Skip to main content
. 2020 Aug 5;2020(8):CD011504. doi: 10.1002/14651858.CD011504.pub3

Alaofe 2016.

Study characteristics
Methods Study design: PCS
How were missing data handled? in 2008, enumerators repeated surveys with each woman in the agricultural groups who had been interviewed the year before, if possible. For the village sample, enumerators returned to previously sampled HHs and interviewed the same respondent, wherever possible. If an original respondent was not present and another woman aged > 18 years in the HH could answer the questions, she was interviewed and this was noted. If a respondent's HH could not be found, a neighbouring HH was substituted and this was noted.
Randomisation ratio: N/A
Recruitment method: HH surveys conducted for each woman in the women's agricultural groups and for a random representative sample of HHs in each village, with women aged > 18 years as respondents. Surveys conducted following installation of the PVDI systems but before any harvest. Any women who were away from the district at the time of the survey were omitted.
Sample size justification and outcome used: NR
Sampling method: all HHs of women who were involved in local women's agricultural groups were sampled from 2 intervention villages and 2 matched‐pair control villages (similar in terms of location along the same roads, administrative status and size). A random, representative sample of 30 HHs in each village was also selected from each village.
Study aim or objective: to evaluate the impact of SMGs on crop production diversity and dietary diversity in the Kalale district of Northern Benin.
Study period: November 2007 to November 2008
Unit of allocation or exposure: HHs
Participants Baseline characteristics
Intervention or exposure group (n = 116)
  • Age: WG, n: children aged 5 years: 30; children aged 5–17 years: 51; adults: 55; adults aged > 65 years: 11; NWG, n: children aged 5 years: 40; children aged 5–17 years: 55; adults: 60; adults aged > 65 years: 16

  • Place of residence: NR

  • Sex: NR

  • Ethnicity and language:

    • language, n (%): WG: Bariba 16 (28.6), Peulh 3 (5.4), Boko 34 (60.7), other 3 (3.6); NWG: Bariba 6 (10.0), Peulh 16 (26.7), Boko 31 (51.7), other 7 (11.7)

    • religion, n (%): WG: Muslim 49 (87.5); NWG: Muslim 56 (93.3)

  • Occupation: WG, n: crop production 55, livestock production 9, small vendor 9, other trade/service 9, salaried job 2, housework 7, student 49, unemployed 12, retired 2; NWG, n: crop production 49, livestock production 15, small vendor 15, other trade/service 14, salaried job 8, housework 8, student 38, unemployed 13, retired 4

  • Education: literacy, n (%): WG: 22 (39.2); NWG: 14 (23.3)

  • SES: mean HH size: WG: 7.8 (SD 3.5), NWG: 8.0 (SD 4.5); median per capita consumption expenditure: WG: USD 173.29, NWG: USD 120.33

  • Social capital: NR

  • Nutritional status: median food consumption (% of total consumption expenditure): 62 (village A); 61 (village B)

  • Morbidities: NR

  • Concomitant or previous care: NR


Control group (n = 98)
  • Ethnicity and language:

    • language, n (%): WG: Bariba 4 (10.5), Peulh 26 (68.4), Boko 7 (18.4), other 1 (2.6); NWG: Bariba 5 (8.3), Peulh 42 (70.0), Boko 6 (10.0), other 7 (11.7)

  • Religion, n (%): WG: Muslim 37 (97.4); NWG: Muslim 51 (89.5)

  • Occupation: WG, n: crop production 30, livestock production 9, small vendor 9, other trade/service 6, salaried job 1, housework 2, student 22, unemployed 5, retired 1; NWG, n: crop production 48, livestock production 19, small vendor 19, other trade/service 16, salaried job 2, housework 4, student 28, unemployed 12, retired 2

  • Education: literacy, n (%): WG: 22 (57.9); NWG: 15 (25)

  • SES: mean HH size: WG 5.9 (SD 2.7), NWG: 7.0 (SD 3.4); median per capita consumption expenditure: WG: USD 156.87, NWG: USD 131.28

  • Social capital: NR

  • Nutritional status: median food consumption (% of total consumption expenditure): 59 (village A); 62 (village B)

  • Morbidities: NR

  • Concomitant or previous care: NR


Overall group characteristics: NR
Inclusion criteria: HHs in 4 selected villages in the Kalale district
Exclusion criteria: none reported
Baseline differences: fewer literate women were in the intervention group than in the control group (P < 0.05). HH size was lower in control WG (mean 5.9 (SD 2.7)) than the intervention WG (mean 7.8 (SD 3.5)) (P = 0.2).
Total number completed and analysed: intervention group: WG: n = 56; NWG: n = 60; control group: WG: n = 38; NWG: n = 60.
Total number enrolled per relevant group: intervention group: WG: n = 56; NWG: n = 60; control group: WG: n = 38; NWG: n = 60
Total number randomised per relevant group: N/A
Attrition: NR
Description of subgroups measured and reported: for the intervention and control groups, there were 2 subgroups: HHs with women who participated in a local women's agricultural group (WG) and HHs where no‐one belonged to a women's agricultural group (NWG).
Interventions Intervention/exposure group: income generation through SMGs
  • Food access intervention category: increase buying power

  • Intervention category: income generation

  • Description: income generation through SMGs: a drip irrigation system combined with a solar‐powered water pump. The water source was from a year‐round stream in 2 villages and from a borehole in the other 2 villages. Each SMG was used jointly by the women in each village.

  • Duration of intervention period: 12 months

  • Frequency: ongoing

  • Number of study contacts: 2 (November 2007; November 2008)

  • Providers: Solar Electric Light Fund (SELF), an NGO

  • Delivery: system installation and training of local technicians took place in 2007 in time for the dry season beginning in November. Supported by funding from the World Bank Development Marketplace competition in 2006, with the expertise of the International Crops Research Institute for the Semi‐Arid Tropics (ICRISAT) in Niamey. A project team oversaw the installation and maintenance and provided continued training for farmers.

  • Co‐interventions: none reported

  • Resource requirements: hiring of a project team (director, solar technician and agricultural technician) for each village for the training of local farmers and additional technicians such as masons and electricians. Purchasing or donation of PVDI systems.

  • Economic indicators: study authors provided an economic analysis of the PVDI system in terms of installation and operational costs, compared to a liquid‐fuel pump drip irrigation system.


Control: no intervention (usual circumstances of hand‐watered irrigation)
Outcomes HH food expenditure: proportion of income spent on food; foods purchased in dry season
Dietary diversity: variety of fruits and vegetables consumed
Identification Sponsorship source: quote: "We would like to acknowledge the Solar Electric Light Fund (SELF) for implementing the PVDI project, ICRISAT technicians for their extension work with project farmers, and l'Institut de Recherche Empirique en Economie Politique (IREEP, Cotonou, Benin) for their enumeration of the HH surveys. This project was supported by an Environmental Ventures Projects grant from the Woods Institute for the Environment at Stanford University."
Country: Benin
Setting: Kalale district in northern Benin, which is a rural setting without an electricity grid, no secondary school and "100 km from a paved road;" 85–90% of HHs totally depended on agriculture for livelihoods. Many women's agricultural groups were engaged in small‐scale vegetable production before project implementation; as such, this PVDI project fit within social and cultural norms.
Authors' names: Halimatou Alaofe; Jennifer Burney; Douglas Taren
Email: halaofe@email.arizona.edu; burney@stanford.edu; taren@email.arizona.edu
Declarations of interest: quote: "The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article."
Study or programme name and acronym: Solar Market Gardens (SMGs)
Type of record: 2 journal articles
Trial registration: N/A
Notes  
Risk of bias
Bias Authors' judgement Support for judgement
Random sequence generation (Selection bias) High risk CBA study. No randomisation of intervention.
Allocation concealment (Selection bias) High risk CBA study. No random allocation of intervention.
Baseline characteristics similar (Selection bias) Low risk Although there were some baseline differences between the intervention and control groups, the regression analysis was adjusted for variables such as age, education, local languages, occupation, HH size and consumption expenditures.
Baseline outcome measurements similar (Selection bias) Low risk Quote: "At baseline, there was no significant difference in the variety of fruits and vegetables produced and consumed between the 4 groups." "… there were no significant differences in food purchases during the dry season between the 4 groups at baseline."
Blinding of participants and personnel (Performance bias) Low risk Blinding not possible, but unlikely that a lack of blinding affected outcomes.
Blinding of outcome assessment (Detection bias) High risk NR by study authors but outcomes were self‐reported and likely to be influenced by lack of blinding.
Protection against contamination (Performance bias) Low risk Allocation by village and contamination was unlikely.
Incomplete outcome data (Attrition bias) Unclear risk It is NR how many HH from the baseline survey per group were N/A for endpoint survey.
Selective outcome reporting (Reporting bias) Unclear risk Study protocol N/A.
Other bias Low risk Misclassification bias: low risk. Measurement bias: low risk. Structured questionnaire on HH food consumption. Recall period of 1 month during dry season. Seasonality bias: low risk. Follow‐up survey conducted during the same season.