How IPA cut survey time in half with SurveyCTO's offline functionality

"With SurveyCTO's offline publishing features, we were able to really simplify our data flow. Instead of having to re-enter information every time, we were able to have a list of pre-existing political candidates, update that dataset locally on every phone, and then, when a previously-nominated candidate came up again, you just had to select their name instead of re-entering them."

Meet IPA

Innovations for Poverty Action (IPA) is a leading research and policy nonprofit that works to solve global poverty with evidence. Since 2002, they have conducted hundreds of evaluations in over 50 countries and built long-term relationships with local decision-makers in 20 countries in Africa, Asia, Latin America, and the Caribbean.

Offices: Africa, Asia, Latin America
Sector: International development
Employees: 500+

The Challenge: inefficiencies and duplicate work in household surveys

In 2023, IPA Sierra Leone undertook an important effort to encourage local participation in the political process by empowering rural people to directly nominate council candidates. Because of the deep ethnic divisions that split Sierra Leone’s politics, potential candidates find it impossible to win an election without a party’s financial support or party symbol. Especially in rural wards, candidates have little means of establishing connections with a party, and the parties likewise are unable to discover high quality candidates.

IPA Sierra Leone sought to discover the most popular nominees in rural wards across the country. They hired enumerators to survey local households on who they would like to see run for council in their community, find the most commonly recommended names, and survey those potential candidates for their qualifications. Then, they presented information about the most qualified candidates to local parties in order to close the information gap.

But because of the sparse internet access in these rural communities, IPA Sierra Leone found that their pilot study struggled with significant inefficiencies. The enumerators had to re-enter a candidate’s information each time they were nominated, then manually count each entry in order to find the most popular candidates. Without a reliable internet connection, enumerators were forced to duplicate work and risk inaccurately assessing their results.

The Solution: using offline dataset publishing to pre-fill data

IPA Sierra Leone streamlined their workflow by utilizing the offline dataset publishing functionality included among SurveyCTO’s advanced offline features. Offline dataset publishing allows enumerators to save survey results to a local dataset on their phone, even without internet or server access.

With offline dataset publishing, enumerators could save each candidate’s information to the offline dataset the first time they were nominated — then, when they were nominated again, the enumerator could simply select their pre-filled information from a list rather than re-enter it all over again. This functionality simplified IPA’s data flow and automatically counted the number of votes each candidate received.

The Results: survey time reduced by half, 50% increase in survey respondents, 50% decrease in costs

IPA Sierra Leone found that by reducing the amount of duplicate entries in their candidate identification surveys, they were able to reduce their survey time from 15 minutes to 7 minutes. As a result, they saw a 50% increase in the number of surveys they were able to conduct in each community, while simultaneously reducing their respondent-related costs by 50%. This significant improvement in efficiency allowed them to effectively double their sample size at no extra cost, giving them a much more accurate picture of candidates’ popularity.

Reducing duplication also improved data quality. Instead of relying on fuzzy text-matching to count each candidate’s nominations, enumerators could trust the accuracy of SurveyCTO’s automatic count. IPA Sierra Leone was able to present this high-quality data with confidence to local political parties, allowing the parties to make well-informed decisions and connect with high-popularity rural candidates prior to the June elections.

"When we used the advanced offline features and the new data flow, we reduced our survey time from 15 minutes to 7 minutes, which led us to 50% more surveys per community. We basically doubled our sample size for free."