A Plan in Action

Thursday, June 19, 2014

After a few weeks of doing odds and ends for ObservaCoop we have finally landed our first large assignment for the summer: geocoding a portfolio European Commission projects in Mexico and using ArcGIS to create maps and a final report for the European Union. It was really exciting and somewhat intimidating at first, since we found out that we, as Summer Fellows, would be largely responsible for the execution of this project. Since none of the Fellows at Observacoop, myself included, have any real experience managing such an endeavor, the best we could do was smile and say “ya lo hacemos!” After all, it was going to be a learning experience for all of us – it would be our first time managing a project and it would be the interns’, our primary geocoders, first time geocoding.

It’s not so much the tasks that we need to do as the mentality. As far as I know, there are six basic components to the process from start to finish: geocoder training, the geocoding plan, geocoding, quality assurance, mapping, and a final report; each is pretty straightforward for those who have participated in geocoding for some time. However, given this experience, I’m a little nervous. Normally, geocoders have at least two to three months of training before they go “live,” or work on actual projects. The ObservaCoop interns, whom we trained, have only had three weeks to let anything sink in, and most of them have been occupied with other tasks at ObservaCoop, such as the design of our new website. It’ll be rocky at the start as they learn to apply the skills from their training to European Commission projects, whose documentation is a little different from that of the World Bank, which were used during training.

There are a total of 60 possible projects in the portfolio, but at this point we have only received documentation of any sort from 32 of the 60. Clay did a quick review of all the projects and estimated that we would have upwards of 1,000 projects to geocode, which can’t be too bad between the seven people (three AidData Fellows and four ObservaCoop interns) working full time on the project. However, after a more comprehensive review of the documents and a trial geocoding of one project, we noticed that we would have to supplement the original training that we gave the interns so that they could maneuver through the documents easily – while some EU projects adhered to the template that ObservaCoop distributed for data collection, others merely sent their annual reports or presentation slides for the coders to review and pull information from.

Additionally, we will have to discuss and standardize precision code procedures for Mexico. Precision codes are assigned to identify the political-administrative level of the location added to Toolkit – the smaller they are, the more specific the location. Generally these codes tend to be similar within a country. For example, in the U.S., a city would receive a 1, a metropolitan area would receive a 2, a county 3, and a state 4. Mexico, however, has some interesting political administrative levels between the various states. Most notably, Oaxaca has seven apparent administrative levels, four more than its neighbors. Additionally, Mexico City, as a federal district, is going to have juggle with its identity as both a city and a state. With the help of the interns, we’re currently reviewing state legal codes to determine the best separate the sub-state information for these special cases so that their precision codes align with those of the others.

That aside, however, the seven of us are set to begin the geocoding process starting next week and we hope to finish this portion of the overall project by July 12, giving us about three weeks to do everything. As these things tend to happen it will probably take a little bit longer than that, but if we can finish around mid-July we’ll leave ourselves plenty of time to make maps and conduct analysis of our data.


  1. Your post resonated with me because I have also found that one of the most important elements of any research undertaking is maintaing a positive, flexible attitude. Whether it’s interns with only a few weeks of training, or glitches in a database you need to use to collect information, there will always be obstacles that cannot be avoided or fully mitigated. It sounds like your team decided to dive in headfirst in spite of these challenges!