Guide on Gender Analysis of Census Data Full Draft of 6 December 2012 Contents

Appendix 5: From Understanding the Gender Data Gap to Improving the Production and Analysis of Gender Statistics

Download 3.92 Mb.
Size3.92 Mb.
1   ...   24   25   26   27   28   29   30   31   32

Appendix 5: From Understanding the Gender Data Gap to Improving the Production and Analysis of Gender Statistics

The phrase gender data gap refers to the fact that the coverage of gender issues, including women’s lives and realities, is generally inadequate in mainstream data and statistics. This occurs to some degree everywhere but is worse in some regions (Africa and Oceania) and for some types of statistics (e.g. social services, disability, time use and unpaid work, gender-based violence, access to resources, access to decision-making processes, informal economy) (UNSD, 2005).

There are several reasons for the existence of a gender data gap including:

  • Underreporting of women in censuses and surveys in many countries. To get women out to respond, information campaigns have to explain the purpose and relevance of data collection; women have to be allowed and feel safe to speak as individuals; and enumerators, male and female, need to be trained to specifically seek the inputs of women.

  • Questionnaires are primarily developed by men/male statisticians who may bring gendered biases in the focus and phrasing of questions. There is a tendency to focus on the more public sphere of men’s lives, as opposed to the more private spheres of women’s lives (for example unpaid work is usually not reflected nor counted). Questions may use pronouns that lead to assumptions, such as using the term “he” when asking about the head of household.

  • A lack of routine disaggregation (by age and sex at the very least) which, in combination with the male norm in language and society, blurs the input and contribution of women. Speaking of “rural workers”, “smallholders” or “secondary school students” makes many people think of males rather than females and may hide important gender differences within these groups. This also leads to certain categories of women, such as indigenous women, women with disabilities, adolescent girls, and elderly women, to be particularly underrepresented. Further levels of disaggregation or multivariate analysis are needed to provide a fuller picture, and in some instances, additional questions need to be asked to identify special social groups.

  • Inadequate budgets, low human and technical capacities, lack of gender mainstreaming and inadequate concepts and methods hinders the quality of data being collected and tabulated and also the publishing of the information.

The following will examine some of the major recommendations to producing and using high-quality gender statistics as well as outline the global standards, potential problems and country examples.
Recommendation 1: Data must be relevant to Gender Equality Issues- Need for strong alliances between data users and producers

Why is this important? Gender statistics are the foundation of gender-based analysis and go well beyond sex-disaggregation. As such, producers of gender statistics are required to know more about the topics of gender, and users of gender statistics must know about the scope and limitation of the data. Collaboration between the producers and users of gender statistics will help ensure that gender statistics are more meaningful and user-friendly and include issues from multiple angles while also keeping in mind policy and planning imperatives. Raising the awareness of the institutions that produce statistics about gender issues may also help increase the demand for gender sensitive data for public policies.

Global standards: Strategic objective H.3./206.c) of the Beijing Platform for Action asks Statistical Services to “Involve centres for women's studies and research organizations in developing and testing appropriate indicators and research methodologies to strengthen gender analysis, as well as in monitoring and evaluating the implementation of the goals of the Platform for Action”. It is now widely understood that other governmental (Ministry of Fmaily/Women/Social Affairs for example) and civil society organisations (non-governmental organisations, universities, media, trade unions, etc.) should also be involved in this process.
Potential problems: A potential problem is the asymetry of power relations: Women’s machineries are often marginalised in government, suffer from a chronic lack of resources and have limited/no understanding of the availability of data or how to use data for programme/policy development. Consultations wth civil society organisations are sometimes mere window-dressing in response to growing national and international pressure to engage with them. If data users are not at the table at the time census questionnaires are designed and tabulations are planned, it will be hard to present gender statistics in a user-friendly way. Data may not be published “just in time” for crucial policy decisions if there is no interaction between policy-makers and data-producers.
Country example: Uganda. The Uganda Bureau of Statistics (UBOS) regularly promotes Public Seminars, organized in collaboration with policy makers and users of statistics (stakeholders), to present analysis and research findings using data routinely collected by the Uganda Bureau of Statistics (UBOS), including census data and gender indicators. One of the main purposes of the seminars is to educate the public about the use of statistics.
Internationally: Over the last ten years, alliances between producers and users of statistics have been strengthened through international meetings on gender statistics. For example, a Working Group on Gender Statistics was created during the fourth meeting of the Statistical Conference of the Americas of the Economic Commission for Latin America and the Caribbean (ECLAC) in 2007. This inter-institutional and inter-agency (UN Women and the Gender Affaires Division/ECLAC) initiative has led to progress in statistical activities across the region.

(Germany example, p. 141, UNECE? What is this ?) You should be included as an example of alliances between producers and users, the international meetings of gender statistics that have been carried out for more than ten years (without interruption). They have been regarded as a good practice, and that is an inter-agency effort (UN Women and Gender Affairs Division/ECLAC) and inter-institutional. The meeting is part of the activities of the working group on gender statistics of the Statistical Commission of the Americas.
Recommendation 2: An enabling institutional environment is needed for the production of gender statistics – political will, a legislative framework, sufficient funding, commitment and buy-in from senior leaders

Why is this important? Gender statistics emerge from demand, whether from politicians, specific ministries, women’s departments or the public. Political will for gender statistics can be thought of as support from political leaders that results in policy change. It generally originates with a single individual or small group, often situated somewhere within the state apparatus (best if at a high level) who become the “champion” for gender statistics, build coalitions and, as part of those coalitions, raise funds for the issue, push for its firm anchoring in the legislative framework and try to sustain the policy change once achieved. Where political will is lacking, funds are unlikely to be allocated to gender statistics and an enabling environment (legislative and administrative framework) is unlikely to be created.

Global standards: Many countries incorporate CEDAW provisions on gender equality and women’s empowerment into domestic law, in particular their civil, penal and labour codes. Countries also reform their Gender Equality Architecture, conduct Gender Audits and initiate Gender-Responsive Budgeting to monitor whether and how political translates into funding and impact. Gender analysis of sex-disaggregated data is needed to monitor progress in compliance with CEDAW and other international and national instruments at country-level.
Text Box 13: Gender-Responsive Budgeting

Gender-responsive budgeting (GRB) is government planning, programming and budgeting that contributes to the advancement of gender equality and the fulfillment of women's rights. It entails identifying and reflecting needed interventions to address gender gaps in sector and local government policies, plans and budgets. GRB also aims to analyse the gender-differentiated impact of revenue-raising policies and the allocation of domestic resources and Official Development Assistance.

 Source:; accessed 1 March 2011.
Potential problems: As UNSD pointed out, it is essential to “distinguish between a national plan on gender statistics and a law on gender statistics”, since “without a specific law, there is no guarantee that gender statistics will be included in the work plan each year” (UNSD 2009). Moreover, the laws on gender statistics must also be relatively detailed, clearly spelling out administrative and budgetary implications. Where a law on gender statistics exists, budget negotiations will be easier and the gender statistics architecture is more likely to dispose of the means it needs to fullfil its functions.

Country examples: Kyrgyz Republic: Obliges State bodies, institutions of local self-government and heads of legal entities regardless of forms of ownership to submit appropriate information on gender issues to the National Statistical Committee (Art. 30).

Philippines and Nepal:, CEDAW Monitoring Bodies have been set up that are responsible for producing and diffusing gender statistics.

Italy: A detailed law on gender statistics makes producing sex-disaggregated data from population censuses compulsory, including data on households and institutionalized persons, as well as on different household typologies. (UNECE/UNDP, 2004)

Korea: A new statistical act was released (2007), giving the Korean Statistical Office the jurisdiction to expand the topics included in data collection, in order to provide more relevant data for gender analysis (UNSD, 2009).

Spain: The 2007 statistical legislation of the Andalusia region included the provision for gender mainstreaming which created an enabling environment for the production of gender statistics, also leading to laws requiring gender budgeting and the monitoring of its effectiveness (UNSD, 2009). [Colombia, Trinidad & Tobago, Ecuador, Mexico and Peru (Costa Rica has a law project) have laws requiring the NSO to collect information on time use and non-remunerated work for unremunerated satellite accounts (set of experiences that have shared best practices in the context of the work carried out by ECLAC and UN Women and the institutions mentioned above]
Recommendation 3: Gender data must be reliable and adhere to international standards – Need for common definitions and quality standards to enhance comparability

Why is this important? Gender statistics are essential for monitoring progress on international treaties and agreements as well as providing countries with comparable data to track progress being made in promoting gender equality. In order to do so, comparison of data must be possible both over time and between countries. However, concepts such as household and household headship, marriage, economic activity, informal sector, etc. pose numerous measurement problems. The main benefit of having common definitions and quality standards is disposing of evidence that is sound and reliable and that is a solid base for informed policy decisions.

Global standards: While the Principles and Recommendations for Population and Housing Censuses, Revision 2, form the normative basis for census taking and census analysis, no such reference framework is available for gender statistics. (From UN Women: the cited examples have not referred only to censuses. It is important to cite the guide of ECLAC/UNIFEM/UNFPA, that include some recommendations about census work). Existing informal guidance documents include a number of publications by the UN Statistics Division and other international agencies, including development banks (see appendix 2).

Potential problems and benefits: In terms of definitions, the main issue is to balance between “validity” (measuring what you went out to measure, which implies using a culturally appropriate approach) and comparability. When standard definitions are too forced, categories lose their meaning on the ground and hence are no longer useful tools for policy-making. In terms of quality, the main issues are capacity and resources: To produce highest quality gender statistics, National Statistics Offices need the appropriate human and financial resources to pilot test and analyse, train enumerators, invest in data editing, etc.

Regional example: The Gender Statistics programme of the Secretariat of the Pacific Communities has created a website (PRISM) collating statistics from 22 countries and territories in the Pacific on common indicators used for monitoring gender equality (see Other regional observatories: Latin America and the Caribbean Gender Equality Observatory ( with the support of UN Women.
Recommendation 4: There is severe underreporting on gender issues – Need for capacity-building for NSOs

Why is this important? Underreporting on gender exists both in terms of who gets enumerated/surveyed and in terms of what gets published. Therefore, the institutionalization of gender mainstreaming is crucial to engendering all statistical operations. On an organizational level, this means training staff on gender statistics and gender analysis as well as establishing Gender Units in NSOs. On an operational level, it implies ensuring gender responsiveness in questionnaire elaboration, field operations and their supervision, as well as in data management, analysis, publication and dissemination.

Global standards: Data is supposed to adequately reflect the situation of all citizens (male and female, old and young, indigenous populations and persons with or without disabilities, etc). Being inclusive of every citizen and asking truly relevant questions is therefore imperative in order to achieve high data quality. Collecting data on women and girls and on gender-specific issues is one step into the direction of achieving a fuller, more valid picture of the situation in a country.

Potential problems:In many countries, high turnover of staff (sometimes due to unattractive working conditions and remuneration) makes capacity-development a never-ending challenge. Specialized gender teams, inter-sectoral and inter-agency gender working groups, and gender departments established in the National Statistics Offices, can ensure that the participation of gender statistics experts in other relevant official statistical productions is institutionalized. Capacity-development is also needed for effective data dissemination strategies: “The analytical capacity of national statistical systems should be strengthened to ensure that available data are used productively and the findings communicated more successfully to the appropriate audiences, especially to the policy makers and the media […]” (UNDESA, 2006: 29-30).

Country example: Nepal: Regional training workshops on gender sensitization were conducted in joint partnership with international agencies and included participants of senior and mid-level officers of the Central Bureau of Statistics.

Challenge 5: Micro-data on gender issues is dispersed - Need to unify what is available, upstream small-scale data

Why is this important? At the national and international levels, there is still a large gap with regards to the availability of data on social services, disability, time use and unpaid work, gender-based violence, access to resources, and the informal economy, to name just a few important gender concerns. Such data are needed by in-country NGOs, line ministries and local governments to monitor projects, prepare advocacy materials and negotiate budgets. If National Statistics Offices could act as a clearing house for disparate small-scale data at the sub-national level, or effectively cross-analyse data from various sources (vital statistics, census, surveys, etc.), much could be gained for understanding gender issues “on the ground”.

Global standards: The UN’s Inter-Agency and Expert Group Meeting on the Development of Gender Statistics emphasized the need of two specific issues: 1) “the review of existing national data collections to identify and develop inventories of sex-disaggregated statistics and gender related indicators”; and the need to 2) “the re-coding, re-tabulation and re-analysis of micro-data from surveys or censuses, from a gender perspective” (idem: 08). Similarly, “one of the strategies recommended by FAO for filling the gender data gap is to promote the coordination, integration and re-tabulation of agricultural data by sex and age at the sub-national level (FAO 1999:23)”

Potential problems: A high level of technical expertise is needed for cross-analysing data collected on small samples of different population groups at different points in time. Despite huge efforts in combining sources, data quality may be too low for meaningful analysis.

Country examples:


Recommendation 6: Information is available but not used - Need for better communication and dissemination of gender data

Why is this important? It is not enough for gender data to simply be available. To make a difference in the lives of women and men, girls and boys, these data need to be disseminated – and ideally used – by policy-makers, opinion leaders and the general public. A variety of formats are required to achieve this. Along with reports and press releases, dictionaries and metadata are important tools for ensuring that the underlying concepts are actually comprehended and that the results are well interpreted. Online resources and services, like query databases and downloadable files, are now a large part of the public’s information toolkit. Documents, tables, maps and images available on the internet facilitate the access and visualization of data.

Global standards: Several issues are at play here: The public’s right to information is key, as is the moral imperative to feed knowledge back into the public domain. Policy makers can be reminded of norms such as transparency and accountability to invite them to cement the knowledge-base on which their policy-decisions are based.

Potential problems: If NSOs are increasingly offering a broader range of formats for dissemination, users of gender statistics also need to be more proactive in accessing and using the existing data. Benefits include a better-informed public, and – in the best case – better informed public policies. A collateral benefits is heightened visibility for the NSO (which can have budgetary repercussions).

Country examples:


Appendix 6: How to Apply this Guide in a Country Context

While there are many ways in which this guide can be used, the following provides four key steps for carrying out a gender analysis project at the country level. It emphasizes collaboration between the producers and users of gender statistics by involving NSO and gender experts from the government, academia and civil society. As such, these steps will help ensure that gender statistics are more meaningful, user-friendly and address the key gender issues relevant to that country.

Step One: Selection of Gender Issues:
Participants: NSO, Gender experts from government and civil society, including research institutions

Format: Workshop

Documentation: 10 Key Question Tool, Census Questionnaire, this manual (table of contents: chapter 3)

Purpose: To identify the key gender issues that can be analysed with the census data obtained in country X.

Roles: Gender Experts provide an evidence-based overview of the key gender issues in the country, ideally using the 10 Key Question Tool below and considering the 10 gender issues. Statisticians explain what can and cannot be measured with census data on the basis of the country’s census questionnaire.

Expected Outcome: Consensus on what statisticians should compute
Box : The 10 Key Questions Tool

1. Who does what? [activities]

2. How? With what? [access to resources]

3. Who owns what? [ownership of assets]

4. Who is responsible for what? [obligations]

5. Who is entitled to what? [claims, rights]

6. Who controls what? [income, spending]

7. Who decides what? [power]

8. Who gets what? [distribution]

9. Who gains and who loses? [redistribution]

10. Why - What is the basis for the situation? [rules, norms,


(Questions 1-9 can be combined with the additional question, "And With Whom?' in order to capture the social relations involved)
Step Two: Census Analysis – Preparation of Tabulations and Computation of Sex-Disaggregated Indicators
Participants: NSO, Gender experts from research institutions, external consultants

Format: Desktop study/in-depth statistical analysis

Documentation: “Principles and Recommendations”, this manual (sections 4 and 5 of each chapter in Part II)

Purpose: To provide the raw data, tabulations and indicators needed for answering the key gender questions identified in Step One.

Roles: Statisticians and researches perform high quality data analysis

Expected Outcome: Tabulations and indicators are available and of high quality
Step Three: Interpretation of Data, Suggesting further Analyses
Participants: NSO, Gender experts from government and civil society, including research institutions

Format: Workshop

Documentation: Tabulations and indicators produced by NSO, this manual (sections 2 and 6)

Purpose: To make sense of the data and suggest further analyses going into more depth with some key findings

Roles: Statisticians walk participants through the analyses carried out, outline problems encountered and summarise the gender differences identified; Gender Experts discuss what may be underlying the gender differences documented

Expected Outcome: Consensus on additional variables that need to be taken into consideration (and technically can be) in order to shed light upon the findings
Step Four: Advocacy Material is devised
Participants: Gender experts from government and civil society

Format: Workshop

Documentation: Key national policies, this manual (section 8)

Purpose: To identify how the indicators, tabulations and results of multivariate analysis can be used to inform and advocate for current and future national gender equality policies and initiatives, or for reporting purposes

Roles: Experts provide an overview of the key gender policies and initiatives currently on-going and planned in-country and select critical data for evidence-based advocacy

Expected Outcome: An advocacy plan with clearly defined roles/responsibilities/timeline (cite key references)

1; last accessed on 3 December 2012. The data indicates that the majority of countries is succeeding in their census planning and taking. However, as conducting a census is a complex and costly process that requires great efforts in capacity building, some countries and regions have been forced to delay or even cancel their censuses. Some of the challenges that countries are facing include: administrative organization, funding constraints, post-conflict situations, humanitarian crisis, natural hazards, etc.

2 If the cost is too large for the NSO, it may opt for distributing the information through the IPUMS programme of the University of Minnesota, which designs user samples for release to the public, guided by the specifications provided by NSOs.

3 In addition, economists often use the term “equity” in a sense that is completely distinct from the one explained in this section, namely to refer to debt-free assets in the form of real estate, bonds or particularly stocks.

4 This discussion parallels the one of “equality of opportunities” versus “equality of outcomes” in regard to the role of the school system. While some consider schools as the great equalizers of opportunities between children of different social backgrounds (or, in this case, different sexes), others (e.g. Jencks, 1972) have demonstrated that, even in societies like the United States, which places a lot of emphasis on the principle of equal opportunity, only a relatively small proportion of the inequality of outcomes can be explained in terms of differences in access to education.

5 Note that in French and Spanish, the usage of the terms is the opposite: fécondité or fecundidad for actual reproductive outcomes and fertilité or fertilidad for biological capacity.

6 The update is being undertaken to incorporate new developments, take account of new trends such as those brought about by the advent of AIDS, and adapt the older techniques to the possibilities created by more modern computational tools such as EXCEL. For more information, see

7 Nevertheless, omission and displacement of births in DHS data are not trivial, and of course, as in any survey, one has to account for sampling errors, which limits the possibility of using DHS data for small sub-groups.

8 For example, Argentina, Azerbaijan, Bahamas, Costa Rica, Kazakhstan, Mexico, Palau, Peru, the Seychelles and Thailand ask the traditional fertility and child survival questions, but do not disaggregate them by sex.

9 For a common definition of live birth, see: natmethods.htm.

10 Brazil, Botswana, Burundi, Cayman Islands, China, Djibouti, Dominican Republic, Ecuador, El Salvador, Fiji, Iran, Liberia, Malawi, Maldives, Occupied Palestinian Territories, Republic of Congo, St. Lucia, Samoa, Sudan, Swaziland, Tokelau, Trinidad and Tobago.

11 Note, however, that the adolescent birth rates used for monitoring MDG 5.B are estimated independently and not based on the UN population projections.

12 The same problem affects the Own Children Method for fertility estimation.

13 In the 2010 census round, however, two censuses have asked a question how many additional children women intended to have, namely Republic of Korea (2005) and Kazakhstan (2009).

14 definition adapted from WHO website:

15 It is quite easy to show a strong correlation between maternal mortality and selected gender indicators but this does not imply a causal relationship, as both are correlated with the overall level of development.

16 Note that in India the sex ratio is computed the other way around, as the number of girls over the number of boys.

17 Where “date of birth of last live-born child born” is not disaggregated by sex, one needs to look at the age and sex of the youngest child in the household and – if under 1 year old – verify if its age/birthday is compatible with the declared date of last birth.

18 Some anomalies in sex ratios at birth can be explained in biological terms. For example, a study by the Arctic Monitoring Assessment Program in 2007 found abnormally low sex ratios, in the order of 50, in some arctic communities in Russia, Greenland and Canada, which it attributed to high levels of endocrine disruptors in the blood of inhabitants, particularly PCBs and DDT. Other studies (e.g. Rocheleau et al., 2011), however, have contested the effect of PCBs on human sex ratios at birth. There is also some discussion among geneticists as to whether sex ratios vary naturally according to race, maternal and paternal age and birth order (e.g. Erickson, 1976; Imaizumi and Murata, 1979; Ruder, 1985; Chahnazarian, 1988). Historical data from Europe suggest considerable heterogeneity between families, with boys predominating in some and girls in others, in proportions that differ from what one would expect if the process were purely random (Garenne, 2008 b). In the case of Africa, Garenne (2008 a) found that sex ratios declined with maternal age and birth order. Due to the fact that he used DHS data, no information on paternal age was available. He concluded that these findings are consistent with James’s (1989, 1996) theories about the biological factors of the sex ratio, in particular, the effect of concentrations of sex hormones (e.g. progesterone, gonadotropin, estrogen, testosterone). Higher levels of gonadotropin and progesterone were found to be associated with more female births (lower sex ratios). Conversely, higher concentrations of male hormones (e.g. testosterone) seem to favour high sex ratios. The African data do not seem to suggest any deliberate sex selection. Oster (2005) has argued, based on existing medical literature and analysis of cross country data and vaccination programmes, that parents who are carriers of hepatitis B have a higher offspring sex ratio (more boys) than non-carrier parents. Since China and some other countries have high hepatitis B carrier rates, she suggested that hepatitis B could explain up to 50 per cent of Asia’s “missing women”. However, Lin and Luoh (2008), using data from a large cohort of births in Taiwan, found only a very small effect of maternal hepatitis carrier status on offspring sex ratio, a conclusion which was later endorsed by Oster as well (Oster et al., 2008).

19 When sex ratios began to rise in Armenia in the 1990s, for example, at first the tendency was to attribute this to the aftermath of conflict in the region. It was not until further analysis established that the imbalance was limited to second and third birth orders that the sex selection process was recognized for what it was.

20 This is based on Article 5a) of the original CEDAW text, which states that “States parties shall take all appropriate measures to modify the social and cultural patterns of conduct of men and women, with a view to achieving the elimination of prejudices and customary and all other practices which are based on the idea of the inferiority or the superiority of either of the sexes or on stereotyped roles for men and women.” Similarly, General Comment 14 of the CEDAW states that “States parties' reports also disclose that polygamy is practiced in a number of countries. Polygamous marriage contravenes a woman’s right to equality with men, and can have such serious emotional and financial consequences for her and her dependants that such marriages ought to be discouraged and prohibited.”

21 This difference in enumeration methods can have implications from the viewpoint of a gender analysis. For example, if the census coincides with a period of the year in which many of the men are temporarily absent from their homes because of seasonal activities (temporary harvesting labour) and if the census is conducted according to the de facto criterion, more women than usual will be classified as living without their husbands. If the census is a de jure census, the missing men will be counted in their households of usual residence, but they may not be considered heads of households, even if they normally act as such.

22 PPP refers to purchasing price parity, which measures the relative purchasing power of different countries’ currencies over the same types of goods and services, adjusting for inflation. PPP helps provide an accurate comparison of standards of living across countries (World Bank, 2011).

23 The change from USD 1 to USD 1.25 was introduceed to correct for inflation of the US dollar.

24 In a hot deck imputation information from other respondents with similar characteristics is used to make imputed that are best suited for the missing information. See: United Nations, (2008 a) Principles and Recommendations for Population and Housing Censuses Revision 2: 70.

25,,contentMDK: 20242879~menuPK:435055~pagePK:148956~piPK:216618~theSitePK:430367~isCURL:Y~isCURL: Y,00.html.


27,,contentMDK: 22405907 ~menuPK:6626650~pagePK:148956~piPK:216618~theSitePK:430367,00.html.


29 Washington Group on Disability Statistics (WG) (s.d.), The Measurement of DisabilityRecommendations for the 2010 Round of Censuses. p.1.

30 Based on UNSD/CAPMAS (2000). Gender and Development: An Information Kit I. Cairo, CAPMAS.

31 The World Plan of Action of the First Conference also stated that regional action should include regional standing committees of experts from countries in order to “give leadership in the methods of reporting on the situation of women and in the development of indicators for assessing the progress made towards the goals of this Plan in conjunction with regional statistical bodies and international efforts to this end” (World Conference of the International Women’s Year, World Plan of Action, Paragraph 207, 1975).

32 The Beijing Platform for Action made reference to the need to develop and strengthen statistical systems in several issues, such as labor and economic activity (including female contribution in the unremunerated and domestic sectors), health of girls and women of all ages, incidence of violence (including domestic violence, sexual harassment and other different forms of violence against women and girls), and sharing of power and decision-making.


34 Activities no longer considered as ‘household activities’ include: production of agricultural produce, gathering of fruits etc. and their storage; processing of primary products (produced or bought) and the collection of water; other processing activities, sold or not, like weaving, dress making and furniture making. (Tempelman, 1999).


Download 3.92 Mb.

Share with your friends:
1   ...   24   25   26   27   28   29   30   31   32

The database is protected by copyright © 2022
send message

    Main page