We have introduced a number of initiatives in the last five years that have reduced respondent load, but a strategic framework for the issue has not yet been formalised. The purpose of Statistics New Zealand's Respondent Load Strategy is to reduce the load on respondents while maintaining output quality.
Where future information needs outweigh the additional load of introducing new surveys, we will make every reasonable effort to minimise the increase in respondent load.
By adopting this strategy, we also intend to demonstrate best practice for the management of respondent load across the Official Statistics System. Official statistics producers have agreed to the framework provided by Principles and Protocols for Producers of Tier 1 Statistics and respondent management is the third protocol in the framework. Aligning with this protocol will make it easier for partner agencies in the Official Statistics System to adopt successful elements of the strategy as we implement it.
The strategy has been structured around the goals outlined in the Statement of Intent 2007. A summary of the initiatives associated with these goals is as follows.
Demonstrate to respondents the value of the information collected
We must ensure that there is a reasonable value exchange when respondents supply data. We can achieve this through communicating how government uses the data and by promoting ways that respondents can use the data.
Understanding respondents and their attitudes towards surveys is critical to knowing how perceived load contributes to non-response. It is also important to know if non-respondents share common characteristics, as this may affect data quality as well as the development of new strategies to improve or maintain coverage.
Minimise the load on respondents
When direct surveying is the only option, many methodological decisions can still help minimise the load imposed. These options include improved sample rotation techniques for business surveys and better data modelling.
We have made great use of administrative data over the last 10 years, which has significantly reduced the amount of direct surveying, particularly for small businesses. The Respondent Load Strategy reinforces this commitment by requiring a thorough investigation of possible administrative sources as a first step for all new survey developments. The strategy also promotes wider reuse of data collected across the Official Statistics System.
An advocacy position will be created to ensure that we address respondent issues and senior managers receive objective advice.
Assessing respondent load is a well-established principle that we already apply when developing new collections. The main initiative in this area is to strengthen how we make this assessment and shift it to the front of the approval process. Survey approval documentation must also be explicit about the load imposed balanced against the value of the collection.
The over-sampling of Māori is a major issue for social surveying and we must investigate alternatives to alleviate this.
Make it easier to respond to surveys
Targeting surveys at the most appropriate source is a good survey design principle that is already well established at Statistics NZ. While this principle is closely linked with the use of administrative sources and the reuse of other survey data, it also relates to targeting surveys at the most appropriate respondent.
Better maintenance of contact details improves the management of respondents in existing surveys, and is another way of ensuring that we direct surveys to the people in the household or organisation who can provide the information.
The way in which a survey is administered can have a large bearing on the perceived load associated with its content. By developing response mode options, such as supplementing paper surveys with Internet forms (as used for the 2006 Census of Population and Dwellings) or telephone interviews, respondents can choose the method they find most comfortable.
We will continue to look at the possibility of integration with standard business reporting tools (for example, eXtensible Business Reporting Language (XBRL)) as an alternative to manually completing forms.
Providing basic information relating to present and future survey obligations is also important for helping respondents plan their survey completion activity.
We must also ensure that frontline collection areas are responsive to Māori issues and have Māori language capability where required.
Identify and manage areas of unreasonable load
Measuring respondent load is fundamental to ensuring we develop policies that balance the need for official statistics against respondent load. Measuring load enables agencies to understand and report on existing levels of load, and provides a benchmark against which to measure changes in trends and/or the effect of new policies or initiatives. The consistent measurement of respondent load will enable comparisons of respondent load across the Official Statistics System.
The initiatives that will contribute to this include creating a standard time-taken question for surveys, and developing a consistent way of editing the data collected. It will also be important to improve our record-keeping so that we maintain survey participation information centrally. Because respondent load is a multi-dimensional issue, we will need to develop other indicators relating to concepts like complexity and sensitivity.
Developing parameters to help identify more heavily burdened respondents will also support this.