We convene thought leaders and leading practitioners to discuss ideas and act on them collaboratively.
Creating closed feedback loops requires a variety of tools and organizations at various stages. Because of ourgrowing network of member organizations, we are able to stimulate and facilitate both conceptual as well as operational collaboration among organizations.
Convening is done through LabStorms, BackTalks, the annual summit, and collaborative experimentation.
Feedback Labs is facilitating experiments to learn what can improve citizen engagement and ultimately help close the feedback loop. One experiment uses innovative technologies that allow the environment to "speak", thus providing data that could galvanize citizen engagement. The other experiment seeks to understand whether mobile phone-based surveys can be used for statistically representative information collection.
In this experiment, environmental sensors are being deployed to measure air quality in urban transportation centers in Brazil.
While air quality sensing has been around for a while, most sensors require an Internet signal to report data, as well as fixed power infrastructure-- two parameters that are not always available. For this experiment, a prototype sensor was developed that overcomes these barriers by using Short Message Service (SMS text messaging) to transfer data and solar panels to meet energy requirements.
For the Feedback Labs community, this experiment is exciting because it helps us understand two things:
Can technology help the environment to "speak"? Will the remote sensors allow for timely and accurate data collection?
Under what conditions is information help empower citizens? Will the information about air quality help mobilize citizens and CSOs to seek change? Will the information pressure the government to take action?
Mobile Phone Surveys in Poor Countries
100 times cheaper, 10 times faster than a traditional household survey
Traditional household surveys in developing countries are important for getting information about citizens. Unfortunately, they are also time-intensive and costly. In this case, the experimenters used mobile phone-enabled surveys to try something faster and cheaper: Interactive Voice Recognition (IVR) to contact citizens in Afghanistan, Ethiopia, Mozambique, and Zimbabwe.
The experiment asks:
Can mobile phone survey platforms reach a nationally representative sample?
To what extent does linguistic fractionalization affect the ability to produce a representative sample?
How effectively does monetary compensation impact survey completion patterns?
For the Feedback Labs community, this experiment is exciting because it shows how listening to citizens can be done quickly and cheaply, in a way that potentially allows aid and philanthropy to adapt and iterate more nimbly to the on-the-ground realities of project implementation. For the full working paper from this experiment, please see the Center for Global Development's working paper.