International Visitor Survey Revision 2008
Last updated:
July 2008
Introduction
The Tourism Stragegy Group has recently redeveloped the International Visitor Survey (IVS). The redevelopment of the IVS incorporates two methodological enhancements aimed at improving the reliability and quality of the results of the survey.
This paper outlines the results of methodological changes to the IVS:
- Weighting methodology
- Treatment of capital expenditure
Other changes to the IVS include:
- Managing the survey methodology and processes in-house and only outsourcing the data collection.
- New questionnaire tool and data provider.
These changes are being made to ensure that future estimates produced from the IVS accurately reflect international visitor activity in New Zealand.
All IVS estimates from the June quarter 2007 onwards incorporate the new weighting methodology and treatment of capital expenditure. To assist users in using the new estimates this has also been applied to all quarters back to the March quarter 2003.
Prior to 2003, the IVS used a different sample design and capital items cannot be easily identified. Therefore there is a break in the time series and results before, and after, the year ended December 2003 are not directly comparable.
The revised IVS series was released on Friday, 25 July 2008 and includes data to the year ended March 2008.
Revised Series
The following graph provides a comparison of the revised IVS expenditure series using the new weighting methodology and treatment of capital expenditure with the previously published IVS series and Statistics New Zealand (SNZ) IVS series used in the Balance of Payment statistics.
Weighting Methodology
The previously published IVS series uses a two stage weighting method developed in 2003 when the sample design changed.
In the first stage of weighting, the sample is weighted to flight stratum population totals from SNZ International Travel and Migration statistics. This stage of weighting accounts for the unequal probabilities of selection.
In the second stage of weighting, the weights are adjusted to country of residence by age by gender population totals, also from the International Travel and Migration statistics. This stage of weighting adjusts for discrepancies between the sample and the population.
A review of the weighting methodology in 2007 recommended the inclusion of a third stage of weighting to purpose of visit and length of stay population totals (also from the International Travel and Migration statistics).
The revised weighting methodology will:
- Correct the sample for known discrepancies between the sample and the target population.
- Ensure that estimates of the number of visitors by key visitor characteristics (main markets, purpose of visit) are consistent with the International Travel and Migration statistics.
- Reduce the variability of key estimates that are correlated with variables used in the weighting.
Treatment of Capital Items
The previously published IVS expenditure estimates include the purchase of capital items such as property by international visitors in New Zealand. Although the purchase of capital items in the IVS was relatively rare, the high unit value of this type of expenditure had a significant impact on survey estimates and they were often identified as outliers.
The treatment of this type of expenditure was not consistent over time. Often large and influential expenditure on capital items were down-weighted to a weight of one but smaller less influential expenditure on capital items were left to represent other purchases of capital items by non-sampled visitors.
A review of the treatment of capital items reported by visitors in the IVS in 2008 recommended non-tourism expenditure be removed from the IVS following the United Nations World Tourism Organisations (UNWTO) International Recommendations for Tourism Statistics (IRTS)
In particular, expenditure relating to the purchase of financial and non-financial assets such as land and real estate for private or business use and expenditure on goods for resale purposes be removed from the IVS
The revised treatment of capital expenditure will:
- Align the IVS with best practice guidelines from the UNWTO IRTS 2008.
- Improve the quality and stability of IVS expenditure estimates.
- Align IVS expenditure series with SNZ’s Balance of Payment series.
The result of changing the weighting methodology and removing capital items in the IVS are lower expenditure estimates than previously published but estimates that better reflect tourism related expenditure and the characteristics of international visitors travelling to New Zealand.
The following graph shows the main changes to the IVS since its establishment in the late 1990s. This highlights the key differences between the data prior to 2003, and after 2003.
ˆ Back to top