Our Approach

All of our projects begin with a tailored research design and a set of research questions, developed in collaboration with our partners to ensure it addresses their needs by identifying evidence gaps to be filled by citizen voice.  

We design our projects for impact. By this, we mean that we always have the ultimate objective in mind: for citizen’s voices to influence and shape policy and programmes towards their realities. As a result, we work closely with our partners throughout the research design, implementation, and insight generation, in order for them to contribute to and engage with the process, fully take on board our findings, and become advocates for our approach.

We spark dynamic and inclusive discussions in popular forums, including radio and social media which engage citizens on their own terms and language. Working with carefully selected and locally trusted media partners, we develop open-ended questions for audiences which they can respond to via free SMS, instant messaging, or social media. We consider how these questions will be processed and answered by audiences, then test and tailor for comprehension and cultural contexts. 

Follow-up SMS survey messages ask for socio-demographic information (e.g. gender, age, and location) or individual circumstances to allow for statistical comparisons between groups. Within these forums, opinions and beliefs are expressed spontaneously, resulting in authentic citizen-generated data.

Our approach allows us to gain insights from conversations held in local languages, and on a scale previously unattainable for qualitative methods. We leverage technology as an enabler for civic empowerment, using it to engage with, not extract from, citizens. We mix human qualitative research techniques with responsible applications of data science and AI to analyse the rich local language voices flowing from digital discussions.

In 2018 we launched Coda2, the second version of Coda, our software tool built in collaboration with the University of Cambridge to allow interpretation and effective labelling of large volumes of textual data. Coda2 leverages AI to augment the interpretive capacities of researchers rather than to automate data analysis, enabling them to listen to and interpret a larger set of conversations in local languages than they would have been able to unassisted.  Human interpretation is crucial as it allows us to draw insights from citizen voices, using natural curiosity, empathy, and compassion. 

A careful combination of qualitative and quantitative approaches allows us to increase both the scope and depth of our work and provide nuanced insights to our partners.   

Always leaning towards innovation and enhancing impact, we discuss and deliver insights to partners through workshops, interactive reports, and data visualisations. See our projects for example reports.