International Visitor Survey methodology
Information about the International Visitor Survey (IVS) methodology, including its purpose, quality targets, population, sampling structure and collection, data processing, analysis and outputs.
International Visitor Survey (IVS) details
International Visitor Survey (IVS)
Measure total annual expenditure by international visitors in New Zealand
Sample based survey
Top country markets
Australia, China, USA, UK, Germany, Japan
Target relative margins of error (RME)
Overall – 5%, top country markets – 10%
Target sample size
8,900 per year
Expected response rate
International Visitor Survey (IVS) purpose
Tourism is New Zealand’s largest export earner, and plays a critical role to New Zealand’s economic success. The International Visitor Survey was developed to provide accurate, quarterly, national information on the expenditure of international visitors to New Zealand, as well as their behaviours and characteristics.
The main purpose of the International Visitor Survey is to measure the total annual expenditure by international visitors in New Zealand .
The survey also:
- provides data for determining the exports of travel services in the Balance of Payments, and tourism expenditure in the Tourism Satellite Account
- measures the amount of expenditure per visitor for the top 6 key international markets (Australia, China, USA, UK, Germany and Japan)
- determines the activities international visitors participate in, the transport and accommodation types used, and places visited
- provides demographic information about international visitors, their motivation for visiting New Zealand, and their satisfaction with their visit to New Zealand.
Note: 'Expenditure' refers to spend by travellers aged 15+, and excludes individuals whose purpose of visit to New Zealand was to attend a recognised educational institute and are foreign-fee paying students. Expenditure also excludes international airfares.
The IVS is one of New Zealand's Tier 1 statistics, and its result is a key input to the National Accounts.
Stats NZ's website has more information about these:
- Stats NZ website — Tier 1 statistics(external link)
- Stats NZ website — National Accounts(external link)
As the IVS was designed to provide estimates at the national level, regional breakdowns of expenditure can't be produced. The Monthly Regional Tourism Estimates (MTREs) provide monthly estimates of spend by region, regional tourism organisation (RTO) and territorial authority.
The following organisations are involved with the IVS:
- Kantar TNS New Zealand is commissioned for data collection and processing
- we conduct analysis and dissemination of survey results
- the survey is carried out with the assistance of the airports, security staff, airlines and Stats NZ.
Survey quality targets
To produce accurate estimates of each of the outputs, the survey was designed to meet a:
- 5% margin of error for the total annual expenditure
- 10% relative margin of error (RME) for each of the top 6 country markets.
To achieve these margins of error, a minimum of 8,900 international visitors to New Zealand are sampled per year. The sample also has target allocation sizes for each of the top 6 markets to ensure that there are enough respondents to meet the 10% threshold.
In order to achieve these desired sample sizes, the response rate of eligible visitors must be taken into account. There are a number of reasons the IVS may not be completed by visitors, including:
- invalid email addresses
- they don't want to respond (‘non-response’)
- invalid responses.
The survey is designed assuming a response rate of 30%. To achieve this sample size, details of approximately 30,000 qualifying visitors need to be collected.
We use the World Tourism Organisation definition of an international visitor to define the population of interest for estimating total expenditure of international visitors to New Zealand:
“persons who travel to a country other than that in which they have their residence, but outside their usual environment, for a period not exceeding 12 months and whose main purpose of visit is other than the exercise of an activity remunerated from within the country visited.”
Given this definition, the population excludes:
- individuals who reside in New Zealand (New Zealand was their last place of residence for 12 months or more)
- individuals who have been in New Zealand for longer than 365 days
- those in transit who haven't formally entered New Zealand (they haven't filled out an arrival card and handed it into Customs)
- individuals who visit New Zealand to serve in the armed forces, for diplomatic or consular business, or aircrew on duty or between flights
- individuals whose next country of permanent residence is New Zealand.
The target population also excludes:
- Those aged less than 15 years, because parental consent is required to interview those under 15, and this is not practical at the interview screening stage.
- Those whose primary purpose of visit were to attend a recognised educational institute and will be foreign fee paying students. The spending of this group is estimated separately as their spending patterns are significantly different from other international visitors, which can create problems in calculating estimates of total spend.
Survey population differs slightly to the target population
The survey population differs from the target population due to legal and administrative reasons. Individuals in the target population not covered by the survey population are those who:
- have formally entered New Zealand but haven't stayed in New Zealand for at least 1 night during their trip
- depart New Zealand from a port that isn't Auckland, Wellington, Christchurch or Queenstown international airport — for example, Dunedin, Rotorua, or any seaport
- depart New Zealand on private aircraft
- don't understand one of the survey languages (English, Japanese, Chinese, Korean and German) sufficiently to be able to answer the questions
- don't have an email address to which the survey can be sent.
There are also individuals not in the target population but who are included in the survey population. These include those whose main purpose of visiting New Zealand is to work in New Zealand.
This population is included due to difficulty in distinguishing between:
- visiting New Zealand for business purposes (included in the target population)
- visiting New Zealand for the purpose of working for an employer in New Zealand (excluded from the target population).
The sampling schedule is constructed to include international visitors to New Zealand departing from Auckland, Wellington, Christchurch and Queenstown international airports. Previously, time periods with a certain number of flights depending on the airport, were randomly selected and used for collecting the sample. The probabilities of a time period being selected are based on the number of flights during that period over the previous year.
We have now moved to a flight-based approach where 6 sample units (email addresses) are collected per flight. We use this to estimate the number of flights to be sampled per strata.
The International Visitor Survey uses a 2-step collection process:
- The first step involves screening departing visitors during the selected time periods for eligibility, and collecting email addresses:
- Interviewers approach departees at random to ask them if they would be willing to fill in the survey.
- Each person approached is provided with a brochure that explains the purpose of the survey and the benefits to New Zealand (the brochure is available in English, Japanese, Spanish, French, Chinese, Korean and German).
- If the person agrees, the interviewer checks they are eligible and obtains a few details, including an email address, text or QR code depending on preferences on how people would like to receive the survey.
- The second step, where the bulk of the information is captured, is via an online questionnaire, which is available in English, Japanese, Chinese, Korean, French, Spanish and German. A link to the questionnaire is sent to those eligible who have agreed to participate.
All variables in the IVS are subject to editing to ensure validity and consistency in responses. This includes:
- data cleaning checks performed on the data to identify potentially incorrect data and manually edited if needed
- adjusting expenditure data to represent expenditure in New Zealand dollars, per person, for travel in New Zealand
- adjusting expenditure — visitors on packages and tours are unable to remove the airfare from their expenditure, so the airfare component is calculated and removed
- in addition to airfare cost removal, some packages and tours span more than one country, so the expenditure must be adjusted to remove costs in other countries.
The sample is weighted to represent the survey’s target population
Each respondent within the sample is weighted to represent a fraction of the total number of all international visitors departing New Zealand within the survey’s target population. Survey response weights are adjusted to reflect the unequal probabilities of respondent selection from the composition of the target population, and known discrepancies between the sample and the population definitions.
The weighting process is broken down into 4 stages. The aim of each stage is to adjust for known discrepancies between the sample and the population for:
- ages groups by airport (stage 1)
- gender by age groups within country of residence (stage 2)
- purpose of visit and length of stay bands within country of residence (stage 3)
- totals for each of the top 6 country of resident markets (stage 4).
Detected outliers are treated
Outliers are detected using the Banff outlier treatment developed by Statistics Canada. The Banff process identifies cases that are too far from the median value, but allows the user to make an assessment as to how far from the median is acceptable. The identified outliers are then treated to reduce their weighting and impact on the results.
For more information see the document below.
Once the data has been processed, it's analysed to calculate statistics and identify trends. This involves looking at the data in the long-term, as well as the short-term, and breaking the data down to look factors such as total spend, average spend, spend per day, length of stay etc. The data can also be split by factors such as country of residence, purpose of visit etc.
“The outputs from the analysis are released at 11am on a predetermined date, which can be found on our Data release calendar.
The outputs include:
- Data visualisations on the Tourism Evidence and Insights Centre (TEIC), broken down by country or travel purpose, including
- total spend
- average or daily average spend
- visitor arrivals
- average length of stay