Review Findings

• Datasets Are Technically Fit for Use at a National Level
• High Respondent Burden May Compromise the Quality of the Data
• Lack of Reliable Data for Smaller Geographic Areas
• Unmet Demand for Qualitative Measures Such as Satisfaction and Motivation
• Dissemination of Data Can Be Improved
• Need to Understand the Needs of Users and to Increase Their Awareness and Knowledge of Data
• Integration of the Datasets Requires Aligning the Classifications and Concepts
• Data Management Processes Need to Be Implemented
• Implementation Capability Is a Key Requirement

The scope of the analysis undertaken in this review relates to the Core Tourism Datasets which comprise the six datasets as set out in Figure 1 below:

Figure 1. Core Tourism Datasets

Source: Tourism Research Council New Zealand, 2002

The analysis of these datasets undertaken during the course of the review process enables the following conclusions to be drawn:

Datasets Are Technically Fit for Use at a National Level

The core tourism datasets are fit for use at a national level and so provide reliable data at this level. Each of the datasets were initially designed primarily to provide national data, including to to input into Tourism New Zealand's marketing programme, the Tourism Satellite Account and the Balance of Payments. The survey methodologies are sound and in line with current best practice.

The survey objectives need to be reviewed and clearly defined to state why the data is needed and what it is used for, and provide a clear focus to the surveys. The questions in the questionnaires should be reviewed to ensure that the design achieves the survey objectives.

High Respondent Burden May Compromise the Quality of the Data

There is a high respondent burden in the CAM, IVS and DTS. It is estimated that respondents spend 10 to 18 hours a year completing the CAM questionnaires, and IVS and DTS interviews can take considerably more than the usual 10 to 15 minutes. This time commitment can lead to respondent fatigue and consequently a reduction in the quality of the responses.

Lack of Reliable Data for Smaller Geographic Areas

Some of the national level data collections generate information for lower level geographic areas, including at regional or local levels. For instance, the CAM can provide data down to area unit level (for instance, suburb or town) and this information is available subject to confidentiality requirements. The DTS can provide reliable data for regions with large populations, and data is available from the IVS for regions with high volumes of overseas visitors. Forecasts are provided for regions. The TSA and IVA provide national level data only.

However, data available at a regional level does not meet the needs of all regional or local-level decision-makers, for instance many Regional Tourism Organisations (RTOs) or local operators. RTOs require data at a range of regional or local levels, and many operators would like data at an even finer geographic level. Consultation with such users is required to determine what data is required (and what it will be used for) and what geographic level is required.

Data for smaller geographic areas will not be available from the IVS and DTS even if the sample size was increased. IVS data availability is determined by travel flows, that is, where the majority of overseas visitors go. As the IVS is a sample survey, data for visitors who travel outside the main tourism centres is likely to have a high sampling error because of the low number of those visitors actually surveyed.

The DTS sampling is based on the population of each region - the higher the population, the more sample that is allocated to that region. The nature of the survey means there is little ability to allocate sample by destination of trip. To do so would require a complete survey redesign and would be expensive to implement.

As it is unlikely that demand for data of smaller geographic areas can be met from the current surveys, other options will have to be investigated. Other data collection and collation methods could be used, such as supplementary surveys and data modelling for small area estimation.

Unmet Demand for Qualitative Measures Such as Satisfaction and Motivation

Currently, only the IVS contains questions about satisfaction, motivation and quality of experience. These questions provide feedback about the trip overall, and do not gauge satisfaction about any particular destinations or activities.

There are currently no equivalent questions in the DTS and as such the satisfaction and motivations of domestic visitors cannot be assessed.

Dissemination of Data Can Be Improved

The release of data from all datasets on the one website, and the introduction of Harmoni for compiling tabular data, will enhance the accessibility of the datasets. This approach is under development for the TRCNZ website.

Publishing the metadata (information about the data, e.g. method, definitions and quality) on the website will help users understand and interpret the data. A catalogue of available information needs to be developed to raise awareness of what data is available and where or how to obtain it. A move to releasing DTS data in line with IVS data on a quarterly basis will improve the timeliness and utility of the data.

Need to Understand the Needs of Users and to Increase Their Awareness and Knowledge of Data

The needs of users for the data justify its collection, frequency of release and the range of data items involved. To better meet the needs of users, an understanding of their characteristics and their requirements is required. Not all needs will be met and there will always be a trade-off between user needs, budget constraints, data quality, respondent burden and maintaining confidentiality.

The tourism industry is a fragmented and cross-sectorial industry which is reflected in the diversity of users (and potential users). There is likely to be a range of knowledge and skill levels, frequency of use (from occasional to intensive) and level of analysis (from the broad level to the detailed).

Options for finding out more about current and potential users and their needs include:

• conducting a survey of users (however, this will not generate responses from those who are not currently users)
• interviewing key people (in consultation with Regional Tourism Organisations and the Tourism Industry Association) ensuring we have a good cross-section of users
• providing feedback mechanisms
• establishing user groups.

The outputs from the core tourism datasets should be easily, widely and equitably accessible. To increase usage and help users understand the data there is a need to:

• provide documentation on survey methodology, outputs and quality measures
• increase the awareness of the availability of tourism statistics
• provide guidelines on how to use and interpret the statistics
• establish which are the most effective dissemination channels
• monitor changes in user requirements.

Integration of the Datasets Requires Aligning the Classifications and Concepts

Key to the integration of the datasets is the use of common classifications and concepts, and a process for managing changes to these classifications. The following classifications are recommended for review:

• accommodation type
• activities
• age group
• country
• main reason for travel
• regional tourism organisation
• transport type
• travel companions.

Definitions and concepts to be reviewed including:

• day trip (DTS)
• domestic visitor (DTS)
• length of stay (IVA)
• usual environment (TSA).

Data Management Processes Need to Be Implemented

Alignment of the classifications and definitions will facilitate the integration of the core tourism datasets. To ensure this is not compromised there needs to be a process for introducing changes to the classifications or surveys. This process will include assessment of the impact of the proposed change, including subsequent changes to other datasets.

Both processes and protocols for data dissemination need to be implemented. These should be guided by the principle of easy, equitable access to data that is safeguarded against the release of confidential information. Adequate permanent funding for dissemination needs to be sought to ensure the greatest value is obtained from the data generated.

Implementation Capability Is a Key Requirement

The steps needing to be undertaken to accrue the benefits from the review are both comprehensive and some will involve considerable additional work.

As such, it is critical that appropriate human and financial resources are made available to enable the recommendations to be acted upon over the next few years with the appropriate degree of depth and completeness.