Skip to main

Impact case study database

The impact case study database allows you to browse and search for impact case studies submitted to the REF 2021. Use the search and filters below to find the impact case studies you are looking for.
Waiting for server

Africa’s voices: Strengthening citizen participation in governance and social programmes in East Africa

1. Summary of the impact

Research on digital technology and political communication by Dr Srinivasan and his team led multiple development agencies in East Africa, including the United Nations Children’s Fund (UNICEF), the United States Agency for International Development (USAID) and the World Bank, to integrate citizens’ views in their strategy and operations. Srinivasan co-founded a spin-out charity, the Africa’s Voices Foundation, in order to translate research on interactive radio into a mixed-method approach through which governments and aid agencies could increase citizen participation. This charity has continued to draw upon the research in its subsequent work, and the research approach is now used extensively in Somalia and Kenya. Africa’s Voices was awarded the Market Research Society’s President’s Medal for Social Impact in 2016.

2. Underpinning research

Since 2010, Dr Srinivasan, Lecturer in Governance and Human Rights at the University of Cambridge, has led an interdisciplinary research programme that includes political and social scientists, linguists, and computer scientists. Their research demonstrated the importance of interactive radio shows for deepening civic engagement in East Africa. They developed new data collection techniques to process high volumes of digital text from radio shows, and mixed methods social research approaches to interpret data and apply their findings in policy contexts. The research had four phases.

Phase 1: Identification of an important but under-studied phenomenon (2010-2011)

In 2010, Srinivasan identified a gap in knowledge about a popular socio-political phenomenon in many African countries: interactive call-in or text-in radio shows. A University of Cambridge study led by Srinivasan found that citizens actively engaged with one another and government through interactive public affairs shows in Kenya, Zambia and Uganda [R1]. It also found some stations were better than others at curating discussion with diverse citizens and policy makers.

Phase 2: Determining how social and political factors affect show popularity (2011-2015)

Comparative research in Kenya and Zambia, led by Srinivasan, with postdocs from the University of Cambridge and partners from the Universities of Zambia and Nairobi, discovered when and how interactive broadcast shows might enhance political accountability. They found that the popularity of interactive shows resulted from authorities, presenters and audiences investing them with different meanings [R2]. Audiences participated more when they experienced the show as a constructive social interaction [R3]. Also, the show format affected the scale and dynamism of participation. These factors explained why many formulaic approaches to interactive shows on governance and development failed to generate active and diverse engagement. They showed that it was necessary to design shows that take into account how and why actors participate [R3].

Phase 3: Designing a methodology to study public opinion (2012-2015)

Abreu Lopes and Srinivasan implemented a quasi-experimental pilot project in eight countries to test ways to generate participation and analyse data using interactive radio. There were two main findings [R4]. First, the mode of asking questions affected participation (e.g. it increases when audiences can provide an open text explanation in their own language). Second, shows had to be analysed as a space for social interaction, similar to a focus group. Researchers also confronted a methodological challenge: manual coding could not readily be scaled up for multiple languages and large data volumes. In 2014 Abreu Lopes was seconded to IBM Research-Africa to test how computational linguistic techniques might address this issue.

Phase 4: Application of social and methodological findings to policy (2015-2020)

In late 2014, Srinivasan co-founded a spin-out charity, the Africa’s Voices Foundation (AVF), to utilise the research to assist policy actors in civic engagement. Research has since been conducted collaboratively between the University and AVF. This included technologists (Blackwell, Church) and computational linguistics (Buttery, Caines) at the University of Cambridge. They focused on improving the quality and speed of data processing, and designing tools to interpret large volumes of complex textual data. In 2016, with Srinivasan, Blackwell led a team in developing a new open-source software called CODA. CODA uses human-computer interaction design for more efficient and reliable qualitative coding of text data at scale [R5]. The team’s earlier research, on designing mixed methods to analyse open text data, guided the software’s development in order to preserve the richness of citizen contributions. In 2017-2019, Srinivasan, Riha, Abreu Lopes and Denyer-Willis evaluated the interactive radio method’s efficacy in health emergencies.

3. References to the research

[R1] Srinivasan, S. and Abreu Lopes, C. (2016). Africa’s voices versus Big Data? The value of citizen engagement through interactive radio. In O. Hemer & T. Tufte (Eds.), Voice & matter: Communication, development and the cultural return (pp. 155-171). Nordicom. ISBN: 9789187957314.

[R2] Srinivasan, S. and Diepeveen, S. (2018). The power of the “audience-public”: Interactive radio in Africa. International Journal of Press/Politics, 23(3), 389-412. [DOI]

[R3] Srinivasan, S. and Abreu Lopes, C. (2020). Mediated sociability: Audience participation and convened citizen engagement in interactive broadcast shows in Africa. International Journal of Communication, 14(2020), 2985-3006. [Link]

[R4] Abreu Lopes, C. and Srinivasan, S. (2014). Africa’s Voices: Using mobile phones and radio to foster mediated public discussion and to gather public opinions in Africa. CGHR Working Paper 9. CGHR, University of Cambridge. [Link]

[R5] Blackwell, A., Church, L., et al. (2018). Computer says ‘don’t know’ - Interacting visually with incomplete AI models. In Proceedings of the Workshop on Designing Technologies to Support Human Problem Solving DTSHPS’18 (pp. 5-14). Lisbon, Portugal. [Link]

Indicators of quality: [R2, R3] are articles in journals of international standing that have been through rigorous peer-review processes. [R1] is a peer reviewed book chapter, [R4] is a working paper that underwent internal quality review, and [R5] is a conference proceeding. Threshold quality is also evidenced through support from several external competitive grants, including:

Sharath Srinivasan, PI - Broadcast media, ICT-generated public opinion and political accountability in Africa (ES/J018945/1) - ESRC-DFID - 2012-2014 - GBP246,457.

Sharath Srinivasan, PI - Socio-cultural factors in Zika virus outcomes in Lusophone African countries - Wellcome Trust -2016-2017 - GBP242,642.

Alan Blackwell, PI - Visual Analytics for African Languages - Research Fund Institutional Grant 2016 - EPSRC Global Challenge - 2016-2017 - GBP39,167.

4. Details of the impact

Through the Africa’s Voices Foundation (AVF), a spin-out charity that applies the research findings, the research has enabled aid actors and governments to better solicit and understand citizens’ views in Somalia and Kenya, and be more responsive to citizens’ needs. The method is effective in contexts where in-person research is unfeasible or expensive, including remote or conflict-affected areas, and humanitarian emergencies. In 2016, the world’s leading professional social and market research association, the Market Research Society (MRS), awarded AVF the President’s Medal for Social Impact. The MRS President said, “ In a country with insecure and inaccessible regions, Africa’s Voices has developed an admirable partnership with UNICEF Somalia. Their work helps to amplify the voices of hard-to-reach communities and is proof positive of research as a force for good” [E1].

Srinivasan, with Lord Cairns, established AVF in late 2014 as a non-profit organisation and registered UK charity (no. 1159589). Srinivasan was seconded to AVF from 2017-2019 to ensure that it utilised the research effectively, and remained involved as an advisor since. Utilising the University of Cambridge team’s research, the charity offers a unique service: it convenes large-scale discussions among citizens using interactive radio [R2][R3], and deploys a mixed methods approach to analyse large numbers of SMS contributions [R5]. AVF then provides qualitative and quantitative analysis of citizen perspectives to governments and aid agencies.

Based in Nairobi, AVF has grown into a sustainable charity with an annual programme income of GBP1,400,000 (2019), and a staff of over 20 [E2, p.144]. AVF has worked with over 15 development actors, including: European Union (EU) International Organization on Migration, Mastercard Foundation, Oxfam Kenya, Trocaire, UNICEF Somalia, United Nations Development Programme (UNDP) Somalia, United Nations Office for the Coordination of Humanitarian Affairs (OCHA), USAID, and the World Bank [E2, p. 41, 102, 104, 141-2]. AVF’s contribution to their programming and strategy is evidenced in its ongoing work in Somalia, detailed below.

  1. Greater inclusion of citizen views in UN and World Bank annual planning

AVF assisted UN agencies and the World Bank to incorporate citizen engagement into planning processes. Since 2018 UN Somalia’s Humanitarian Needs Overview (HNO) (the UN’s major annual planning document in Somalia) has relied on AVF’s approach. The 2019 HNO [E3] stated that feedback through interactive radio enabled them to adjust activities in response to people’s concerns. It affirmed the value of AVF’s approach to incorporating citizen voices to their planning:

"Interactive radio programmes and SMS messaging by the Africa’s Voices Foundation additionally garnered feedback from close to 9,000 individuals across every region in Somalia. A very high proportion of respondents (87%) indicated that they felt the consultations had made them feel more included in decision-making, and the same proportion further reported that they would like to see this process repeated." [E3, p. 16]

AVF also contributed to the 2020 HNO, evidencing changes in citizen views [E4, pp. 11, 26-28].

In 2019, AVF helped the World Bank to review its strategy in Somalia by convening interactive radio shows and analysing public views. The World Bank Country Manager asserts about AVF,

“*Their work tackles an important challenge for the World Bank in Somalia and many other country contexts: namely inclusive approaches to civic engagement and rigorous methods for drawing insights from citizens into more accountable and participatory policy planning processes.*” [E5]

AVF’s approach led to internal reflection in the World Bank team. The Country Manager states:

“As a direct result the team is re-considering the importance of health programming and community engagement. This kind of intervention is a game-changer for civic engagement efforts in Somalia.” [E5]

  1. Innovation in social programme design, including in health emergencies

AVF’s interactive shows and analyses contributed to important innovations in aid agencies’ social programming. AVF assisted UNICEF Somalia in better engaging target populations and gathering detailed social insights. In 2016-2017, AVF designed and delivered more than 25 shows on 21 radio stations for UNICEF Somalia. Over 52,000 Somalis participated. AVF’s insights on citizens’ views informed internal discussions in the development of UNICEF Somalia’s 2018-2020 Gender and Child Protection programme strategy [E2(b)]. UNICEF Somalia’s 2017 Annual Report attests to AVF’s contribution:

“*Africa’s Voices Foundation utilized a semantic analysis tool for capturing and categorizing the responses received, producing several short analytical products around the recipients’ insights on behaviours and key themes. This feedback has informed the development of a new targeted strategy for behaviour change for the different sectors which will continue to be implemented in 2018.*” [E6, p.11]

The rigour of the analysis was key to its adoption. The UNICEF Chief of Child Protection wrote:

“The Africa's Voices programme is one of the most articulate pieces of research I've seen on social norms in Somalia. It is the first time I've been given the tools to start applying this kind of scientific rigour to social norms programming and I'm finding it quite exciting and thought-provoking. The insights should be critical in our strategies for UNICEF's social norms pillar in 2018.” [E2(a), p. 30]

The Somalia Stability Fund (SSF), a large multi-donor fund, used AVF’s approach to achieve more inclusive participation in its programmes. In 2019, SSF invested USD1,000,000 in AVF to deliver a nationwide gender equality and social inclusion programme. The SSF Head of Risk, Finance and Procurement identifies two changes that occurred as a result [E7]. First, SSF achieved a greater scale of engagement with over 900,000 Somalis listening (according to an independent representative telephone survey) and over 53,000 Somalis participating. The show’s design [R2] was key to this. Second, data analysis of audience contributions was new, and used to tailor communication to encourage gender equality and social inclusion. A testimonial on behalf of the SSF Team Leader states:

“What makes this programme truly innovative and a step beyond other interactive media programmes that SSF has funded previously is that the data generated from citizen engagement is analysed and allowed AVF to identify key issues for audiences and weave these back into programmes in a way that positively challenges audience perceptions about gender norms.” [E7]

Finally, AVF’s support for early COVID-19 responses evidenced the adaptability of the research methodology to support programming. The team’s research on how to use it in public health interventions enabled AVF to move quickly to support COVID-19 risk communication in Somalia and Kenya. In April 2020, AVF collected and analysed over 7,500 individuals’ views on COVID-19. It identified that stigma and misinformation were widespread, and addressing these required working with trusted religious authorities. Based on this work, SSF and UNICEF commissioned AVF to conduct further COVID-19 programming in Somalia, responding to their earlier findings and bringing religious leaders and content into radio discussions [E8].

  1. Strengthened public accountability through improved communication

AVF adapted its application of the research to respond to donor and government needs. A key innovation was the creation of a Common Social Accountability Platform, which was designed to improve citizen-authority relations through interactive radio and analysis. The EU, the Horn/East Africa-wide Regional Durable Solutions Secretariat, International Organization on Migration, UNICEF, UNDP, UN Habitat and the World Bank utilised this platform in Somalia to foster better citizen-authority relations. The Durable Solutions Coordinator in the Integrated Office of the UN Resident Coordinator for Somalia explains:

“The work conducted with AVF showed clearly that, through radio shows and … text messages free of charge, groups that traditionally are excluded and marginalised in the public arena have the possibility to express their opinions on decisions that affect their lives. … We have had a very enthusiastic and positive feedback from local government actors who indicated that this methodology has enhanced their insight into the perceptions of the public on their work.” [E9]

For example, AVF worked with the Banadir Regional Administration, federal authority for the Mogadishu region, to address the city’s displacement crisis. AVF identified ways that people thought that authorities might better support vulnerable citizens, such as ‘crowd-funding’ - a salient proposal in a society dependent on remittances but which had not featured much in prior programming [E10]. Citizen engagement also helped to strengthen citizen-authority relations. In a post-project evaluation, a female participant underscored this impact:

I feel involved because community consultation is always the best thing to do. I personally believe that I am part of the decisions in the community and we very much appreciate those who made this safe space to talk, like the radio presenters, the leaders involved.” [E10, p. 8]

Donor, government and citizen feedback to AVF identified a shared interest in more interactive citizen engagement with authorities. In 2019, Srinivasan and Church decided to create a social business with AVF, katikati, to expand technological capacities in [R5] to manage two-way SMS conversations at scale in any language. These University of Cambridge researchers designed and developed the technology, and worked with AVF to apply it to governance challenges [E8]. In late 2019, they won a global impact award (USD150,000) from Twilio to develop an open-source capability for humanitarian organisations to use the katikati technology in global crisis response [E11]. In mid-2020, AVF deployed katikati in Kenya and Somalia to respond to stigma and misinformation about COVID-19, in collaboration with UNICEF, UNDP and SSF [E8].

5. Sources to corroborate the impact

[E1] Testimonial: Market Research Society President, 4 December 2016 (included in email from Events & Awards Manager).

[E2] Reports: Africa’s Voices Foundation annual and programme reports, 2017-2019: a) Africa’s Voices. Annual report 2017 (pp. 1-40); b) Africa’s Voices/UNICEF. (2017). Beliefs and practices of Somali citizens related to child protection and gender . Findings from Africa’s Voices Foundation’s interactive radio programmes aired during January and February 2017 for UNICEF Somalia (pp. 41-94); c) Africa’s Voices. Annual report 2018 (pp. 95-122); d) Africa’s Voices. Voice. Dialogue. Evidence. Action. Annual report 2019 (pp. 123-145).

[E3] UN/OCHA. (2018). Humanitarian needs overview 2019: Somalia. [Link]

[E4] UN/OCHA. (2019). Humanitarian needs overview 2020: Somalia. [Link]

[E5] Testimonial: World Bank Somalia Country Manager, 4 February 2020.

[E6] UNICEF. (2017). UNICEF Annual Report 2017: Somalia. [Link]

[E7] Testimonial: Head of Risk, Finance and Procurement, Somalia Stability Fund, 7 February 2020.

[E8] Testimonial: Executive Director, Africa’s Voices Foundation, 30 January 2020.

[E9] Testimonial: Durable Solutions Coordinator, Durable Solutions Unit, Integrated Office of the UN Resident Coordinator for Somalia, 22 January 2020.

[E10] Africa’s Voices. (2019). Common social accountability platform: Results and findings from citizen-led discussions on displacement and durable solutions in Mogadishu. (In partnership with the Regional Durable Solutions Secretariat and the Banadir Regional Administration). [Link]

[E11] Blog: Reilly, E. (17 December 2019). Twilio.org awards nearly USD3.7 million in grants to 26 nonprofits to support crisis communications. Twilio. [Link]

Additional contextual information

Grant funding

Grant number Value of grant
206015/Z/16/Z £246,457
EPSRC GCRF Institutional Funding £39,167
206015/Z/16/Z £242,642