Rapid mobile data collection training with Open Data Kit and Formhub in Liberia

Check the full post - including the amazing pictures - on terra incognita on wordpress. 

This post was originally posted on the Formhub blog in October 2012. I have made some changes to the text and added some pictures of my own.

As a part of the project to support the Liberian government in the preparation of their energy sector Master Plan, Modi’s Lab conducted a two-week training to prepare key government staff from four different Liberian government organizations in the use of mobile rapid data-gathering tools to estimate the urban and rural energy demand in the country. I had the opportunity to join the team to support on this 2-week long training in Monrovia.

After the end of the civil war in 2003, the Liberian power sector was largely destroyed due to warfare. Currently less than 1% of Liberia’s population has access to electricity. With the exception of independent power producers (IPP) and limited municipal mini-grids in Robertsport, Gbarnga, and Sanniquellie, Liberia’s rural (non-Monrovia) population has no access to electricity. Access to modern energy will be a critical element if Liberia is to restart the process of economic and social development. The cost to develop the sector is expensive, minimizing this costs is a complex problem that requires taking into consideration different technologies available, different network architectures and the penetration level within a community is essential to planning. To identify the optimal solution to this problem it is essential to gather sufficient spatial data to estimate the residential, commercial and industrial spatial power demand.

For two weeks, 20 government employees from the Ministry of Lands, Mines and Energy (MLME), the Liberian Institute of Statistics and Geographic Information Services (LISGIS), the Rural and Renewable Energy Agency (RREA) and the Liberian Electricity Company (LEC) participated in a practical training that allowed these different agencies to understand their different data needs and collaborate in the design and authoring of expert surveys. The intensive 8 day training covered all aspects of Formhub, the mobile rapid data gathering technology adapted by this Earth Institute engineering lab that focuses on developing and implementing engineering solutions for the developing world. The training included the excel syntax to author mobile surveys, the use of Android phones to conduct surveys and the steps to clean, process the resulting information and analyze it with the use of analytical software such as GIS and statistical software packages.

The practical nature of the training allowed government institutions to select a current need for collecting data, defining the set of questions required to study the issue at hand and pilot the survey to test it before making a full scale data gathering effort. For example, RREA designed a survey to understand the energy needs of rural populations in their use of charcoal and other biomass to help plan the introduction of fuel-efficient cook stoves.

Other efforts where aimed at understanding the operation of Independent Power Providers, entrepreneurs that sell electricity produced by diesel generators. Because the government has not been able to provide energy to all Liberians, IPP’s cover a very important need, but because of the unregulated nature of their business usually provide a service that is expensive, unreliable and that does not contribute to the extension and improvement of the power grid of the country. Piloting the survey in the field with real IPP’s allowed the different teams in the field to grasp the use of this new technology but also to understand the problems that enumerators face on the field and modify the survey to better respond to the data analysis needs.

Doing a pilot was very illustrative for the government officials but also for the Columbia University team. The scale of operation of some of these power producers goes well beyond what we can consider a micro-grid. Some had a customer base of over 150 customers, with circuit breakers to cap energy consumption to usually 1 amp but up to 10-15 amps for businesses. These IPP are really making a profit by charging flat-fees of up to 45$ per Amp, the scale is large and there is absolutely no use of safety standard for the wiring or the equipment.

During the training we were exposed to the common difficulties that Liberians experience everyday. Common power outages and tension spikes affected the equipment and created some minor but bothersome disruptions several times a day (try to charge 40 Android smartphones with power outages every hour). Every time the A/C stopped working, I would brace myself for the heat that 25 people with their respective laptops would release in a small conference room. Low bandwidth, recurrent inaccessible Internet forced us to be creative in our methods, pushed our resourcefulness to the limits.

What was interesting was putting the whole technological platform to the test of least developing country conditions, many innovations in the field of ICT4D work great in the big cities of the U.S. but fail completely as soon as you try them in least developed countries. We are proud to announce that Formhub was successful in facing the challenge of functionality under stressful situations (reduced bandwitdth, power shortages, no cell phone signal).

The last couple of days we visited each one of the organizations to make sure of what data exists and which are the data gaps. All institutions were provided with as many as 9 Android smartphones to start piloting their surveying efforts. The utility company wanted to start working right away on their Cross Border Needs Assessment, 18 communities close to the border with Ivory Coast will have access to power coming from the neighboring country. Understanding the power demand of such rural communities will be extremely valuable to parameterize the Earth Institute modeling efforts and improve the accuracy of the different scenarios for the energy grid planning.

Although data collection is a fundamental step for planning, the analysis of this data will also be a challenge, capacity building of these institutions in this area will be fundamental to truly make a difference in access to modern energy services in Liberia.

This technology has a very strong potential to truly increase work productivity in several ways. Not only is the possibility to obtain field data in large amounts real. Indeed any project can benefit from accurate planning based on actual facts. On top of the long term benefits the technology also permits to reduce the efforts of conducting a survey. For example, data clerks that digitize the paper surveys results can now be sent to the field to collect more data.

During our stay in Liberia we learned that the World Bank was able to collect information on all water points across the nation using 70 LG Optimus Smartphones during 30 days, a project called FLOW. The Formhub platform is still being improved everyday and Columbia University is hoping that it will allow developing countries to make large national infrastructure facilities inventories possible with very little investment.

Open Data Kit and Formhub are both Open Source tools, if you are interested in learning more about this tool please refer to the following short presentation to get an idea of what you can do with this amazing tool. After my return from Liberia I helped Modi’s Laboratories to improve it’s online tutorial.

Peak children - On the future of the planet population.

Check the full post on Wordpress

Hans Rosling illustrates us on the realities of the world. Dividing the world in developed and developing countries is an obsolete categorization. No matter the religion or ecosystem in which societies live in, there has been a reduction in the number of children per woman in the world. Although population is growing exponentially, Hans Rosling explains to us how he knows that the world population will stop exponential growth: while we are still debating peak oil, the world has reached peak children…

Pretty amazing right? There is also the full version of this explanation, why is it inevitable that we will be 10 Billion in three generations? Hans Rosling explains in an admirable way on what he calls the filling gap. In any case he destroys the wrongly constructed perception that population growth has anything to do with religion, it has to do with children mortality, labor productivity per worker and regulation on child labour, women education and possibility to join the labour force and of course accessibility to family planning.

As a strong believer in the use of reliable data and facts to dispel or confirm opinion based ideas, I think Hans Rosling educational videos should be translated to every language and shown to every secondary student in the world. Look at the shortest TED talk ever and how he manages to explain the idea of economic convergence in less than 60 seconds…

A green revolution needed in Haiti

Link to my latest post on Wordpress A few months ago, I had the opportunity to visit Haiti as part of an Earth Institute visit to meet some of the newly appointed government officials from Haiti’s new Prime Minister Laurent Lamothe. Our main objective was to gain some on-site understanding of the main bottlenecks in the agriculture sector. There are a lot of studies and papers about this subject but seeing the rice fields with your own eyes adds a new dimension to any desk research.  I was lucky enough to have professor Glenn Denning leading the mission. He has a very long and successful career as a rice specialist in Southeastern Asia and his coaching during our travels in the country allowed me to quickly grasp the agricultural policy situation in Haiti. Haiti’s mountainous landscape limits the area that is suitable for investments in the intensification of agriculture.  The Artibonite valley, in the central plateau, is Haiti’s “rice bowl”, producing up to 80% of the national rice production. Thanks to the construction of the Peligre Dam in the 1950s and a major irrigation system financed by German aid in the 1970s, the area has the potential to secure the country’s rice stocks and reduce food insecurity.

The Haitian government created the ODVA (Organization for the Development of the Artibonite Valley) in 1949 to take advantage of this area’s potential and to start what could be called a “Haitian Green Revolution”. Almost sixty years later and despite the relentless efforts of the organization to build and maintain water management infrastructure, the area produces yields comparable to what other tropical countries achieve with traditional rain-fed organic agriculture. The organization provides credit and in-kind transfers of basic imported inputs like seeds and fertilizer; it supports farmers in the acquisition of small mechanization tools. With these extension services and the irrigation infrastructure provided by ODVA the yields in the region should be up to three times higher for certain crops like rice. To better understand the reasons of these shortcomings we spent a full day in the Artibonite Valley visiting farmer fields, irrigation canals and meeting with ODVA staff. We learned that one of the main issues that the organization is facing is the lack of funding to maintain infrastructure. One of the key elements is the inability of the organization to charge for the water it provides farmers with. The effective full subsidy of water resources, covered by ODVA,  has critically limited the financial capacity of the ODVA to maintain secondary and tertiary canals in the valley. Without these canals, effective drainage and water management at the plot level is impossible causing yield levels  to remain stagnant.  This situation is worsened by the fact that the water comes from an upstream dam managed by EDH (Electricity of Haiti) whose work is totally uncoordinated from the needs of farmers in the valley. The Peligre dam represents more than 25% of the power generation capacity of the country and is fundamental to the country’s functioning; as a result water discharge is not synchronized with the agriculture needs in the region. The irrigation system provides either too much or too little water during the year. Another aspect to bear in mind is that imported rice is highly subsidized or even free, making the profitability of local production inexistent and forcing farmers to produce rice only for subsistence purposes. If all these factors weren’t enough, one needs to consider the land tenure issues in Haiti, having a higher-than-average productivity might lead to claims of land ownership by other farmers that could actually put in jeopardy any investment in the plot irrigation infrastructure. There are just no incentives for farmers to invest in productivity increases. Supporters of market-based mechanisms argue that part of the solution to this inefficient situation would be to charge farmers for the water delivered to their plots to build and maintain secondary and tertiary canals. Another important aspect would be to reduce the subsidies in the rice market to boost local production, a protectionist measure that Haiti’s commercial partners probably wouldn’t support. Both options would mean political suicide given the current political deadlock in the country. The reality is that the very precarious situation of Haiti has created a vicious cycle in the agriculture sector. There are absolutely no incentives for the private sector to invest in Haitian agriculture and there are very important pressures by the donor community to avoid any kind of important government intervention. Our job and the reason the ministry of agriculture has requested technical assistance from the Earth Institute is to support the ministry’s argument that the donor community has to invest in the sector having the government of Haiti act as the implementer of the projects. In this sense, the government needs to understand how much investment is needed in the sector to achieve yield levels that are in line with the two or even three times larger yield levels that similar Caribbean countries have already achieved. The Earth Institute has been working with yield, production and fertilizer data to model the possible economic impacts of strategic agricultural interventions to boost productivity. Factoring this political gridlock into the models is impossible; economists often avoid theses types of situations by making certain key assumptions to avoid the hard realities that occur in the development arena.  As a recent graduate from the MPA in Development Practice from Columbia University, I recognize that any form of effective policy recommendation will require going beyond the important desk based quantitative modeling. A true political reform is needed, one that recognizes the importance of starting a green revolution to Haiti.