Research procedure
Verify reported community actions
- Select villages after 90 days
- Interview local informants independently
- Compare accounts with action records
Illustrative summary of the source-described verification in 65 treatment villages: 41 in Indonesia and 24 in Tanzania. Interview about ten informants per village. The 91.9% verification rate applies to queried actions reported complete in the subsample, not all programme actions.
See source →How
When to use
Research teams evaluating social accountability programs need an independent method to verify self-reported community actions and assess implementation progress. Field staff conduct semi-structured interviews with local leaders and health workers following project completion. Researchers deploy this method to validate participant-reported accomplishments across program villages.
Conduct key informant interviews with local village stakeholders to verify self-reported social action plan completion.
- Select a representative subsample of treatment villages following the conclusion of the 90-day intervention period.
- Recruit independent data collection teams, such as SurveyMETER in Indonesia and Ideas in Action in Tanzania.
- Identify approximately ten key informants per village, including village leaders, health facility managers, midwives, and community representatives.
- Administer semi-structured interview protocols to inquire about specific planned social actions, implemented steps, and observable outcomes.
- Compare key informant testimonies against facilitator-recorded action plan completion data to establish verified completion rates.
Why it works
Key informant interviews reduce reporting bias and capture nuanced contextual insights regarding action execution and facility impacts.
Example
In 65 treatment villages across Indonesia (41 villages) and Tanzania (24 villages), survey teams conducted interviews with approximately ten key informants per village right after the 90-day follow-up meeting. In Indonesia, key informants confirmed 138 out of 196 queried actions, while in Tanzania, informants confirmed 81 out of 99 queried actions, verifying 91.9% of completed actions across the subsample.
About the study
Participants
65 rural treatment villages across Indonesia (41 villages) and Tanzania (24 villages), interviewing approximately ten key informants per village.
Study Design
Mixed-methods verification study embedded in two randomized controlled trials, using independent survey teams to audit self-reported social action plan completion data.
Results
Key informant interviews verified that 91.9% of the queried actions reported as complete in the subsample were carried out as described.