Four Lessons Learned Collecting Data in Venezuela During a Global Pandemic
Back in March 2020, Premise Data was the first solver organization awarded under the BetterTogether Challenge. Premise aimed to help local organizations in Venezuela make better decisions and deliver more responsive humanitarian aid by using real time data collected through the Premise app. But just as Premise kicked off the project, the world changed. As a result, key aspects of the project implementation, and in particular how data collection was conducted, had to be adapted. Here are four lessons learned from the experience crowdsourcing data in Venezuela during 2020 in the midst of the COVID-19 pandemic.
1. Prioritize health and safety first!
Premise, through its data-collection app, originally planned to send its contributors to public spaces to collect data that could help map out local conditions (i.e. availability of food or medicine). But when COVID-19 hit, the strategy changed right away to designing surveys that could be answered by everyone from their homes, without putting them at risk of contagion. The pivot was immediate. The team redesigned the strategy to focus only on tasks that could be collected from everyone participating in the app in a safe way.
2. Be empathetic to the local context, using culturally appropriate language and traditional references.
Localizations of the survey instruments should not only be about direct translation but adapting into local contexts.For example, Premise uses a language that is friendly and approachable so that anyone reading the question while completing the online application in Venezuela couldn understand the meaning. Premise also adapts specific questions to the local customs and place so that it is relatable.
3. Be inclusive!
To make the data relevant, it needs to be inclusive of the diversity of the local population, ensuring women, and marginalized groups are represented. Data is only relevant if it represents all individuals of a specific population in the sample. Historically, national and international datasets in the humanitarian space have offered data over-representing specific populations, such as men, which has posed a big problem. To tackle this, Premise Data collected responses from a representative sample per state and 50% of the responses per locality came from women contributors.
4. Be receptive to feedback from your users in order to best meet needs.
Premise keeps an active team responding to all contributors' questions incoming through the app and alternative channels, such as Facebook, Instagram, and Whatsapp. We constantly learn and adapt to keep them engaged. For example, Premise prioritized reading and addressing messages from contributors who reported how to improve specific features of the app. One of the frequent messages received was to add more tasks and surveys to complete! Also, once it was safer for contributors to go to public spaces, we mapped out unprotected open water sources in specific locations, to ensure that our contributors would be in constant communication when the locations we were sending them to had nothing to report in the area. This allowed the team to refocus and redesign the strategies for this specific task, resulting in a very compelling water sources mapping that is now being used by national and international organizations.