While Official Statistics Research was being established, nine interim projects were funded for the 2004/2005 fiscal year.
Using small area estimation methods to enhance statistics on Māori | Investigating methods to create synthetic datasets | Understanding the effect of changing thresholds in confidentiality rules | Developing a methodology for constructing regional spatial indexes | Developing an inward foreign affiliates trade statistic (FATS) for New Zealand | Investigating ethnic responses | Investigating selective editing | Updating telephone sampling methods | Investigating the translation of employees across the labour market to higher paying jobs
The work was conducted jointly by external researchers and Statistics New Zealand staff in order to maintain the collaborative nature of the programme. Projects were completed in 2005 and all final reports are available in the Official Statistics Research Series. OS Research invited the public to scheduled Officials Statistics System Seminars throughout 2005 and 2006. The results are widely applicable to government agencies that contribute to the Official Statistics System.
For any inquiries, please contact OS Research at: osresearch@stats.govt.nz.
1. Using small area estimation methods to enhance statistics on Māori
Small Area Estimation
Priority areas for small area estimation will be determined by the Māori Statistics Unit of Statistics New Zealand in order to develop and identify the sources of information and appropriate methods to use to deliver better, more timely information and improve the range and quality of statistics about Māori.
Team members:
- Systemetrics Research Ltd
- Statistics New Zealand.
Back to top
2. Investigating methods to create synthetic datasets
Synthetic Datasets
Synthetic datasets mimic genuine data in that they preserve the statistical properties of the original. The primary methodology being examined is smoothing data for tables (which would provide some confidentiality protection), and could extend to creating unit records.
Team members:
- Christchurch School of Medicine University of Otago Statistics New Zealand.
Back to top
3. Understanding the effect of changing thresholds in confidentiality rules
Unique and Disclosure Control Design
The first objective of this project is to better describe the risks associated with disclosure and information loss in a way that helps senior management make a good decision about the appropriate levels for Census 2006 tables. The second objective is to provide a means for discussing how information loss and disclosure risk occurs when there is a change in the census small population threshold value (areas with populations below the threshold may only have totals published).
Team members:
- MasterWork Software Ltd Statistics New Zealand.
Back to top
4. Developing a methodology for constructing regional spatial indexes
Developing a Methodology for Constructing Spatial Cost of Living Indexes
Two objectives best describe the goals of this project:
- to develop an appropriate methodology for constructing regional spatial cost of living indexes; and
- to assess the fitness for purpose of existing CPI price data at a high level.
Team members:
- University of New South Wales Statistics New Zealand.
Back to top
5. Developing an inward foreign affiliates trade statistic (FATS) for New Zealand
Investigation of New Zealand's Inward Foreign Affiliate Trade Statistics Using Existing Data Sources
The project involves an investigation of FATS definitions, what other countries have achieved in the area, what data Statistics New Zealand has available, and what can be used to create estimates for New Zealand's inward FATS. Following this, relevant variables in the New Zealand context will be established and the type of analysis these statistics could support will be determined.
Team members:
- New Zealand Institute of Economic Research Statistics New Zealand.
Back to top
6. Investigating ethnic responses
Ethnic Intermarriage in New Zealand
This research will progress implementation of the Review of the Measurement of Ethnicity recommendations and provide background for future ethnic measurement. The research includes investigating the relationship between ethnic intermarriage and multiple ethnic responses on a survey. Ethnic dynamics of families and households will be explained by studying ethnic residential concentration and diversity at the household level, alongside geographic ethnic concentration and diversity. Finally, intergenerational transmission of ethnicity will be investigated by using census household data to analyse factors such as parent and child ethnic comparisons, including families where 'New Zealander' responses were given, and studying how ethnicity may be related to spatial dimensions.
Team members:
- Victoria University of Wellington
- Statistics New Zealand.
Back to top
7. Investigating selective editing
Feasibility study of Selective Editing
This project is a feasibility study exploring the implementation of the selective editing methodology. Data from Overseas Trade and the Quarterly Employment Study (QES) will serve as case studies. Included in this study will be an international literature review of current best practices, the development of high-level principles for Statistics New Zealand and the public sector, and recommendations for specific Overseas Trade and QES needs, as well as the general applicability of the methodology for the public sector.
Team members:
- University of Canterbury
- Statistics New Zealand.
Back to top
8. Updating telephone sampling methods
The purpose of this project is to update Statistics New Zealand and the public sector about how changes in telecommunications technology or the telecommunications environment impact on the sampling of individuals or households using telephone numbers as the sampling frame.
Team members:
- Statistics Research Associates, Statistics New Zealand.
Back to top
9. Investigating the translation of employees across the labour market to higher paying jobs
This research is being undertaken to improve understanding of the nature of wage rate adjustment to changing labour market conditions.
Team members:
- University of Auckland
- Statistics New Zealand.