Stats NZ has a new website.

For new releases go to

www.stats.govt.nz

As we transition to our new site, you'll still find some Stats NZ information here on this archive site.

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
Non-Sampling Error in the Initial Design Phase of the Survey Process

Introduction

The design phase is the most effective phase to concentrate on to minimise the non-sampling error present in our surveys. This phase has the most significant effect on our overall data quality, and the importance of this area is reflected in the high priority placed on the control of the sources of non-sampling error in this phase. In addition, if we can control the sources of error in this area well, there are great benefits further on in the next two phases. For example, if the questionnaire design is very good, we may also reduce the respondent error and therefore will be able to reduce resources put into that area. This is an area where Statistics New Zealand invests a considerable amount of resources.

Select an area of interest from the two areas below to find more detailed information on the sources of non-sampling error. Learn why they can be a problem, what their impact is, what Statistics New Zealand currently does to minimise the associated error and what we will be doing in the future. The sources of error encountered in each area are presented according to their priority.

The Statistics NZ Business Frame

The Business Frame is a database of individual private and public sector businesses and organisations engaged in the production of goods and services in New Zealand. The prime objective of the core Business Frame is to provide an up to date and accurately classified list of all economically significant enterprises for designing, selecting and operating Statistics New Zealand’s economic and financial survey program.

The Business Frame is a critical part of the business survey process since it is the
starting point from which all other stages flow from. A large amount of work continues to ensure the frame is as complete and up-to-date (consistent with the real world situation) as possible. The majority of sources of error in this area are considered to be of high priority since the frame forms the population that we start with when we select a sample for almost all the economic surveys and therefore makes a significant contribution to the overall quality of our surveys.

More detailed information on the sources of error in the Business Frame.

top 

Questionnaire design at Statistics NZ

There are two different broad types of questionnaire design: the rollover of the questionnaire from one period to the next, and the new development or redevelopment of a questionnaire. The rollover of a questionnaire is usually accompanied by minor changes, for example, the date, logo and signatures. The new development or redevelopment of a questionnaire involves several stages, including setting/clarifying objectives, implementing client specifications and field testing. With the many changes that happen in technology and the economy, questionnaire design also needs to evolve to keep up-to-date with developments/changes to information sources, methods of capturing responses (for example, email, Internet, scanning), terminology, new business concepts, legislation, tax rules and shifts in the economy/new industries.

The questionnaire, in whatever form, is the tool from which we collect the data. If the questionnaire does not ask exactly what is required, in a way that the respondent can understand and answer, the resulting data quality will not be as high as required. For example, the data may be biased if lots of businesses similarly misinterpret a question and provide an incorrect answer. The design of the questionnaire needs to be as useful and appropriate as possible, for us and for respondents. A lot of work is done to produce high quality questionnaires, and this will continue in order to effectively control the sources of non-sampling error present in this area.

More detailed information on the sources of error in Questionnaire Design.

 

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
Top
  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+