Comparison between these years should be made with caution.
Interviews were evenly split across calls made to household landlines (50%) and mobile phones (50%) using a dual-frame sample design.
The sample size increased from 80,000 to 120,000.
All interviews were made through calls to household landlines.