Able and Mostly Willing: An Empirical Anatomy of Information’s Effect on Voter‐Driven Accountability in Senegal

The forthcoming article “Able and Mostly Willing: An Empirical Anatomy of Information’s Effect on Voter‐Driven Accountability in Senegal” by Abhit Bhandari, Horacio Larreguy and John Marshall is summarized by the author(s) below. 

Electing competent and hard-working politicians is a pillar of democratic governance. It is often hoped that informing voters about their politicians’ performance may help citizens to select high-quality politicians and hold underperforming ones to accountAcademic research has made strides toward causally linking political information and accountability, though the evidence to date remains mixed. On one hand, the recent Metaketa initiative—which coordinated experimental projects in six countries to address this very question—found that making indicators of incumbent performance available to voters had little effect on their beliefs or vote choices. On the other, information’s pro-accountability effects are clearest in studies of broadcast, print, and social media.  

Given the complexity of the theoretical chain linking political information to voter-driven accountability, it is hard to know where the logic of accountability breaks down. Could limited effects be due to difficulties in disseminating information to voters? To voters’ inability or disinterest in processing information? To the lack of relevance of the information to voters? Or perhaps even to voters’ unwillingness to hold politicians to account?  

Furthermore, existing research has focused predominantly on electoral accountability – that is to say whether votes are cast against incumbent politicians that failed to meet expectations and for incumbents that exceeded them. However, citizen interactions with politicians once in office may be just as important, especially in the Global South where interactions with representatives often extend only to the months of an election campaign. Such contact represents a valuable opportunity to communicate an individual’s or community’s information, preferences, and requests to politicians.  

We designed a randomized field experiment around Senegal’s 2017 parliamentary elections to dissect the chain that connects the receipt of incumbent performance information to two forms of accountability: choices at the ballot box and requests of politicians after the electionPartnering with a Senegalese civil society association, we implemented the project across 450 villages in five rural Senegalese districts – places where information about legislators’ performance is otherwise hard to come by. Compounded by Senegal’s party-dominated system where citizens typically vote based on party rather than candidate in parliamentary elections, the low-information environment presented an opportunity to assess whether voters are willing and able to hold politicians to account based on information they receive. 

Over the month preceding the electionnine voters aged 20-38 per village could receive leaflets containing information about MP responsibilities, incumbents’ activity in parliament and local development policy outcomes, and/or information comparing the performance of the current incumbent to the previous one in a given district. To focus on how voters respond to such information upon receiving it, rather than whether the information reached or was in engaged with by voters, enumerators spent up to five minutes explaining the leaflet in person. Via baseline, endline, and follow-up surveys, we measured whether (and when) information impacted voter evaluations of the incumbentefforts to contact politicians, and actual votes in the community. 

Our varied treatment conditions, factorial experimental designs, and extensive measurement of citizen beliefs and behaviors enable us to anatomize the process through which information influences voter-driven accountability. Absent the dissemination challenges that may have affected previous studies, our study is able to examine which types of information voters care about, whether they internalize this information, and how – if at all – they use this information to hold their politicians to account.  

We find that rural Senegalese voters processed performance information in a sophisticated manner vis-à-vis their prior expectations, cared primarily about performance related to local development, and found information that compared their current incumbents with previous incumbents to be most informative. Their more positive evaluations of the best-performing incumbents endured for at least a month after treatmentIn contrast, providing information about legislator duties had little effect on voters’ beliefs or accountability-seeking behavior, suggesting that voters are well aware of politician responsibilities to bring resources back to their community.  

The information campaign also affected electoral and non-electoral efforts to hold incumbents to account. Voters durably requested greater contact with the best-performing politician after the election, and votes for such incumbents increased among likely-voters and the voters who prioritized local projects. In the often tight-knit communities that we studied, disseminating leaflets to nine younger voters was even sufficient to influence polling station level voting outcomes  

Upon receiving information, we conclude that voters are able and mostly willing to use information to hold politicians accountable. When the information provided to them is relevant, voters—even in rural settings characterized by limited access to education and information as well as low levels of political competition, where researchers might have least expected to have observed our resultsseek to hold politicians to account.  

By showing that accountability does not break down with voters in such contexts, which are all too common in the Global South, our findings shift the onus to improve accountability back to governments, donors, and civil society organizations. Such institutions may require more powerful incentives or greater capacity to disseminate accurate and relevant information that reaches a largely receptive electorate. Furthermore, our findings suggest a need for accessible fora through which enthused voters can communicate with their representatives.  

About the Author(s): Abhit Bhandari is Postdoctoral Research Fellow, Institute for Advanced Study in Toulouse at University of Toulouse Capitole, Horacio Larreguy is Associate Professor, Government Department at Harvard University and John Marshall is Assistant Professor, Department of Political Science at Columbia University. Their research “Able and Mostly Willing: An Empirical Anatomy of Information’s Effect on Voter‐Driven Accountability in Senegal” is now available in Early View and will appear in a forthcoming issue of the American Journal of Political Science. 


  1. I am delighted to see the American Journal of Political Science focusing on Senegal. A French American born in Paris, I lived in Senegal from 1988 to 1992, where I headed the World Bank office in Dakar. A major factor behind Senegal’s political vibrancy and awareness is the existence of a free and active press, with some of its many titles mimicking the French press, such as the “Canard Enchaine.”

Speak Your Mind



The American Journal of Political Science (AJPS) is the flagship journal of the Midwest Political Science Association and is published by Wiley.

%d bloggers like this: