The impact of sensitization workshops of solar water pump in Vietnam

In the study, both husbands and wives from treatment households are invited to attend standardized sensitization workshops on solar water pumps (SWP), delivered by district-level agricultural extension officers. To address potential bias from selective participation (attrition), control group households are similarly invited to placebo workshops on healthy diets—unrelated to agriculture or energy. All households receive identical invitations with non-specific titles ("Disseminating Knowledge to Support the Community") to ensure comparable participation rates across groups.

2025-04-14

2025-05-07

Our key outcomes include Willingness to Pay (WTP) for the solar water pumps, intention to install solar water pumps, household’s agricultural production, household income, food security, women’s time use, agricultural performance, participation in domestic chores and childcare of husbands and wives.

A. At the household level

1. Willing to pay for a SWP (continuous variable):
Definition/questionnaires: Total money that households are willing to pay for a SWP corresponding to their land area

2. Main reason for not installing a SWP (categorial variable):
Definition/questionnaires:
Do not know. . . 1
Do not have money . . .2
Not economically efficient . . . 3
Others . . . . . . .4

3. Intention to install a SWP (Binary variable)
Definition/questionnaires:
If you can get a preferential loan, will you install a solar water pump?

4. The share of agricultural land areas that are irrigated (Continuous variable)
Definition/questionnaires: The share of agricultural land areas that are currently irrigated in the total agricultural areas

5. Outputs and total sale of crops
Definition/questionnaires: Total outputs and total sales from crops: disaggregated by annual crops, perennial crops, and fruit crops.

6. Costs of cultivation.
Definition/questionnaires: Total cost of agricultural production. This variable is computed using information on costs of crop production inputs.

7. Income from crops: Total outputs minus total costs of crop productions

8. Per capita household income: Total income of households divided by household size

B. At the individual level
1. Main employment status during the past 12 months (Binary variable)
Definition/questionnaires: A number of binary variables measuring the main job types: Employer; Self-employed; Family workers; Member of a cooperative; Wage workers; Unemployed; Not working

2. Working hours during the past 7 days (Continuous variable): Total working hours during the past 7 days

3. Perception of gender issues (Discrete variable)
Definition/questionnaires: Husbands and wives are asked 8 questions about the role of men and women in life:
- A woman's most important role is to take care of home and cook
- Taking care and feeding the kids are the mother's responsibility
- The most important role of men is to earn money
- Men should do big thing not house work
- Men should not cook or wash clothes
- Men should not wash dishes after meals
- On the whole, men make better political leaders than women do
- On the whole, men make better business executives than women do
The responses are: 1. Totally agree; 2. Partially agree; and 3. Disagree

4. Decision makers in family (discrete variable)
Definition/questionnaires: Husbands and wives are asked questions about who the main decision-maker is for the following issues:
- Household production and business
- Major household expenditure consumption (such as: motorbike, land, TV . . . )
- Minor household expenditure consumption (such as: food for daily consumption or other household needs)
- Childcare
The mutually exclusive responses: 1. Solely husband; 2. Mainly husband; 3. Both husband and wife; 4. Mainly wife; 5. Solely wife.

5. Doing housework (discrete variable)
Definition/questionnaires: Husbands and wives are asked questions about doing housework such as washing clothe, buying food, cleaning house, and childcare during the past 7 days

6. Time use on different activities (discrete variable)
Definition/questionnaires: Time that individuals spent on different activities during the 24 hours of a typical day in the last week (sleep, Care for others; Take care of yourself; Domestic tasks; Tasks on family crop; Tasks livestock, aquaculture and other activities related to agriculture; Tasks on nonfarm business; Activities for pay or for money outside of household; At school/college/University; Studying at home; Leisure)

In Đắk Lắk province, the research team collaborated with the provincial Department of Agriculture and Rural Development to identify four districts with favorable conditions for solar water pump (SWP) adoption: Buôn Đôn, Cư M’gar, Krông Buk, and Krông Bông. From these districts, 30 rural communes with the highest number of agricultural households were selected. Each commune then identified four villages suitable for SWP adoption, resulting in a total of 120 study villages.

Villages were randomly assigned to treatment and control groups within each commune:

Treatment group: Two villages per commune were randomly selected. In each village, 10 households with the largest areas of perennial crops (e.g., fruit trees or coffee) were identified with help from village leaders. These 600 households (from 60 villages) form the treatment group and are invited to attend sensitization workshops on the benefits of SWP.

Control group: The remaining two villages per commune follow the same household selection process (10 households per village), totaling 600 households. These households form the control group and are invited to attend placebo workshops on healthy diets, unrelated to agriculture or energy.

In total, the study covers 1,200 households across 120 villages, with 60 workshops conducted for each group.

Not available

We conducted the random assignment of villages to treatment and control groups using custom Stata code to ensure transparency and replicability

Villages

Yes

120 villages

1200 observations

Treatment group: 60 clusters (villages), 600 households.
Control group: 60 clusters (villages), 600 households.
Total: 120 clusters and 1,200 households.

5%