The production of gender statistics concerns the entire national statistical system and covers data from different sources and statistical fields (Hedman, Perucci and Sundström, 1996). Gender statistics may be considered a field of statistics (see, for example, UNECE and World Bank Institute, 2010); however, the production of gender statistics should not be misunderstood as being limited to the compilation of sex-disaggregated statistics from various statistical fields and their dissemination in gender-focused publications, reports or databases. Similar to the obtention of other statistics produced by national statistical systems, the obtention of gender statistics involves such stages as planning, data collection, data analysis and dissemination. Obtaining reliable gender statistics that adequately reflect differences and inequalities in the situation of women and men in all areas of life requires a strategy of gender mainstreaming in all stages of data production. This means that gender is brought into the “mainstream” of all statistical activities, rather than dealt with as an “add-on” (United Nations, 2002).
- + Mainstreaming a gender perspective in statistics
- Mainstreaming a gender perspective in statistics means that gender issues and gender-based biases are systematically taken into account in the production of all official statistics and at all stages of data production (Hedman, Perucci and Sundström, 1996; United Nations-, 2001a, 2001b, 2002, 2006). The Inter-agency and Expert Group on Gender Statistics and the various editions of the Global Forum on Gender Statistics organized by the United Nations recognized that it is important to institutionalize gender statistics in all sectors in order to secure its sustainability (United Nations, 2006, 2009, 2012). Gender statistics produced as an “add-on” field are often marginalized and fail to reach a wide range of users, including policymakers in domains other than gender equality, analysts and researchers. Moreover, their production may be more dependent on irregular economic and human resources. Mainstreaming a gender perspective in the national statistical system may also lead to more efficient coverage of gender issues and better coordination among data collection programmes in producing gender statistics.
National statistical systems need to regularly collect, analyse and disseminate data that address relevant gender issues. Gender statistics should document women’s and men’s participation in and contributions to all social and economic areas and reflect the underlying causes and consequences of gender inequality (Hedman, Perucci and Sundström, 1996; United Nations, 2002). The coverage of gender issues by official statistical systems and the adequacy of such systems should be regularly reviewed, as recommended in paragraph 207 (b) of the Beijing Platform for Action (United Nations, 1996). The review should make clear whether relevant gender issues, as defined by major data users, are covered by existing data collection programmes and made available to users. Based on this review, the strategy of gender mainstreaming can involve collecting new types of data expanding data collection in some areas to fill existing knowledge gaps and better disseminating data already collected (Hedman, Perucci and Sundström, 1996; United Nations, 2002). The strategy of gender mainstreaming should be based on strong collaboration between users and producers of data, strong internal coordination within the national statistical office and the national statistical system, and data-sharing agreements between the national statistical office and other agencies of the national statistical system or other producers of data.
Gender-sensitive concepts and methods should be used in data collection in all statistical fields. In order to provide reliable comparisons between women and men, gender statistics need to correctly measure women’s and men’s participation in and contribution to society (Hedman, Perucci and Sundström, 1996). Conventional concepts and methods used in data collection are often inadequate to reflect the realities of women and men. For example, some women’s activities and contribution to the economy and society are not adequately captured in statistics if old concepts of work and labour force, which do not take into account all forms of work, are used (see, for example, the section entitled “Work” in chapter II).
The units of enumeration and the units of data collection should be adequately chosen to support the production of data that would show meaningful gender differences. For example, gender statistics in agriculture should be based on an adequate coverage of all agricultural holdings, including smallholdings, where women are predominant; should include information on farm labour disaggregated by sex, age and other social and economic characteristics; and should cover aspects of management and ownership of agricultural resources at the most disaggregated level possible, such as the subholding and individual levels.
Furthermore, new concepts and new methods of data collection should be used for the production of gender statistics. For example, recent methodological developments in time-use surveys and violence against women surveys and changes towards more comprehensive statistics in national accounts so that unpaid work is covered should be integrated in the production of national statistics.
Improvement of content, methods, classifications and measurements from a gender perspective should be made part of the ongoing efforts to improve all statistical sources – censuses, surveys and administrative systems (Hedman, Perucci and Sundström, 1996). Mainstreaming a gender perspective in data collection programmes involves review and revision of the conceptual basis of data collection tools, review and revision of coding and classification systems and terminologies, gender training for all personnel involved in data collection, media campaigns that include gender-specific messages, gender-sensitive selection of field interviewers, and review and revision of tabulations and data presentation and dissemination (Corner, 2003).
The presentation and dissemination of gender statistics should reach all potential target groups. Most often, existing data are not fully exploited for obtaining gender statistics (United Nations, 2009). Furthermore, data are often analysed and presented without considering users’ needs and therefore fail to reach the target audiences (Hedman, Perucci and Sundström, 1996). However, the presentation and dissemination of data is a crucial area of work in gender statistics. Gender statistics and the results of data-based gender analysis should be disseminated to a wide range of users with a clear language that highlights gender-based causes and consequences and their policy implications (United Nations, 2002).
The dissemination of gender statistics should not be limited to gender-focused reports and databases. Restricting the activities concerning gender statistics to the compilation and dissemination of sex-disaggregated data and gender-sensitive indicators in gender publications only limits the audience and users of data. This restrictive approach may also perpetuate the perception that gender statistics are useful for women’s or gender advocates only. Gender statistics need to be taken into account not only in policies and programmes created to reduce gender inequality, but in all policies and programmes. It is important that statistics made available on a regular basis to policymakers include a gender dimension. As recommended in paragraph 207 (d) of the Beijing Platform for Action, Governments should use more gender-sensitive data in the formulation of policy and implementation of programmes and projects (United Nations, 1996). Presentation of gender statistics in regular statistical products produced by national statistical systems increases the accessibility of gender statistics and their chances of being taken into account in policymaking. If these documents fail to highlight the importance of the goal of gender equality and to incorporate relevant gender perspectives, an important opportunity is lost (United Nations, 2002).
Mainstreaming a gender perspective in data collection and presentation should be seen as part of the overall process of improving the quality of data produced by national statistical systems. Four components of overall statistical data quality are particularly impacted (UNECE, and World Bank Institute, 2010):
(a) Relevance. This is defined as the degree to which statistics meet the needs of users. Gender mainstreaming in statistics entails taking into account users’ needs. Gender statistics aim to address gender issues that are defined as relevant by policymakers, advocates, researchers and the public.
(b) Accuracy. This is defined as the closeness of statistical estimates to true values. Gender mainstreaming in data collection has a crucial role in reducing bias in data collection. For example, use of gender-sensitive data collection tools can prevent underreporting of women’s economic activity, underreporting of violence against women and undercounting of girls, their births and their deaths.
(c) Accessibility of data. Data on a variety of topics that are often associated with women’s interests are becoming available, such as statistics on time use, violence against women and family-work balance. Many gender statistics programmes also aim to make relevant gender-sensitive statistical information accessible to a wide range of audiences.
(d) Clarity. This is related to the presentation of data as well as to the availability of information on data quality and appropriate metadata. Gender mainstreaming pays particular attention to disseminating statistics in formats that are easily understood by a wide audience and making clear the limitations of data collected on the basis of concepts and methods that are not gender-sensitive.
- + Implications of gender mainstreaming in statistics at the organizational level
- Leadership. Mainstreaming a gender perspective in national statistical systems requires political will at all levels, not only in national statistical offices but also in the statistical services of other Government agencies and in all institutions that provide administrative data (United Nations, 2006). Sensitizing and raising the awareness of both users and producers of data is critical in linking gender statistics to policies. Moreover, policymakers and heads of national statistical offices should be fully involved in capacity-building and leadership in mainstreaming gender statistics in national statistical strategies (United Nations, 2006, 2009).
Legal framework. The development of gender statistics should be specified within the legal framework of official statistical systems (United Nations, 2006). Of crucial importance to improving the availability of gender statistics is the specification of formal requirements for sex-disaggregation and the incorporation of a gender perspective within the national statistical legislation that regulates the production and dissemination of official statistics. In order to expand the range of information available for gender analysis, requirements need to be established not only for statistics already officially collected by the national statistical office, but also for other sources of data, particularly administrative data being collected and disseminated by other Government agencies and by organizations in the public and private sectors (United Nations, 2006).
Cooperation between users and producers of statistics. A dialogue should be fostered between national statistical offices and interested stakeholders, including women’s groups. The general approach in the development of gender statistics has involved efforts to promote dialogue and understanding between statisticians and the various users of statistics—policymakers, representatives of non-governmental organizations, activists and researchers (United Nations, 2000). Dialogue between national statistical offices and interested stakeholders can enable data users to understand, gain access to and use gender statistics more effectively and help to increase the capacity of statisticians to identify and understand gender issues and to present data in formats that better address the needs of users (United Nations, 2006).
Collaboration in developing and improving concepts and methods. National statistical offices need to work with international and regional organizations and agencies and academic and research institutions to mainstream gender in the development and revision of concepts, definitions and methods of collecting data on topics where methods are inadequate (United Nations, 2006). This collaboration extends to all methodological issues, including the design of survey questionnaires or modules within questionnaires, the revision of international classifications and standards and the development of analytical methods and appropriate indicators, among others.
Training. Statisticians should be trained in how to incorporate a gender perspective into their regular work, from the design of data collection tools and fieldwork to data analysis and presentation. At the second Global Forum on Gender Statistics, held in Accra in 2009, it was recognized that, despite global efforts and advocacy for gender statistics, including capacity-building activities, there still persist gaps in knowledge of gender statistics among many professionals in national statistical offices (United Nations, 2009). Regular training is a key component in ensuring the mainstreaming of gender in national statistical systems and the sustainability of programmes. In particular, producers of statistics need to be trained to become more proactive in making the value of gender statistics visible to Governments, the public and other stakeholders (United Nations, 2006).
Refocusing of the activities and position of gender units and gender focal points within national statistical systems. In paragraph 206 (d) of the Beijing Platform for Action, it was recommended that statistical services should designate or appoint staff to strengthen gender statistics programmes and ensure coordination, monitoring and linkage to all fields of statistical work and should prepare output that integrates statistics from the various subject areas (United Nations, 1996). Since the Fourth World Conference on Women, many countries have embarked on gender statistics programmes (United Nations, Economic and Social Council, 2010). In many instances, however, the activities of gender units and gender focal points are focused narrowly on the compilation and dissemination of sex-disaggregated data (United Nations, Economic and Social Council, 2010). In other cases, the activities are extended a step further, to the preparation of outputs that disseminate the compiled and analysed information on women and men. However, as recognized in paragraph 309 of the Beijing Platform for Action, strategies must be developed to prevent the inadvertent marginalization of gender units as opposed to the mainstreaming of the gender dimension throughout all operations (United Nations, 1996). National statistical offices can benefit from extending the functions of gender units. These extended functions would allow gender units to escape potential marginalization and to be more involved in the quality of statistics produced and their relevance for policymaking.
In particular, gender units can play a catalytic role in initiating and monitoring the process of integrating a gender perspective in national statistical systems, especially at the early stages (Hedman, Perucci and Sundström, 1996; United Nations, 2006). They can also play a crucial role in reviewing the existing production of gender statistics and in developing a gender statistics programme. Gender statisticians, meanwhile, can be more involved in the planning of data collection, including the coverage of gender issues and the use of gender-sensitive concepts and methods. They can also help to review data collection instruments and regular publications in order to ensure that a gender perspective is integrated in all statistical fields and programmes. Furthermore, gender units can play an important role in gender sensitization training as well as in training on how to avoid gender bias in data collection.
Through their contacts with national machineries for women and non-governmental organizations, gender units and gender focal points can facilitate communication between the producers and some of the end users of gender statistics (United Nations, 2006). These units can provide information to users and help them to understand the uses of existing statistics. At the same time, gender units and gender focal points can increase awareness among statisticians of the need to produce or disseminate statistics that address gender concerns and to develop gender statistics in such new areas as time use, violence against women and unpaid work.