In broad terms, the process of producing gender statistics is similar to that for other fields of statistics. There are mainly four different types of existing data sources: population censuses, population-based sample surveys, business surveys and administrative records. While the information they provide is complementary, they are based on different methodologies, which affect the type, range and quality of gender information they can provide.
The process of production of gender statistics involves the following key steps:
- selection of topics that need to be investigated
- identification of the data needed to understand gender differentials and women’s and men’s roles and contributions in the different spheres of life
- evaluation of existing concepts, definitions, and methods to produce unbiased gender-relevant information
- development of new concepts, definitions, and methods where necessary that adequately reflect the diversities of women and men in society
- development of the data collection instrument: choice of unit of enumeration about which to collect information; survey design, such as definition of sample size, questionnaire development and testing; training of enumerators, ensuring all will use a standard approach in data collection and avoid sources of gender bias
- collection and processing of data using practices that will deliver reliable results
- analysis and presentation of statistics in easy-to-use formats; dissemination of statistical products to a wide range of users including policymakers and planners.
For further information see
Developing Gender Statistics: a Practical Tool, Chapter 3 (2010)
Handbook for Producing National Statistical Reports on Women and Men, United Nations, New York (1997)
Engendering Statistics - A tool for change, Statistics Sweden (1996)