What the red dots are for, or why we map (part 1: Iraq)

Mercy Corps has recently embarked on a strategy to introduce mapping tools to its conflict management and protection programs around the globe – from Iraq to Kenya to Nepal. Rather than develop a one-size-fits-all approach, Mercy Corps is looking to design context-specific mapping components that match the programming needs of different conflict management and protection projects. I’ve been involved in designing the systems for the first two projects to adopt mapping: the Iraq Conflict Resolution and Reconciliation program and the Libya Protection Program. The Iraq system will use Ushahidi and FrontlineSMS to collect data on mediated disputes and perceptions of conflict; the Libya system will use a combination of Google products to collect data on demographics, socio-economic conditions and conflict indicators at the IDP sites where Mercy Corps works.

These tools and methodology were chosen to match the analysis and programming needs of the Mercy Corps teams and their partners.  It’s been a very interesting design process, which has made me reflect on why (and how) we map. This is the first of a two-part post exploring some questions we can ask when choosing to map in support of peacebuilding and protection outcomes.

Understand the user: the Iraqi Centre for Negotiation and Conflict Management

Mercy Corps Iraq is developing a dispute resolution tracking and early warning system for its Iraq Conflict Resolution and Reconciliation (ICRR) program. The main purpose of ICRR is to support the Iraqi Centre for Negotiation and Conflict Management (CNCM), a network of Iraqi mediators who intervene in local disputes to prevent them from escalating. The network has been working since 2005, has a very strong track-record, and enjoys widespread support. The Mercy Corps team has worked with CNCM since its inception, and has an excellent understanding of how network members go about their work. With this in mind, the team set about providing CNCM with a mapping tool tailored to the mediators’ needs for information.

There is a wealth of information already available in the network. Most of the disputes network members are involved in are recorded in audio or written form, and stored by Mercy Corps. CNCM members receive frequent requests for dispute resolution and mediation through their personal networks, effectively providing them with an early warning system. The challenge in developing a better information system for CNCM was to find ways to systematize information collection and analysis, so that the network can proactively address emerging trends rather than reactively respond to requests.

Identify the main function: dispute mapping and early warning

The CNCM Dispute Mapping and Early Warning System will collect data on trends in the disputes handled by the Centre, data related to the conflict context in Iraq, and data on changing perceptions of people on key triggers of conflict. The collection and visualization of these sources of data will allow the ICRR team to better monitor past activities and better respond to new developments.

Monitoring: Data on disputes, systematized with date and location information, will allow ICCR staff to monitor the outputs of the Center, by understanding when, by whom, where and what disputes are mediated. Cross-referencing data on disputes with data on perceptions will allow ICRR staff to monitor the impact of the Center by understanding how disputes mediated affect perceptions over time.

Early Warning: Looking at how information on past mediated disputes compares to context data will allow the ICRR team to better understand how the conflict context affects disputes that arise, so they can work with Center members to better plan interventions. Tracking perceptions on triggers of conflict over time will give Centre members advance warning that tensions may be rising in an area, allowing them to intervene early.

Decide on data collection functions: web based and SMS

To fulfill its main function, the system needs to collect and visualizes three types of data:

  1. trends in the disputes handled by the Centre;
  2. changing perceptions of people on key triggers of conflict
  3. data related to the conflict context in Iraq

Mercy Corps staff already collect disputes data (1), and all that is needed is a more systematic way of organizing it. A web form that can later be edited as new information emerges or details change, and that systematizes some of the categories is sufficient. Context data (3) for Iraq is readily available, so the system just needs to be able to display it in a useful way.

Perceptions data (2), on the other hand, is a new source of data not previously collected. After some discussion, the team settled on running an SMS survey of perceptions of conflict triggers. This survey consists of 10 questions that are sent to a group of pre-selected key informants. Informants are asked to respond saying whether they strongly agree, agree, disagree or strongly disagree to the statements made in the question. Informants receive one text message each day for 10 consecutive days. The survey is re-run every two weeks, asking the same questions to the same informants.

Understand what analysis is needed: eye-balling patterns

The purpose of collecting disputes, perceptions and context information is to provide actionable data to help network members plan interventions that prevent disputes or tensions from escalating to conflict. To this effect, the ICRR team will look for spatial and category patterns in the data collected. However, given the nature of data collected (non-random, incomplete), the team will not use statistical tools: any conclusions drawn from the analysis indicate that an area deserves attention or further investigation, not that causality can be demonstrated. With this in mind, the team has two broad areas for data analysis:

Monitoring: Identifying spatial patterns helps draw out lessons from past experience in dispute mediation. Specifically, identifying clusters can help tailor training according to most pressing issues in certain areas. Looking for spatial correlations can provide context information for mediators who are intervening in a dispute. Finding disputes that tend to happen together might point to a larger (policy) issue that CNCM members can intervene on.

Early Warning: Finding patterns over time can help CNCM members act on real-time information. In the short term (when the system first starts running), focus should be on looking for spikes in indicators. When a spike in an indicator occurs, the relevant CNCM members should discuss whether an intervention is necessary to prevent a dispute. In the longer term, finding time correlations will help CNCM members identify perceptions that are leading indicators of disputes. These identified leading indicators are the early signals that system users should pay most attention to. CNCM members could then subscribe for alerts on identified leading indicators, which would provide real-time early warning that a type of dispute (linked to that perception) has a higher likelihood of occurring.

Choose your mapping technology: Ushahidi

The system will run on two separate Crowdmap platforms that share data. The Iraq Monitor platform will track disputes handled by Center members; the Iraq Warn platform will collect data on perceptions of conflict triggers (via a link with FrontlineSMS). Context data will be mapped as background on both platforms (where relevant), and data from each platform can become context information for the other platform.


The team picked Crowdmap because it meets its data collection needs closely and allows for user-friendly customization and management. The mapping functions are also adequate to the level of spatial analysis they intend to conduct. Crowdmap does not have a suitable graphing / data summary function, but given the number of reports and complexity of each report, this can easily be supplemented by using Excel. Crowdmap allows users to download data in CSV format, which can then be manipulated to show basic summary graphs and tables.

The final output – two Crowdmap platforms, a FrontlineSMS platform, a template Excel sheet for analysis and a user manual with a basic analysis tutorial and examples – gives the ICRR program staff and CNCM members a set of practical tools to use in support of the good work they already do. That’s what Iraq taught me the red dots are for: practical decision making, reflective learning and an exploration of new areas for intervention.


8 thoughts on “What the red dots are for, or why we map (part 1: Iraq)

  1. In terms of the Iraqi center, what I understand from your description is that you’ve got a network of mediators who are pretty good at preventing escalation of conflict and now you’re looking to use them as sensors for any emerging issues and early warning signs? Sort of like social workers going into the family and being trained well to detect any issues that may not be (and often aren’t) communicated verbally- they need to be in a heighten sense of awareness to detect issues. If im right so far, an idea comes to mind that may be useful in this context. Just a bit a disclaimer here- I could very well be in a mind frame where I have a hammer and everything therefore looks like a nail BUT we’re working with Cognitive Edge on using micro-narratives for a variety of development problems and this seems like this could fit the Iraqi situation well.

    The methodology involves collecting thousands of stories from people who are prompted to think about a situation but in an indirect way (e.g. you never ask them ‘what issue do you have with your neighbor’ rather ‘can you tell us about an experience from last week that frustrated or encouraged you’) and then the people who provided the story, interpret it themselves (as opposed to experts, anthropologists, any-one 3rd person who didn’t tell the story). By having a clever capture strategy (e.g. embedding it in a process that already exists… asking kids to collect stories as a homework assignment from their parents or grandparents; asking people to tell stories as they come to collect social welfare, etc), you get a continuous stream of narratives that give insights into the word on a street so to speak and that may allow you to detect weak signals, annoyances that you may be able to address before these become hardened. Anyway, GlobalGiving (http://www.globalgiving.org/stories/) has done some work on this as is Nominet Trust (they use it to monitor impact of their projects on youth: http://www.nominettrust.org.uk/knowledge-centre/blogs/evaluation-cognitive-and-digital-edge), and we at UNDP use it in 6 different contexts (can we detect source of resentment in local populations that live in protected areas; what facilitates or hinders citizen involvement in local level decision making; can be monitor social inclusion of Roma; etc). I know of at least 1 case where Cognitive Edge has applied this methodology in extremely security sensitive environment, so if there’s any interest, we can connect? Now I know this is way different than mapping, but I was inspired by your post and wanted to share 

    1. Hey Millie, thanks for the suggestion and for your interest in this project! I’ve looked at the use of this micro-narrative technique by Global Giving before – it’s very interesting and glad to hear UNDP is using it elsewhere! I don’t think it’s appropriate for the Iraq case at the moment, mainly because I’m guessing it’s quite a time-intensive methodology, which would be a problem for the Mercy Corps team that is spearheading this.
      On a separate but related note… I’ve just been doing some work with UNDP Sudan on their conflict prevention and recovery portfolio, and it strikes me this could be interesting for them. Just an idea to explore. Who would be a good person to talk to about this in UNDP? Maybe that’s you 🙂

  2. This is important work—-and the analysis part is something we have been working on—-to gather insights into “meaning”—-perhaps we could discuss how our work is complementary—-especially as you are focussing on qualitative, non structured complex data sets—-we’ve been introducing a process that can quickly “make sense” of that data. See http://www.inthum. org for a quick overview…..

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s