Exploring empowerment transitions of women and men in Bangladesh
•Download as PPTX, PDF•
0 likes•49 views
This presentation was given by Audrey Pereira (International Food Policy Research Institute), as part of the Annual Scientific Conference hosted by the University of Canberra and co-sponsored by the University of Canberra, the Australian Centre for International Agricultural Research (ACIAR) and CGIAR Collaborative Platform for Gender Research. The event took place on April 2-4, 2019 in Canberra, Australia. Read more: https://www.canberra.edu.au/research/faculty-research-centres/aisc/seeds-of-change and https://gender.cgiar.org/annual-conference-2019/
Recommended
More Related Content
What's hot
What's hot (7)
Similar to Exploring empowerment transitions of women and men in Bangladesh
Similar to Exploring empowerment transitions of women and men in Bangladesh (20)
More from CGIAR
More from CGIAR (20)
Recently uploaded
Recently uploaded (16)
Exploring empowerment transitions of women and men in Bangladesh
- 1. Exploring empowerment transitions of women and men in Bangladesh Akhter Ahmed, Hazel Malapit, Audrey Pereira, Agnes Quisumbing, & Salauddin Tauseef April 3rd, 2019 | Seeds of Change Conference | Canberra, Australia
- 2. Empowerment Process of gaining the ability to make strategic life choices, when these choices were previously denied (Kabeer, 1999) Drydyk (2008) proposed that empowerment should be durable Gaining the ability to make strategic life choices Also be able to continue to exert that power and remain empowered over time Empowerment is an ongoing process Not a one-time destination Empowerment is relative To others To oneself at a different point in time
- 3. Current evidence Akter & Chindakar (2019) Decision making, control over financial resources, freedom of mobility, freedom from domestic violence, and marital inclusiveness Created a multidimensional empowerment index (MDEI) India Human Development Survey (IHDS), for the years 2004−5 and 2011−12 54% of women remained empowered over time Age, education, asset ownership, wealth and collective assets foster empowerment durabilityPhoto credit: Kalyani Raghunathan
- 4. Motivation Drawing from the poverty literature Household poverty transitions has been well-studied in Bangladesh Ahmed & Tauseef, 2018; Quisumbing, 2007; Davis & Baulch, 2011 Apply the poverty dynamics methodology to empowerment Study explores transitions in empowerment using a panel dataset For men and women in the same household Uses an internationally validated measure of empowerment Photo credit: HKI and VAARD teams
- 5. Research questions How do men and women transition into and out of empowerment over time? In what domains do these transitions occur, and how do they differ between men and women? What are the factors that facilitate these transitions? Time 1 Empowered Disempowered Time2 Empowered Sustained empowerment Moving into empowerment Disempowered Falling into disempowerment Persistent disempowerment
- 6. Bangladesh Integrated Household Survey (BIHS) Nationally-representative rural household panel survey Two survey instruments: HH-questionnaire with detailed sex-disaggregated data Different modules administered to M and F in each HH Community questionnaire 2011-12 and 2015 The BIHS 2011/12 and 2015 panel dataset is funded by the U.S. Agency for International Development (USAID), designed by the Bangladesh Policy Research and Strategy Support Program (PRSSP) implemented by IFPRI, and administered by Data Analysis and Technical Assistance (DATA).
- 7. What in the world is WEAI? Women’s Empowerment in Agriculture Index Developed by USAID, IFPRI & OPHI Launched in 2012, now used in 53 countries Measures inclusion of women in the agricultural sector Survey-based index - interviews men and women in the same household
- 8. How is the Index constructed? Constructed using interviews of the primary male and primary female adults in the same household Women’s Empowerment in Agriculture Index (WEAI) WEAI is made up of two sub indices All range from zero to one; higher values = greater empowerment
- 10. Empowerment Empowered if respondent is adequate in at least 80% of the weighted indicators We use a binary indicator for empowerment Alkire et al., 2012 Only HHs with no missing indicators for the primary male and female respondent in both time periods Attrition weights calculated following methodology in Fitzgerald, Gottschalk and Moffitt (1998) Sample covers 2,575 HHs Adult men and women
- 11. Respondent characteristics Female Male Test of differenceMean (SE) Mean (SE) Age (years) 36.88 43.56 *** (0.23) (0.24) No schooling 0.48 0.51 *** (0.01) (0.01) Some primary education 0.14 0.12 ** (0.01) (0.01) Completed primary education 0.15 0.12 *** (0.01) (0.01) Some or completed secondary education or higher 0.23 0.25 (0.01) (0.01) Estimates are weighted using HH-level attrition weights *** p<0.01, ** p<0.05, * p<0.1
- 12. Household characteristics Estimates are weighted using HH-level attrition weights *** p<0.01, ** p<0.05, * p<0.1 Household Mean (SE) Negative Shocks Mean (SE) Positive events Mean (SE) Household size 4.34 Any death in household 0.00 Received remittances 0.01 (0.04) (0.00) (0.00) Child <5 years lives in HH 0.38 Any illness in household 0.14 New job or business profit 0.03 (0.01) (0.01) (0.00) Adult >55 years lives in HH 0.29 Dowry or wedding expenses 0.05 Daily per capita HH expenditure (taka) 87.89 (0.01) (0.00) (1.23) Nuclear HH 0.77 Any livestock death or theft 0.07 (0.01) (0.01) Flood damage to house/livestock/crops 0.07 (0.01) Crop loss due to non- flood reasons 0.04 (0.01) Other shocks 0.07 (0.01)
- 13. Empowerment T1 & T2 Improvement in empowerment status for women and men over time (net) Statistically significant differences between women and men within each time period at p<0.01 25% 47%48% 58% 0% 10% 20% 30% 40% 50% 60% 70% Empowered in T1 Empowered in T2 Female (N=2,575) Male (N=2,575) Estimates are weighted using HH-level attrition weights
- 14. Empowerment transitions 15% 30% 33% 28% 11% 17% 42% 25% 0% 25% 50% 75% 100% Female (N=2,575) Male (N=2,575) Sustained empowerment Moving into empowerment Falling into disempowerment Persistent disempowerment All estimates are statistically significantly different between men vs. women at p<0.01
- 15. Empowerment transitions by indicator 29% 12% 33% 27% 22% 20% 17% 42% 15% 21% 15% 6% 20% 16% 3% 20% 10% 24% 16% 8% 13% 10% 20% 13% 24% 24% 15% 4% 9% 4% 11% 12% 19% 16% 23% 0% 25% 50% 75% 100% F M F M F M F M F M F M F M F M F M F M Prod. Decision s Autonm y Asset Owners hip Asset RightsCreditIncomeGroup Public Speakin g Workloa dLeisure Sustained adequacy Moving into adequacy Falling into inadequacy Persistent inadequacy Income Prod. Dec Autonomy Assets Asset Rights Credit Group Public Sp. Workload Leisure
- 16. Rights over assets 41% 81% 20% 4% 27% 10% 13% 4% 0% 25% 50% 75% 100% Female (N=2,575) Male (N=2,575) Sustained adequacy Moving into adequacy Falling into adequacy Persistent adequacy All estimates are statistically significantly different between men vs. women at p<0.01 Respondent can decide whether to sell, give away, or rent/mortgage [asset] most of the time HH assets except chickens and non-mechanized farming equipment Inadequate if HH does not own assets
- 17. Group membership 9% 3% 20% 12% 17% 13% 54% 71% 0% 25% 50% 75% 100% Female (N=2,575) Male (N=2,575) Sustained adequacy Moving into adequacy Falling into adequacy Persistent adequacy All estimates are statistically significantly different between men vs. women at p<0.01 Respondent is part of at least one group in the community Inadequate if no groups in community
- 18. Workload 55% 38% 24% 16% 15% 24% 7% 21% 0% 25% 50% 75% 100% Female (N=2,575) Male (N=2,575) Sustained adequacy Moving into adequacy Falling into adequacy Persistent adequacy All estimates are statistically significantly different between men vs. women at p<0.01 Respondent worked <=10.5 hours in the previous 24 hours Includes productive and domestic/reproductive work
- 19. Methodology Outcome: probability of being in an empowerment state Multinomial regression models Reference category: Persistent disempowerment Controls (shocks 2015 data; 2011-12 others) Individual Household Idiosyncratic Shocks Covariate shocks Other Age HH size Death Flood damage to house/livestock/crops Other shocks Education Nuclear HH Illness Crop loss due to non-flood reasons Positive remittances Child <5 in HH Wedding expenses/dowry New job; business profit Adult 55+ in HH Livestock death/theft Female-headed HH Productive asset loss Income per capita Division
- 20. Selected results: Individual characteristics Falling into disempowerment Moving into empowerment Sustained empowerment Women Men Test Women Men Test Women Men Test Age (years) -0.00 0.00 0.02** -0.01 0.00 0.02*** (0.00) (0.00) (0.01) (0.01) (0.01) (0.01) No education (reference group) Some primary education 0.02 -0.01 * -0.02 0.00 * 0.02 0.10*** ** (0.02) (0.02) (0.03) (0.03) (0.02) (0.03) Completed primary education -0.00 -0.03 0.07** 0.04 -0.00 0.06** (0.02) (0.02) (0.03) (0.03) (0.02) (0.03) Some secondary education or higher -0.01 -0.03 0.00 -0.01 0.03* 0.11*** * (0.02) (0.02) (0.03) (0.03) (0.02) (0.02) Estimates are weighted marginal effects; *** p<0.01, ** p<0.05, * p<0.1 Reference category: Persistent disempowerment
- 21. Selected results - Shocks Falling into disempowerment Moving into empowerment Sustained empowerment Women Men Test Women Men Test Women Men Test Female-headed household -0.90*** 0.73*** *** -2.59*** 0.99*** *** 1.02*** -2.42*** *** (0.09) (0.15) (0.14) (0.24) (0.14) (0.16) Any death in household -1.12*** -0.01 *** 1.14*** 0.06 *** -1.36*** -0.02 *** (0.08) (0.11) (0.11) (0.12) (0.09) (0.12) Any illness in household -0.01 -0.03 0.04 0.05* -0.02 0.05** ** (0.02) (0.02) (0.03) (0.03) (0.02) (0.02) Dowry or wedding expenses 0.01 -0.09** * 0.06 0.03 * 0.06* 0.08* (0.03) (0.04) (0.05) (0.04) (0.03) (0.04) Any livestock death or theft -0.03 0.01 0.02 0.01 0.07** 0.09* (0.03) (0.04) (0.05) (0.05) (0.03) (0.05) Flood damage to house/livestock/crops -0.06 -0.05 0.12** 0.08 -0.13** -0.06 (0.03) (0.04) (0.05) (0.05) (0.05) (0.04) Crop loss due to non-flood reasons -0.01 -0.00 0.13** 0.10* 0.06 0.12** (0.04) (0.05) (0.07) (0.05) (0.04) (0.06) Estimates are weighted marginal effects; *** p<0.01, ** p<0.05, * p<0.1 Reference category: Persistent disempowerment
- 22. Next steps and conclusion There has been considerable progress in improving women and men’s empowerment in Bangladesh Need for better recognition of unintended consequences that may exacerbate gender differences and/or outcomes Role of programs and policies: who are we (not) reaching? Next steps : more iterations; unpacking the data by indicator Suggestions welcome!
- 23. Acknowledgments and references With many thanks to Jessica Heckert, Elena Martinez, Emily Myers, Farha Sufian, Wahid Quabili, the Seeds of Change conference organizers, and the CG Gender Platform for funding Ahmed, Akhter, and Salauddin Tauseef. Climbing up the ladder and watching out for the fall: Poverty dynamics in rural Bangladesh. Vol. 1791. Intl Food Policy Res Inst, 2019. Akter, Sonia and Chindarkar, Namrata. An empirical examination of sustainability of women’s empowerment using panel data from India. Forthcoming (2019) Journal of Development Studies. Alkire, Sabina, Ruth Meinzen-Dick, Amber Peterman, Agnes Quisumbing, Greg Seymour, and Ana Vaz. "The women’s empowerment in agriculture index." World Development 52 (2013): 71-91. Davis, Peter, and Bob Baulch. "Parallel realities: exploring poverty dynamics using mixed methods in rural Bangladesh." The Journal of Development Studies 47, no. 1 (2011): 118-142. Drydyk, Jay. "Durable empowerment." Journal of Global Ethics 4, no. 3 (2008): 231-245. Fitzgerald, John, Peter Gottschalk, and Robert A. Moffitt. "An analysis of sample attrition in panel data: The Michigan Panel Study of Income Dynamics." (1998). Kabeer, Naila. "Resources, agency, achievements: Reflections on the measurement of women's empowerment." Development and change 30, no. 3 (1999): 435-464. Quisumbing, Agnes R. "Poverty transitions, shocks, and consumption in rural Bangladesh: Preliminary results from a longitudinal household survey." (2007). Quisumbing, Agnes R., and Bob Baulch. "Assets and poverty traps in rural Bangladesh." The Journal of Development Studies 49, no. 7 (2013): 898-916.
- 24. gender.cgiar.org We would like to acknowledge all CGIAR Research Programs and Centers for supporting the participation of their gender scientists to the Seeds of Change conference. Photo: Neil Palmer/IWMI
Editor's Notes
- Good morning, everyone, and thank you for being here today. My name is Audrey Pereira and I’m a Research Analyst at the International Food Policy Research Institute. Today, I’ll be presenting some preliminary findings from an analysis that looks at how men and women transition into and out of empowerment over time, in rural Bangladesh. I’m doing this work with Agnes, Akhter and Hazel, who are with IFPRI, and Tauseef, who is with the University of Manchester.
- Naila Kabeer defines empowerment as the process by which one gains the ability to make strategic life choices, when those choices were previously denied. “Drydeck” takes this definition one step further, and proposes that empowerment should be durable. Not only should one gain the ability to make strategic life choices, but he or she should also be able to continue to exert that power over time, and remain empowered. So, in this sense, empowerment is not a one-time, or one-off, destination. It’s an ongoing process. Empowerment is also relative. Because it’s a process, it’s relative to others or yourself at a different point in time.
- The motivation for this study comes largely from the poverty literature and has been well-studied. There is established methodology, including in Bangladesh, that looks at at how households transition into and out of poverty over time. So, essentially in this analysis, we are taking the poverty dynamics methodology and applying it to empowerment. To our knowledge, this is the first study that uses a panel dataset to explore how men and women in the same transition into and out of empowerment over time, and we use an international validated measure of empowerment – the Women’s Empowerment in Agriculture Index or WEAI.
- The motivation for this study comes largely from the poverty literature and has been well-studied. There is established methodology, including in Bangladesh, that looks at at how households transition into and out of poverty over time. So, essentially in this analysis, we are taking the poverty dynamics methodology and applying it to empowerment. To our knowledge, this is the first study that uses a panel dataset to explore how men and women in the same transition into and out of empowerment over time, and we use an international validated measure of empowerment – the Women’s Empowerment in Agriculture Index or WEAI.
- So our main research questions are: How do men and women transition into and out of empowerment over time? In what domains do these transitions occur? And do they differ between men and women? And finally, what are the factors that predict these transitions? In our analysis, we look at four categories or states of of transition; The first is sustained empowerment, or being empowered in both time periods The second is moving into empowerment, i.e. being disempowered in time 1 and empowered in time 2 The third is falling into empowerment, which is being empowered in time 1 and disempowered in time 2 And the last is persistent disempowerment, or being disempowered in both time periods
- The data for this analysis comes from the Bangladesh Integrated Household Survey, which is a nationally-representative rural household panel survey that was conducted in 2011-12 and 2015. And we use both rounds of data in this analysis. The survey contains two instruments – one is a community questionnaire, and the second is a household-level questionnaire that collects detailed sex disaggregated data. Different modules are administered to the primary male and female respondent in each HH, and one of these modules is the Women’s Empowerment in Agriculture Index, administered to the primary male and female respondents, and is what we use to measure empowerment in our analysis.
- So what is the WEAI? It was launched in 2012 as a collaboration among IFPRI, OPHI and USAID. It’s currently in use in 53 countries, and it measures the inclusion of women in the agricultural sector.
- The survey itself is composed of two sub-indices. The first is the five domains of empowerment index, which is a direct measure of an individual’s empowerment in five domains. The second is the Gender Parity Index which measures the relative achievements between women and men in the same household. For our analysis, we use the 5DE as the main measure of empowerment.
- The 5DE is made up of ten indicators across five equally distributed domains. The five domains are Production, Resources, Income, Leadership and Time, and these are the indicators included in each of the domains.
- We define empowerment as being adequate in at least 80% of the weighted indicators. In our analysis, we only include households that are not missing any of the ten indicators for either the man or woman respondent in either time period, and we use attrition weights to adjust our estimates. This leaves us with 2,575 households for our analysis.
- Here’s what our sample looks like. These are weighted estimates and the test of differences measures the difference in means between men and women in our sample. On average, men are older than women, but women tend to have slightly more education than men, and these are statistically significant at the 1% level.
- In terms of household characteristics, the mean household size was about 4, and three-quarters of the households in our sample were nuclear households. Approximately 40% of HHs had a child under 5, and about 30% had an adult over 55. We also look at idiosyncratic and covariate shocks that households experienced between 2011-12 and 2015. About 15% of households had a sick household member, 7% had flood damage and 4% had non-flood damage to crops. About 3% of HHs reported a new job or business profit.
- This is what empowerment looked like in each time period. Between 2011-12 and 2015, we see that the proportion of men and women who were empowered increased. For women, it was by about 22 percentage points, and for men it was by about 10 percentage points. In each time, there were statistically significant differences between men and women at the 1% level. So, overall this is good, because we see that women are catching up to men, but if we focus only on net improvement, it gives us only a small part of the overall picture.
- This graph shows us how men and women transitioned over time. Dark green shows sustained empowerment, light green shows moving into empowerment. So green is good – we want to see that. Yellow shows falling into disempowerment, and orange is persistent disempowerment. Overall, a bigger proportion of men are empowered compared to women, which is the same as the graph on the previous slide. But if we look at the margins, we find that women are making more progress than men, i.e. moving into empowerment, and a larger proportion of men, compared to women, are falling behind. Within each of the four categories, there are statistically significant differences between men and women.
- That was empowerment as measured by the 5DE score. In this graph, we look at transitions by indicator. The ten indicators that make up the 5DE are on the left axis. Overall, there’s more dark and light green, which is sustained empowerment or moving into empowerment, which is what we want to see. But there are a few indicators that are quite interesting.
- The first is rights over assets. This graph is just blown up from the previous slide. A respondent is considered adequate in this indicator if he or she can decide whether to sell, give away, or transfer an asset most of the time. And we ask this question about various HH assets such as land, livestock and farm equipment. In this indicator, we see that men are doing much better than women. Although 20% of women moved into adequacy in this indicator, 27% fell into inadequacy, in this time period.
- The next indicator is group membership. Adequacy is defined as being a member of at least one group in the community; and those in communities without groups are considered inadequate. This is one indicator where we find that men are doing really poorly. Most men are not group members, or there are no such groups in the community. One of the factors that may be driving these results is that organizations tend to focus on groups for women, and such programs or interventions may inadvertently or unintentionally leave men behind.
- The last interesting indicator is workload. A respondent is adequate in workload if he or she worked 10.5 hours or less in the previous 24 hours. This is another indicator where men are faring worse than women, and that a lot of men fell into inadequacy or remained inadequate over time. A small note that I’m not saying that women are not overworked. This definition accounts for childcare as a primary activity but not as a secondary activity, for example she is cooking and watching her child at the same time, and we know that many women multitask. Based on the definition of the indicator, we find that women are progressing faster than men.
- To determine factors that drive the probability of being in a particular empowerment state, we use multinomial regression models following the poverty dynamics literature. Our main outcome is the probability of being in an empowerment state. And our reference category is persistent disempowerment. We control for individual and household-level characteristics from 2011-12, to prevent endogeneity, and shocks and positive events that occurred between 2011-12 and 2015.
- Here are the results on individual characteristics. Our reference category is persistent disempowerment and these estimates are weighted marginal effects. The tests show whether there are statistically significant differences in the coefficients between women and men. A positive estimate indicates increased probability of being in a particular transition category compared to being in the persistent disempowerment one, and a negative estimate is lower probability. So, for example, having some secondary education or higher compared to no education, is positively associated with being in the sustained empowerment state, compared to the persistent disempowerment state for both women and men, but the effect is higher for men, and it’s marginally significant.
- Here are some more selected results, looking at shocks. We find that women who experienced the death of household member are less likely to be in the sustained empowerment or falling into disempowerment category, but more likely to be in the moving into empowerment category. There are a couple of different mechanisms at play here – the death of a household member may signal a loss of income earned, especially if that HH member was the breadwinner, but it could also lead to inheritance. So many of these findings require further unpacking.
- In summary, although ther ehas been considerable progress in improving empowerment in Bangladesh, we need to better recognize the unintended consequences of programs and policies For our next steps, we will be unpacking the data more, running more regressions on individual indicators, and robustness checks. This is still a work in progress, so your suggestions and feedback are very welcome. Thank you.