About this tool
- There are different frameworks for doing a gender analysis
- Different gender analysis frameworks serve different purposes. For example, for the purpose of measurement the WIAD tools are useful while you are interested in the design and planning of a project the Harvard framework is recommended.
This tool provides guidelines to improve existing data collection processes by making them gender sensitive. This tool is about gender-sensitive data collection and gender analysis.
Gender-sensitive data collection allows for:
- Understanding the farmer population (supplier base) and main differences
- What categories of farmers are being reached by current services (and which are currently excluded)
- Informing supply chain interventions; ensuring suppliers (females/males) are targeted with services that work for them
- Identify needs for tailored services and interventions
- How to improve outreach strategies
- Participatory monitoring and evaluation to track progress over time for different types of suppliers
- Communicating results about improvements in farmers’ performance (f/m) and well-being
Gender-analysis allows for:
- Understanding gender relations: how division of labor, access to and control over resources, decision-making processes within the household and norms and values are interlinked and interlocked.
- Inform the design, planning and M&E of robust and inclusive interventions and programmes that contribute to women’s economic empowerment and sustainable livelihoods.
Association for Development of Education in Africa. (2006). A Toolkit for Mainstreaming Gender in Higher Education in Africa. Workgroup on Higher Education, Ghana.
CARE Facilitation tools. Accessible at http://www.care.org/sites/default/files/documents/FFBS_1_Facilitation_Tools.pdf
CARE Monitoring, Evaluation and Learning tools. Accessible at http://www.care.org/sites/default/files/documents/FFBS_6_Monitoring_Tools.pdf
CARE’s Participatory Performance Tracking Tool. A Step by Step Guide for Use Completed January 2015. Accessible at http://www.care.org/sites/default/files/documents/PPT%20Step-by-step%20Guide%20Final.pdf
Anouka van Eerdewijk & Katrine Danielsen (2015) Gender Matters in Farm Power February 2015. Accessible at http://www.kit.nl/gender/wp-content/uploads/publications/551bcea41f1f2_Gender%20Matters%20in%20Farm%20Power%20(final%20150227%20AE%20KD).pdf
Guion, Lisa A., Diehl, David C. and McDonald, Debra (2013). Triangulation: Establishing the Validity of Qualitative Studies. University of Florida, IFAS Extension.
IFPRI (2011) Engendering Agricultural Research, Development, and Extension. Priority setting, research & development, extension, adoption, evaluation. Authors: Ruth Meinzen-Dick, Agnes Quisumbing, Julia Behrman, Patricia Biermayr-Jenzano, Vicki Wilde, Marco Noordeloos, Catherine Ragasa, Nienke Beintema.
International Labour Organization (ILO) SEAPAT South-East Asia and the Pacific Multidisciplinary Advisory Team. The Harvard Framework. Accessible at http://www.ilo.org/public/english/region/asro/mdtmanila/training/unit1/harvrdfw.htm
Ranjani K. Murthy (2015) Toolkit on Gender-sensitive Participatory Evaluation Methods. Technical Report · January 2015. Accessible at https://www.researchgate.net/publication/283509621_Toolkit_on_Gender-sensitive_Participatory_Evaluation_Methods
PADev Participatory Assessment of Development Guidebook. Accessible at http://padev.nl/guidebook.htm
WEAI Resource Center. Accessible at https://www.ifpri.org/topic/weai-resource-center
Mainstreaming gender in data collection and analysis enriches understanding of the supplier base, enabling CLP’s Matching Grantees Partners (MGs) to act strategically and to monitor effectiveness. MGs can also meet expectations of customers, donors and public commitments with regards to tackling gender inequality.
Advantages of gender sensitive data collection:
- Measuring development impact
Including gender indicators is crucial to measuring development: women’s empowerment has a major impact on family well-being, nutrition, education, and health.
- Understanding your supplier base
Gender sensitive data collection will improve the understanding on current access to, and benefits from, specific sustainability programmes, sourcing strategies and services for both men and women.
- Recognizing women’s potential
Women contribute significantly to cocoa production, but often lack land privileges or other rights that allow them to benefit from their contributions. Gender sensitive data collection makes the contribution of women visible and generates insights on how women can benefit from and contribute more to cocoa production.
- Avoiding gender-blind interventions
If you collect data as a basis for designing interventions, ignoring gender will result in missed opportunities; interventions can be counterproductive and reinforce existing gender inequalities.
- Developing gender specific interventions
A better understanding of gender roles and gender issues in the cocoa sector will inform interventions that reduce gender inequalities and contribute to women’s empowerment.
- Monitoring women’s performance and measuring impact
Gender sensitive data collection makes it possible to track changes in performance between men or women.
- Communication of gender sensitive policies
Gender sensitive data enables companies to learn more about the status of women in their supplier base, and what can be done to address gender inequalities. Having this type of data facilitates communication on gender issues to customers, consumers and suppliers. For example, in ‘Women and Chocolate: An Action Roadmap’ Oxfam as part of their ‘Behind the Brands’ campaign demands that Mars, Mondelez and Nestlé deliver gender data to develop gender action plans to address gender inequities in their cocoa supply chains.
What is gender sensitive data collection?
Gender sensitive data collection starts by disaggregating the data by gender. This is straightforward but often overlooked. Gender-disaggregated data collection means differentiating between sex in data collection: indicating whether each respondent is a man or woman. This will provide information on the percentage of males and females in supplier bases, enabling observation and analysis of similarities and/or differences between them, and measurement of changes over time for both men and women.
Although a first step, disaggregating data only by sex is not enough. It does not do justice to other variances among suppliers within the database (e.g. different women face different challenges and opportunities). In order to fully understand gender dynamics and their implications, gender relations must be understood within in their context (see box 4.1). Gender relations refer to the different socially-constructed roles, behaviours, responsibilities and attributes a society considers appropriate for men and women. An example could be that women are responsible for cooking and fetching water. Doing a gender analysis helps to understand gender relations.
Steps to use this tool
Step 1: Define learning objectives
Step 2: Select data collection methods
Step 3: Gender analysis
Step 4: Participatory Monitoring and Evaluation
Step 3: Analyse your data
- There are different frameworks for doing a gender analysis
- Different gender analysis frameworks serve different purposes. For example, for the purpose of measurement the WIAD tools are useful while you are interested in the design and planning of a project the Harvard framework is recommended.
Being sensitive in gender-data collection is not the same as doing a gender-analysis. A gender analysis is conducted to understand the role that gender relations play in for example the adoption of a new technology, or the impact of a training or larger programme. A gender analysis can be used for the purpose of monitoring and evaluation, but also to design interventions, programmes and policies for women’s empowerment.
In doing a gender analysis often three main questions are asked:
1. Who does what?
2. Who has access to and control over what, and who benefits?
3. Who makes the decisions?
‘Who does what?’ refers to the division of labor. This involves both the labor on a farm or in other income generating activities, as well as tasks in the household.
‘Who has access to and control over what’ refers to access to key resources needed for the different activities men and women are involved in. Resources can involve access to productive resources, such as land, access to finance or for example group membership. Because access to resources does not automatically result in benefits, it is important to also ask about control over the resources. For example, whether or not a woman in a cocoa growing community can benefit from access to a loan, depends on to what extent she can also make the decision about what to do with the money borrowed.
‘Who makes the decisions’ is about decision-making processes in the household and gender relations. These involve decisions about roles, the use of resources, or, for example, the freedom of movement. For example, in cocoa growing communities in West Africa it is common that women need to ask for permission from their husbands if they want to travel to a neighboring community.
There are different frameworks used for gender analysis. These frameworks have often different purposes. An example is the Gender Roles Framework or Gender Analysis Framework, developed by the Harvard Institute for International development in collaboration with the WID office of USAID. This Gender Analysis Framework is mainly used for the design and planning of more efficient projects and improve overall productivity. The Harvard framework gives practical guidelines for collecting data at the community and household level and gives. It gives key questions to ask at each stage of the project cycle: identification, design, implementation, and evaluation.
For the purpose of technology adoption another framework for gender-analysis has been developed by KIT. In this analytical framework the three dimensions mentioned above are combined with a fourth dimension: values and norms. Gender norms and values are a set of social rules and assumptions about what men and women should do, how and with what resources, and the status of individuals and their relative value in society. Gender values and norms are constantly changing. In this framework, understanding gender dynamics is not only about understanding each of the dimensions separately, but also about looking at how these four dimensions are interlinked and interlocked.
Figure 5.1 Analytical framework used to understand technology adoption
Source: adopted from Eerdewijk and Danielsen 2016.
For example, looking at the interplay between norms and values and the division of labor helps to understand to what extent these roles reflect ‘social rules’ and if there are rewards and punishments for compliance and non-compliance.
If the purpose of doing a gender analysis is measurement a very useful and practical tool is the “Women’s Empowerment in Agriculture Index” (WEAI). WEAI was launched in 2012 by IFPRI, Oxford Poverty and Human Development Initiative (OPHI), and USAID’s Feed the Future. WEAI is said to be the first comprehensive and standardized measure to directly capture women’s empowerment and inclusion levels in the agricultural sector.
WEAI is a survey based index to measure the role and extent of women’s engagement in the agriculture sector in five domains:
- Decisions about agricultural production
- Access to and decision-making power over productive resources,
- Control over use of income,
- Leadership in the community, and
- Time use.
WEAI training materials include examples of surveys that have been piloted to collect this type of data that enable measurement. WEAI has been used extensively since 2012 by a variety of organizations to assess the state of empowerment and gender parity in agriculture, to identify key areas in which empowerment needs to be strengthened, and to track progress over time.
Other useful frameworks for gender analysis are:
In their analysis they put emphasis on women’s empowerment, which is defined as “the sum total of changes needed for a woman to realize her full human rights – the interplay of changes in:
- Agency: her own aspirations and capabilities,
- Structure: the environment that surrounds and conditions her choices,
- Relations: the power relations through which she negotiates her path.
CARE has developed a global women’s empowerment framework that provides 23 key dimensions of social change which have been shown to be widely relevant to women’s empowerment across many studies and contexts.
CARE has also developed practical gender tools that can be useful to collect data in the field and to raise awareness in communities, groups and households. For example:
- A tool to explore who in the household has authority to make important decisions, and how decision-making could be more equal (page 106).
- A tool to identity the differences between rules of behavior for men and for women; To understand how these gender rules can negatively affect the lives of both women and men (page 109)
- A tool to illustrate the existence of power in relationships and its impact on individuals and relationships (page 113).
Step 2: Selection of data collection methods
- Data collection methods that have not been designed in a gender-sensitive way are unlikely to effectively collect evidence around gender issues.
- Gender-sensitive data collection requires mix methods, both quantitative and qualitative
Step 2 is about how to collect data in a gender-sensitive way. Gender-sensitive data-collection requires more than collecting data disaggregated by sex.
How to ensure that the data is collected in a gender-sensitive way? There are a number of choices that support gender-sensitive data collection. For example:
- Collect information from both men and women. This does not necessarily require interviewing men and women in the same household. Studies that fail to include male and female respondents will be subject to biases.
- Ensure that all data collection methods are context specific. This requires that questions must be adapted to the context.
- The research methods should be easy to understand for both male and female respondents. For example, using a questionnaire in a context where the majority of women are illiterate might result in poor responses or a lack of female respondents. It can help to work with visuals and with a translator/researcher who can communicate in local languages.
- Taking into account timing and location of interviews with respect to availability and possible constraints to women’s participation. Conducting research during hours when women are for example normally are busy collecting firewood skews responses and biases collected data. It helps to agree interview timing beforehand.
- Work with skilled and mixed gender research teams able to conduct gender sensitive research and understand gender issues and social dynamics within the research context. This is particularly relevant when working in gender-segregated societies, in which men and women do not occupy the same public spaces, so men will usually be cut off from women and vice versa.
- Consider working with a gender expert early in the process to define (or refine) the research questions and methodology.
- If the data collection involves sensitive topics, such as domestic violence or ownership over resources, make sure that the data collection takes place in an environment that is considered to be safe, for example a ‘women-only group’.
- Comparing male and female headed households is not the same as doing a gender analysis. Differences between these diverse household types cannot necessarily be attributed to the sex of the household head. Therefore, make sure you collect additional basic demographic data, such as age, education, number of children, etc.
- Make sure you also collect qualitative data. Qualitative data will enable you to address the ‘why’ and ‘how’ questions. This type of data collection is necessary if you want to understand why gender inequalities exist.
Gender-sensitive data collection can involve more resources that you anticipated upon. For example because you employ both a male and female researcher, instead of only one researcher. Or you did not budget for qualitative data collection. Therefore in doing gender-sensitive research, make sure budget for additional costs.
Data selection methods
There are different types of data collection used by researchers or evaluators. Generally a distinction is made between three types of data: quantitative, qualitative and a combination of both (mixed methods).
Quantitative data can, especially for large data sets, be a powerful instrument to present evidence and carry out measurements. The general idea of quantitative research is to gather information that can be inferred (or generalized) to large populations.
Typical examples of quantitative data collection are surveys or questionnaires, which can be used to compare results over time and measure change. In order for surveys to enable gender-sensitive data collection it has to be ensured that the data is collected in a sex-disaggregated way, and that also other additional demographic data is being collected.
What has turned out to be challenging sometimes in survey’s among cocoa producers is to get good representation of female respondents. A lack of women in your sample can have different reasons, for example there can be a bias towards respondents that are head of the households (who are predominantly male), or the timing of the survey or the location where the survey is done constrain women to participate. It also happens that women don’t see themselves as cocoa farmers, and refer to their husbands when it comes to being surveyed, this despite women’s known contributions to cocoa production. Table 5.1 shows how it can be easy to unintentionally exclude women from participating in a survey.
It can also be that women are intentionally excluded, for example because researchers have experienced that women ‘know less about cocoa’, which affect the quality of the survey. Before you make the choice to leave out women, as a researcher you should reflect on your research method and questions, and see how questions can be rephrased or visualized so that they are easier to understand, so women can participate more equally.
Surveys have their limitations, particularly when it comes to collecting gender-sensitive data. The number of questions that can be posed is limited (a questionnaire should not take more than an hour to complete). Most importantly, this type of data does not provide any explanation of the results. Therefore if only quantitative data is collected, understanding ‘why’ gender inequalities exist and ‘how’ these can be addressed is difficult.
So, the collection of only quantitative is not enough to understand why differences between male and female farmers occur. Understanding ‘why’ therefore requires additional qualitative data collection, such as focus group discussions and semi-structured interviews.
Qualitative data collection is better able to address the ‘why’ question and incorporate insights from discussions and dialogues. There are different tools for qualitative data collection, including semi-structured interviews, focus-group discussions and participant observations. Qualitative data should be able to tell the story behind the facts and is also useful in communicating stories to the outside world.
Focus group discussions (FGDs) are a great way to hear from program or project participants and their household members about their experiences and reflections. Being gender-sensitive in the way you organize a FGD requires a conscious decision about the group’s composition. You can choose to conduct FGDs with women and men separately or in mixed groups.
Group composition depends on a number of factors including:
- Society. In a sex-disaggregated society FGD groups should be women- or men-only.
- Objective. If the objective is to facilitate discussion and to sensitize participants on gender relations, then mixed gender groups are ideal. If the aims is to create space for more honest answers, women-only (or men-only) groups work better.
When the participants are selected, the next step is to inform participants of the purpose of the exercise and approximately how long it will take. During the FGD it is important to be aware of the group dynamics, and avoid that some participants dominate the discussion at the cost of others. Table 5.1 highlights that also in women-only groups, some women can dominate over others.
 Based on a toolkit for Mainstreaming Gender in Higher Education in Africa (2006). Association for Development of Education in Africa. Workgroup on Higher Education, Ghana.
There are different tools online available that give guidance for the organization of a FGD, including how to deal with group dynamics.
For example CARE has developed some facilitation tools that help the group members to practice viewing other persons’ standpoints before making a judgement. See page 28 (How Many Eyes? Valuing other Viewpoints.)
Examples of outlines for FDGs can be found in CARE’s toolkit on gender-sensitive participatory evaluation methods (page 162). In the gender and cocoa livelihood toolbox there are also practical examples of qualitative data collection, for example on how to do social mapping.
Practical examples of FGD exercises can be found in the PADev toolbox (PADev stands for Participatory Assessment of Development), for example on how to collect data to assess or evaluate a certain programme in a participatory way.
Qualitative data collection has certain limitations: the quality of such data collection depends heavily on the researcher, and there is a risk that results are subjective and biased (also from a gender perspective). Another limitation is that qualitative data collection takes generally more time, and smaller samples are being used. This can make it difficult to generalize outcomes and to build strong evidence.
Mixed methods involves the use of multiple qualitative and/or quantitative methods (triangulation). For example, results from surveys, focus groups, and interviews could be compared to see if similar results are found. Different research methods can also be used in a complementary way: a focus group discussion may help to prioritize outcomes of a survey, or a statistical analysis can back up anecdotal stories collected from the field. Triangulation is a technique that facilitates validation of data through cross verification from two or more ways of acquiring data.
If both quantitative and qualitative data collection are used, it is important to think about how and when to best conduct qualitative research. For example, if the intention is to conduct a baseline assessment, it makes sense to collect qualitative data first to understand gender issues in the research context, e.g. using key informant interviews or FGD. A short qualitative exercise can also help in determining a control group. During a baseline assessment, qualitative data collection complements survey data, particularly on the ‘why’ questions and to confirm quantitative findings.
We have seen that each research method has its challenges. The gender sensitivity of each method depends on a number of factors. The next table explains these factors for the different research methods and provides some practical recommendations that enhance the gender-sensitivity
Table 1: Research methods and the factors enhancing or diminishing their gender sensitivity
|Method||Factors possibly diminishing gender sensitivity||Comments on Gender sensitivity|
|Field surveys (Quantitative) Field surveys are done by researchers who meet the people they want to survey, and ask them the questions posed in the survey||The race and gender of the researcher influences the responses of participants, especially when the topic of the survey deals with gender or race issues. In sex-disaggregated societies, interviews have to be conducted by researchers of the same sex.|
|They depend on the researcher dynamics and the context.|
|In the sampling, there is usually a bias towards interviewing respondents that are more prominent or more easily accessible (e.g. living nearby roads, being landowner and therefore seen as the farmer etc.). This leads to exclusion of more marginalized groups, of which many are women.|
|Participant Observation (Qualitative) researcher (participant observer) studies the life of a group by sharing its activities||Depends on skills and accuracy of observer, i.e. do you use a gender lens? Characteristics or gender of the researcher can influence the observation|
|Interviews (Qualitative) Interviews are less structured than surveys. The interviewer has the opportunity to probe or ask follow up questions||Like with surveys, the gender or other characteristics of the interviewers might influence participant responses||There is room for gender bias unless conscious efforts are made to mitigate these biases.|
|Interviews are time-consuming and costly. Gender-sensitivity can require additional resources.|
|Focus Group Discussions (Qualitative) A group of people are asked about their perceptions, opinions, beliefs, and attitudes towards a certain subject. Questions are asked in an interactive group setting where participants are free to talk with other group members||In FGD, dominance problems can occur. For example confident people may tend to speak up more or less empowered people may not been listened to. Men in focus groups speak more readily than women.||Sometimes working with women-only groups can mitigate dominance. Also in these groups, the facilitator has to be sensitive about dominance of certain women.|
|The composition of the group is important to establish a safe environment where people feel free to speak|
|Triangulation (Quantitative & Qualitative) Triangulation is a technique that facilitates validation of data through cross verification from two or more ways of acquiring data||A combination of research methods helps to mitigate some of the mentioned sensitivities and/or biases||Potentially high on gender sensitivity.|
Step 1: Defining learning objectives
- Before starting to collect (additional) data (or do additional analyses) it is important to understand what it is that you are interested to know more about, and why this is important for your company. How do you anticipate that gender disaggregated data collection and analysis will contribute to the objectives of your company and your role to contribute to the sustainability of the cocoa sector?
- Identify your research team’s capacity to collect and analyse gender sensitive data.
Without a good rationale behind data collection, it is unlikely that the additional efforts will also translate in improved strategies and interventions. At the same time, without a clear objective, it is difficult to create commitment in the organization for the need to collect disaggregated data and to build capacity to analyze data in a gender disaggregated way. It is important to think beforehand how data will be used, and for which audience.
Defining a learning objective is about determining what issues need to be researched what is it that you are interested to know more about; and why this is important. Objectives can be based on:
- A set of assumptions and/or existing data that needs to be validated or contextualised. For example, women in West Africa contribute to quality cocoa production, so how about women in East Africa?
- A knowledge gap, and ideas about how filling this gap through data collection will contribute to the objectives. For example, what is the contribution of female cocoa farmers to quality production? What kind of reward structure responds to women’s needs?
- Planning certain activities that aim to involve women e.g. what do women do besides cocoa production? When and where do these activities take place?
- Gender sensitive monitoring and evaluation (M&E) which requires a gender sensitive baseline M&E framework. For example, what explains different impacts of a programme [OF1] on male and female cocoa farmers?
Box 5 What plan did Armajaro have to reach their objectives?
Armajaro wanted to learn about:
- What are easy entry-points for improving gender sensitive data collection within the existing data collection system? What additional data do we need to collect?
- What is essential information on gender that we need to know and why?
- What can we learn around the business case for addressing gender inequalities?
- How can we monitor our work in relation to gender?
Armajaro wanted to build on their existing data collection system: Geotraceability
When defining learning objectives, it is important to look at what you would like to verify, test or know. For example it is important to know why you want to know more about the farmers you work with. Do you want to increase productivity, or do you want to ensure both women and men are reached out to? Or both?
This tool zooms in on two main goals of data collection:
- To collect knowledge or to verify and/or contextualize existing knowledge understand a particular situation
- To establish an evidence base for particular types of interventions
There is a lot known already about existing gender issues in cocoa production. One way of using data collection is to test/verify some of these existing insights. For example:
-female cocoa farmers own smaller plots of cocoa land and produce smaller volumes of cocoa;
-female farmers are better adopters of new technologies and are more trustworthy.
Box 6: What do we already know about context specific gender relations in agriculture relevant for West-African cocoa producing countries
Context specific gender relations
- In terms of roles and responsibilities, women often have greater responsibility for family food production and processing (subsistence farming), whereas men have greater involvement in market- oriented production (cash crops).
- Even when men and women work together on cash crop farming, men are usually in charge of the marketing and benefit from the related income.
- Where women are engaged in markets, their responsibility for cooking food and serving it to their family is an important factor influencing their preferences for certain crops (for example, vegetable production for relishes) or varieties (for example, those with certain cooking traits).
- Men and women also play different roles in natural resource management, local organizations, and communication with external actors. This needs to be considered when developing programs, strategies or group and market based programs.
- Women’s responsibilities for childcare and domestic work create labor constraints, limiting the time they can farming.
- In many regions, women are increasingly involved in all types of agricultural production and labor. This is a result of male migration and occupational diversification.
- Labor constraints and other differences in access to resources will affect men and women’s ability to benefit from different types of agricultural technologies and innovations.
- Gender-based differences in task allocation within wage labor systems may result in different health impacts on men and women. This is especially problematic when women’s exposure to pesticides and other agro- chemicals causes increased risk of reproductive difficulties, miscarriages and birth defects in particular. Evidence from plantation systems indicates that women workers often receive less training and instruction than male workers in working with agrochemicals.
If the user of this tool is not sure about the capacity of the research team to integrate gender in the learning objectives, the gender capacity assessment tool available in this toolbox can be used.