New Approaches to Sample Design in Small Island States

Among other planning, evaluation and indicator frameworks, the Sustainable Development Goals require the production and reporting of statistics that can be disaggregated by various sub-populations, such as sex, age, wealth quintile and disability status.  The small size, and geographic spread, of Pacific populations pose significant challenges to achieving disaggregated and statistically significant reporting of progress against development indicators.  The collection of data that can produce disaggregated indicators with a reasonable degree of accuracy, is expensive, particularly considering the geographic context of the Pacific region, and it places significant demand on statistical systems, including respondent burden resulting in ‘survey fatigue’.

In response to these challenges, the Statistics for Development Division (SDD) of the Pacific Community (SPC), in collaboration with partners, is developing has developed a series of sampling guidelines for the Pacific region.  To date, the series of sampling guidelines include:

1. Introduction to surveys and sampling;
2. Overview of sampling theory;
3. Sample design for household income and expenditure surveys (HIES), including case studies for survey practitioners and example in Excel;
4. Sampling considerations for disability related surveys; and
5. Sample design for agricultural surveys.

The draft guidelines are accessible herein.

The production of the series of sampling guidelines is a result of collaboration among multiple people representing multiple technical and development organisations.  For the preparation of the introduction, overview of sampling theory and sample design for HIES, we acknowledge Bertrand Buffiere, Pacific Community, Bruce Fraser, Australian Bureau of Statistics, Kristen Himelein, World Bank, Chris Ryan, United Nations Economic and Social Commission for Asia and the Pacific, Tracey Savage, Statistics New Zealand, and Taggy Tangimetua, Government of the Cook Islands.  For the preparation of the sampling considerations for disability related surveys, we acknowledge Daniel Mont, Centre for Inclusive Policy, and Scott Pontifex, Pacific Community.  For the preparation of the sample design for agricultural survey, we acknowledge Dramane Bako, Food and Agriculture Organisation of the United Nations.  We’d also like to acknowledge the Pacific Statistics Methods Board, particularly Vince Galvin and Ofa Ketu’u, Statistics New Zealand, and Mike Sharp, Pacific Community, for overseeing and coordinating the development of the guidelines.  We also acknowledge the countless Pacific statisticians who provided feedback on the guidelines and ensured the content is suitable in the context of the Pacific region.

While all of the acknowledged technical and development partners made financial contributions, for which we are grateful for, we’d like to acknowledge the World Bank’s Trust Fund for Statistical Capacity Building which provided financial support to SPC for the development of the guidelines.  

The next steps for parts 1, 2 and 3 of the guidelines is to consult representatives from Pacific statistical agencies who are attending a sampling workshop that is being held in Nadi, Fiji, from 24 February to 3 March, 2020, to ensure the guidelines are appropriate and relevant.  The disability and agricultural chapters are in earlier stages of their development and they will be subject to further review and consultation.  Endorsement from the Pacific Statistics Methods Board will be sought prior to publishing the guidelines.  

Please check back here regularly for progress. All comments (below) are welcome.