Uchaguzi

Quick Look

Uchaguzi (“decision” in Swahili) was developed to monitor Kenya's 2010 constitutional referendum.

Beginning Date: 
May 15, 2010
Ending Date: 
August 31, 2010
How many unique hits per month?: 
5000
Project Scale: 
Function: 
Specific Tools: 
Location:
Kenya

Context

Uchaguzi (“decision” in Swahili) is a follow-up project to the first instance of Ushahidi. Ushahidi was developed and launched during the 2007–2008 post-election violence in Kenya. The goal of Uchaguzi was to monitor Kenya’s August 2010 constitutional referendum.

While the initial instance of Ushahidi sprung up quickly in reaction to the 2007–2008 post-election violence, planning for Uchaguzi began well in advance. Months prior to the elections, organizers began talking to potential partners in the Uchaguzi project, including the Social Development Network (SODNET) and the Constitution and Reform Education Consortium (CRECO). CRECO trained 500 official election monitors, and Ushahidi trained them on how to report incidents to the Uchaguzi platform via text message. CRECO also had a partnership with the Interim Independent Election Commission (IIEC): when electoral issues were reported to Ushahidi either by the official CRECO monitors or by members of the public, these issues were reported to the IIEC.

Uchaguzi received over 2,500 messages during the referendum, some of which were forwarded to the IIEC for action. One example is that of a photo sent by monitors to Uchaguzi. Significant campaigning had been done using the color associations of green for “yes” and red for “no.” The photo was of a poster hung at two polling stations in which these colors were reversed. The error was reported to the authorities and the IIEC removed the posters.

Interview

Tell me a little about your project.: 

My name is Jessica Heinzelman, and I’m an Uchaguzi team member. The Uchaguzi platform really grew out of the Ushahidi team’s continued commitment to Kenya. The main core of this team contributed to the development of Uchaguzi, and many more new people were also involved. Months prior to the elections, we began talking to potential partners about the Uchaguzi project, including the Social Development Network (SODNET) and the Constitution and Reform Education Consortium (CRECO).

The goal was really multifaceted: to collect data, to understand what was going on in the country and also to be able to respond to it. CRECO, for instance, had trained 500 official election monitors, and we [the Uchaguzi team] trained them on how to use SMS to report incidents to the Uchaguzi platform. A lot of the reports that we received on election day were from these official monitors. When there were electoral issues reported by either the monitors or the crowd, thanks to CRECO’s relationship with the Interim Independent Election Commission (IIEC), these issues could be followed up on and taken care of. One example was that significant campaigning had been done using the color associations of green for “yes” and red for “no.” A picture of a misleading poster that reversed these colors, which was hung at two polling stations, was sent to Uchaguzi. The error was reported to the authorities and the IIEC removed the posters.

Uchaguzi also had the ability to flag reports of violence for action, thanks to an informal partnership with the Uwiano platform, which was running a separate short code to monitor conflict around the country. Uwiano had relationships with a variety of actors on the ground, ranging from the local provincial police to civil society organizations, local community groups, religious groups, and NGOs that could do mediation or whatever intervention seemed appropriate for the issue.

What's your vision for the project?: 

If you compare Uchaguzi in 2010 to the original deployment of Ushahidi in 2008, much is the same. People could still submit reports via the web and Twitter. Kenya’s blogging and the tech community played a role in reporting and supporting the platform. But this time we had the benefit of being able to publicize the project, and a short code was setup that allowed people to submit reports through text. In the Kenyan context, where mobile phones are much more prevalent than Internet connections, this gave more people the opportunity to report in. Many of the founding principles of Ushahidi held true: creating a space where information was free, open, where common Kenyans could, if they wanted to, participate and have their voices heard about what was happening in their world. So the philosophical vision carried through, but there were a number of new features both in the implementation and in the technical platorm. Most notably, in 2008, there weren’t response mechanisms set up, while with Uchaguzi we had partners with response capabilities and contacts that could follow up on the information we received to make sure something was done with it.

How does your work currently turn into offline change? : 

A lot of the feedback we’re getting is that there is still a difference between citizen monitoring and election monitoring. The crowd can give a lot of important information and rich, deep data about what’s happening on the ground, but there is still a utility for a trusted network of official monitors who can provide information about what is happening from an unbiased and scientific perspective. Ushahidi and its partners saw the value in both models, so what we did with Uchaguzi was to collect both types of information and separate it, visualizing messages received from the crowd with black dots and reports from officially trained election monitors through our partner CRECO with blue. We felt it was important that people had that added information about the data coming in, so they could sort through and decide how much weight to give to each report. One of the challenges of crowdsourcing, of course, is figuring out what is truth, what might be the results of intentional misinformation or exaggeration born from emotion or tension.

In addition to the issue with the misleading posters mentioned earlier, we saw concrete actions being taken based on information submitted to Uchaguzi at the Dobholm Primary polling station, where names were missing from the voter rolls. After intervention by the authorities, the individuals dropped from the list were allowed to vote. On incidences of violence, of which there were thankfully few, our team was in close communication with Uwiano team.

One of the things that we added to this platform was a check box that said “actionable” and another that said “action taken”, so we could go back and fill out the results from our information gathering and the responses from our partners; unfortunately, the police, the monitors and the different actors were so busy doing their jobs keeping the peace and observing the election that they often did not have the ability to feed that information back to us — it was not high on their priority list. Hopefully, we will be able to get more information and think through how to better incentivize closing the feedback loop for future deployments.

Overall, though, it is more about lessons learned for future planning and not really about a quantifiable impact at this point. It is hard to define impact when the election went so smoothly. For us right now, impact is more about setting the precedent and familiarizing Kenyans with these tools so that we can play a greater role should the situation turn south in the future. Thankfully, this time peace prevailed, and the elections were on the whole very well run.

What are the biggest obstacles to your success?: 

The Uchaguzi project was an overall success. We had a good number of reports that were really reflective of the situation: tension in some areas of the country, while others were peaceful. We may have gotten more reports had we started our publicity campaign earlier, but at the same time the reason why we did not get overwhelmed by too many reports is because the elections were so peaceful, there wasn’t much incentive for people to report non-incidents (most of those were coming from our monitors). I don’t think any of us would trade larger stats for a different outcome. As a team, it was a good exercise, and we’re doing a post mortem to run through what kinds of things we can improve for next time, so in that sense it was good too.

Has your work been replicated?: 

Ushahidi is a versatile tool that used by different organizations for a lot of different purposes. One of the most popular ones has been election monitoring. It started with small, grassroots groups that were interested in getting information from the crowd through citizen monitoring, but has now grown into something large. Internationally recognized and accredited elections monitoring organizations, such as IFES, are using Ushahidi for election monitoring in places like Burundi.

Is your work a replication of another project?: 

Ushahidi is a crisis-mapping platform born out of the post-election violence in 2008, when Kenya exploded into conflict. It allowed Kenyans to share information about what was happening around their country amidst a media vacuum and document the events as they occurred. After Ushahidi’s first deployment the founders were approached by different groups who wanted use the platform for data collection and aggregation of information for use in other crowdsourcing projects. In response, the team decided to form Ushahidi, Inc., which is now dedicated to improving the crisis-mapping software for use by groups around the world. While Ushahidi, Inc. focuses mainly on improving the technical platform, the founders all have strong links to Kenya and with the Ushahidi platform born out of the post-election violence, there is a continued commitment within the organization not only to Kenya but also to monitoring issues around elections. The Uchaguzi platform grew out of Ushahidi Inc.’s continued commitment to Kenya.

What transparency/accountability organizations do you work with?: 

We partnered with the Social Development Network (SODNET) and the Constitution and Reform Education Consortium (CRECO), and informally with the Uwiano platform.

Further Questions

How many reports did you receive overall?

We received 1,523 reports, and 794 of those were reports of peace. And that was wonderful.

Did you get in touch with mainstream media about publicizing the project?

We had ads on the front pages of the newspapers in the days leading to the referendum, a couple of radio spots, advertising posters in a number of malls, but we also benefited from having the trusted network of monitors and our partners Uwiano, who had a different short code and were getting a lot of crowd reports as well. Their platform was internal, not public and not visualized on a map, so they saw the benefit of having that component and shared some of their information with us.

Could you tell us a little bit about the people collecting and managing the reports and how it was organized?

We had 12 CRECO staff administrators purely dedicated to processing the election monitor reports. Then we had numerous Uchaguzi volunteers to process the crowd reports, about 70 volunteers in total, including our tech volunteers who helped us fix bugs as they came up. They worked from the iHub in Nairobi. It was very useful that most volunteers were Kenyan and had specific geographical and linguistic (Swahili) knowledge.

Are you satisfied with people’s overall awareness of the project?

Well, our focus was more on awareness of Kenyans and empowering them to report what they were seeing happening around them. We had 4,982 unique visitors on the website, but that wasn’t our main goal. Like the majority of Ushahidi deployments, a lot of visitors were from the US, where Internet access is more prevalent. The second highest country by views was Kenya.

Comments

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd> <img> <h1> <h2> <h3> <h4> <h5> <pre>
  • Lines and paragraphs break automatically.

More information about formatting options