Target Audience: Data analysts, researchers, development practitioners, policy makers, and anyone working with data in South Asian contexts
Prerequisites: Basic familiarity with data collection and analysis helpful but not required
Materials Needed: Laptops/tablets for data exploration, notebooks, sticky notes, flip chart paper
By the end of this workshop, participants will be able to:
Government Report: "96.5% of people live in households with an improved drinking-water source (NFHS-6, 2023-24)."
Feminist Analysis: "An improved source can still be a walk away from home, and the figure says nothing about who fetches the water, how far, or how often. Time-use data points to who: on the survey day, 81.5% of women and girls aged 6 and above did unpaid domestic work, against 27.1% of men and boys (Time Use Survey 2024)."
Same data, different questions, different insights.
Who collects data? Who analyzes it? Who gets to decide what counts as "data"?
Use data to challenge existing hierarchies rather than reinforcing them.
Value experiential knowledge alongside statistical evidence.
Move beyond male/female, urban/rural to understand complex identities.
Multiple perspectives lead to better understanding than single "objective" view.
Numbers don't exist in vacuum - historical, cultural, political context matters.
Acknowledge who does the work of data collection and cleaning - often invisible.
| Data Area | Who's Typically Counted | Who's Often Missing | Why It Matters |
|---|---|---|---|
| Economic Surveys | Formal sector workers, heads of household | Informal workers, women's unpaid labor, care work | Underestimates women's economic contribution |
| Health Studies | Cisgender adults, urban populations | LGBTQIA+ people, children, migrant workers | Health interventions miss vulnerable groups |
| Education Data | Enrolled students in formal schools | Out-of-school children, disability, caste discrimination | Policies don't address systemic exclusion |
| Violence Statistics | Reported crimes, police data | Domestic violence, caste violence, police violence | Underrepresents actual prevalence and patterns |
Small Groups (3-4 people):
Step 1 (5 min): Choose one dataset you use in your work (surveys, admin data, monitoring data)
Step 2 (7 min): Map the power dynamics:
Step 3 (3 min): Identify one way this power dynamic might bias your understanding
Report Back: Each group shares one insight about hidden power structures in their data
Intersectionality: People experience multiple, overlapping forms of discrimination and advantage. Data analysis has to capture this complexity instead of looking at single categories in isolation.
Aggregate Data: "Maternal mortality ratio is 164 per 100,000 live births (Sample Registration System, 2016-18)"
Intersectional Analysis Asks, and the official series cannot answer:
Insight: Caste, geography, age, and ability may intersect to create very different maternal health risks, but an official series that is not disaggregated keeps those differences invisible. Generic "maternal health" interventions may then miss those at highest risk.
Individual Reflection (5 min): Look at a dataset you know well. Pick one categorical variable (gender, occupation, religion, etc.)
Ask These Questions:
Pair Discussion (5 min): Share your analysis with a partner
Large Group (2 min): Quick insights about hidden assumptions in data categories
Remember: Feminist data practice keeps quantitative methods and uses them more thoughtfully and equitably.
Before You Start Analysis:
During Analysis:
When Presenting Results:
Data always reflects the priorities and assumptions of the people who collected it. Feminist data practice examines those choices and uses data to reduce inequality.
Essential Reading:
South Asian Context:
Practical Tools:
Next Steps in ImpactMojo:
This handout is part of the ImpactMojo 101 Knowledge Series
Licensed under CC BY-NC-ND 4.0 • Free to use with attribution • www.impactmojo.in
For the complete Data Feminism 101 course with datasets, analysis templates, and intersectional frameworks, visit the ImpactMojo platform.
