This presentation was given by Catherine Ragasa (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/

1. Knowledge is power
Impact of interactive radio programming on women’s
empowerment and agricultural development outcomes
in Malawi
Catherine Ragasa
International Food Policy Research Institute
Canberra | April 4, 2019

2. Outline of the presentation
 Intervention
 Research questions
 Data sources
 Indicators and analytical methods
 Results
 Some reflections

3. Interventions
 Interactive radio programming piloted in the 2015 (Farm Radio Trust +
DAES)
o Radio messaging (national + community radio stations)
o Daily broadcasting, varied innovations, farmers have choice on agricultural radio
station
o Call-in center (anyone can call for free)
o Mobile platform (anyone can text for free)
o Videos and short text messages sent via mobile phones
o focused on a few districts: Kasungu, Dedza, Mchinji, and Ntchisi
o Listening club (introduced ICT to existing farmers’ groups)
o focused on a few districts: Kasungu, Dedza, Mchinji, and Ntchisi
 Agricultural radio messages
o Crop diversification, focusing on legumes promotion
o Nutrition, focusing on dietary diversity
o Gender equality (jingles, dramas and experts and farmer interviews)

4. Research questions
Is there difference between access to radio and other sources
of information by women and men and age group?
Does access to radio messaging by women or men (or both) in
the household have an effect on crop diversification, dietary
diversity and intrahousehold equality?
Does individual’s access to radio messaging associated with
greater empowerment?

5. Mixed methods
 Household and community surveys (nationally representative; 3000 HH:
July-Sep 2016 (baseline); July-Sep 2018 (follow-up)
o Modified A-WEAI (shorter questions on time use) only for 2018
 Focus group discussions (55 FGDs) (dots on map); Jan/Feb 2017,
Jan/Feb 2019
o Local enumerators fluent in Chichewa, Chibandya, and Chinyika
o Enumerators were experienced in qualitative data collection and facilitators
to encourage active participation and articulation of differing viewpoints
o Discussions were recorded, transcribed, and translated and then
thematically coded using NVivo 11
 2018 data of callers in the call-in centers
 Census and monitoring of state and non-state extension service
providers in 15 districts (shaded districts in map on the right): Dec 2016
- Mar 2017
 In-depth interviews with 30 service providers and 71 extension workers:
Dec 2016 - Mar 2017

6. Indicators and Analytical Methods
(individual and household levels)
 Individual level
o Outcome: Empowerment (measured in terms of pro-WEAI empowerment score)
o Variable of interest: Listens to agricultural radio programming (0/1)
o Controls: other information sources, education, age, household asset, distance
to nearest road, community-level indicators, district and region dummies, etc.
o Cross-sectional data analysis for determining treatment effects:
o Matching techniques and instrumental variable approach (instrument:
ownership of radio)
o Fractional regression model (0-1 outcome indicators)
o Compared focus districts (Kasungu, Dedza, Mchinji, and Ntchisi) with heavy
promotion of mobile platform, videos, and listening clubs and comparable
districts

7. Indicators and Analytical Methods
(individual and household levels)
 Household level
o Outcome:
o Household dietary diversity score
o Legumes promotion (acres planted with legumes)
o Crop diversification (Simpson index for crop diversification)
o Variables of interest:
o Whether woman received agricultural/nutrition information from radio (0/1)
o Whether man received of agricultural/nutrition information from radio (0/1)
o Controls: other information sources, education and age of woman and man;
household asset, distance to road, GIS-related data, community-level data, etc.
o Panel data analysis for determining treatment effects: Correlated random effects
model and instrumental variable approach (instrument: ownership of radio)

8. Fivedomainsofempowerment
Modified Abbreviated WEAI (modified A-WEAI)
1/5
2/15
1/5
1/5 shorter version
Questions on time constraint:
1. If you wanted to do something
(livelihood-related, training-related,
self-care), will you be able to
reorganize your tasks so that you can
make time to do it?
2. If you wanted to do something
(livelihood-related, training-related,
self-care) and could not take your
child with you, is there someone who
could care for your child in your
absence?
3. If you had an additional 2 hours in
your day to do anything, how would
you spend that time?  if answered
“more sleep/rest”, we coded as
time constrained

9. 1. Lack of access to credit and group
membership are the major sources of
disempowerment
2. Similar levels of empowerment by
women and men
0.00
0.20
0.40
0.60
0.80
1.00
%ofrespondents
Panel a. % of respondents achieving
adequacy by gender
Men Women
0.00
0.20
0.40
0.60
0.80
1.00
%ofrespondents
Panel b. % of respondents achieving adequacy by
gender and age group
Young Men Older Men Young Women Older Women
3. Young men are more disempowered in terms of lack of inputs to
production decisions
4. Young men and women are more disempowered in terms of lack of
group membership
5. Older women are more disempowered in terms of heavier workload
6. Older men and women are disempowered in terms of lack of access
to credit

10. 0.042 0.047 0.052
0.035 0.047 0.047
0.006 0.005
0.010
0.004
0.007 0.004
0.069 0.069
0.068
0.071
0.066 0.072
0.007 0.006
0.008
0.006
0.006 0.005
0.130 0.123
0.143
0.120
0.132 0.115
0.063 0.065
0.061
0.064
0.060
0.070
0.000
0.050
0.100
0.150
0.200
0.250
0.300
0.350
0.400
Men Women Young Men Older Men
Young
Women
Older
WomenTotaldisempowermentsub-indicatorscore
Contributors to disempowerment
Work balance
Group membership
Control over use of income
Access to credit
Ownership of assets
Inputs to production decisions

11. Access of agriculture advice, by source
56
40
29
24
13
5 2 2 2 0 1 0 1
0
20
40
60
80
%ofhouseholds
a. In the last 2 years
2016 2018
Collected only in 2018
35
27
17
11
7
2 1 1 1 0 0 0 0
0
20
40
60
80
b. In the last 12 months
2016 2018
Collected only in 2018

12. Access to extension services, by method/approach
47
41
28
15 14
11 10
7 7 5 4 4 4
2 2 2
0
20
1 0
0
5
10
15
20
25
30
35
40
45
50
%ofsamplehouseholds 2016 2018
Collected only in 2018

13. 1. Men have greater access to almost all sources of
agricultural advice
2. Older men have greatest access to almost
sources of agricultural advice
3. Compared to other sources (e.g., govt), radio is
the top source for young men and women
0.00
0.10
0.20
0.30
0.40
0.50
Young Men Older Men Young Women Older Women
Panel b. Sources of nutrition education in the last
12 months
Government extension NGO extension
Lead farmer Other farmers
Radio Mobile apps
Health workers Hospitals/clinics
0.00
0.10
0.20
0.30
0.40
0.50
Young Men
(N=1740)
Older Men
(N=2280)
Young Women
(N=2273)
Older Women
(N=2324)
%ofrespondents
Panel a. Source of agricultural advice in the last
12 months
Government extension NGO extension
Lead farmer Other farmers
Radio Mobile apps
4. Women have greater access to most sources of
nutrition advice
5. Men (both young and older) have greater access
to nutrition advice from radio compared to other
sources and compared to women

14. Results of regression analysis
 Access to radio programming is associated with 4% greater empowerment score of women
and men
o Relative to other sources, access to radio had greater effect on improving empowerment of women
than men
 Hard to explain variations in intrahousehold equality scores (W-M)  almost all explanatory
variables were not statistically significant
o 60% of households achieved intrahousehold equality (e.g., woman is at least as empowered as
man)
o Men’s access to nutrition advice from radio reduces the gap between women and men
empowerment scores
 Positive impact of access to radio programming on increased crop diversification and legume
harvested area
o Access to radio by woman and by man improved crop diversification and legume harvested area
 Positive impact of access to radio programming on improved household dietary diversity
scores (HDDS)
o Access to radio by woman and man improved HDDS
 Contribution of listening club and mobile platform? Households in focus districts (Kasungu,
Dedza, Mchinji, and Ntchisi) have higher treatment effects of radio access than comparable
districts in most models estimated

15. Some Reflections
 Radio is the top source of agricultural and nutrition advice
o 60% of households reported listening to radio at least once a week and 41%
listens to radio at least once daily
o More than half of women (52%) reported listing to radio at least once a week.
o Younger women and men used radio more than other sources for their agricultural
information needs
o Younger and older men used radio more than other sources for nutrition education
o Radio seems to be a critical delivery platform for nutrition education for men,
circumventing strong gender norms on women’s role on domestic work and nutrition while
men are usually teased and laughed at when attending meetings/trainings related to
nutrition
 Our results also show significant effect of interactive radio programming on both
women’s and men’s empowerment scores
o Relative to other sources, access to radio had greater effect on improving empowerment
of women than men
 Mechanisms for this gendered outcome come from lower time burden and
awareness campaigns and messages on gender equality

16. Some Reflections
 Radio programming has been demand-driven at group- and community-level
 At national level, at least yearly review and harmonization of agricultural
radio messaging
 Farmers have to option of which agricultural radio station to listen to
o Reason why certain radio station was selected: The issues discussed are relevant to
my needs
 Types of contents farmers like to have in agricultural radio programs:
o Interviews with farmers
o Experts talking
o Experts answering questions
o Farmers answering questions
o Agricultural news
o Drama, music
o Weather forecast
o Market prices

17. Some Reflections
 Other sources of advice (government and NGO extension) also have
effects on the outcome variables  combining radio messaging and other
sources matters
o Beaman et al. (2015) indicate that technology diffusion and behavior change are
characterized by a complex contagion learning environment in which most
farmers need to learn from multiple people and sources before they adopt
themselves
 Listening clubs linked to the radio program seem to be useful platforms that
strengthen social capital and cooperation among listeners.
 Call centers and mobile platform, in which anyone can call or text for free,
may have helped in greater responsiveness of service provision to farmer’s
demands

18. 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