DAY 2 – Tuesday 3 May – 14:00-15:30
Swiss Tech | Room 1C | Level Garden
University of California Berkeley, United States
Temina Madon directs the Center for Effective Global Action (CEGA), a research network headquartered at the University of California, Berkeley. CEGA creates innovative products, services, and technologies for economic development. Temina has advised the WHO, World Bank, and Gates Foundation. Previously, she held positions in science policy at the National Institutes of Health and U.S. Congress, where she served as AAAS Science and Technology Policy fellow. She has a PhD in health sciences from UC Berkeley and an SB in engineering from MIT.
Development engineering (Dev Eng) applies principles from engineering, economics, and the social sciences to solve challenges arising from global poverty. A core focus of the Dev Eng research community is to improve the measurement of development indicators. How can we cost-effectively capture inputs from low-income, remote, and excluded communities? How can real-time, high-frequency, more reliable information be integrated into social policy and program design? How can we track progress toward poverty reduction—especially in light of the new Sustainable Development Goals (SDGs)?
New technologies—from satellites and mobile data streams, to sensors and administrative “big data”—are revealing the demands, preferences, and realities of people living in poverty. This rich information can inform the design of new products, services, and interventions that target poverty alleviation. It also facilitates the evaluation of new anti-poverty strategies and learning about what works. However, mainstreaming these innovations within the social sector has proven slow, given budget constraints, missing technical expertise, and lack of incentives to improve efficiency.
This session invites researchers and private sector product developers to showcase new measurement technologies and strategies that have been adapted for use in developing country contexts. Presentations will highlight how these innovations affect development practice in the field—in terms of resource allocation, program decision-making, and evaluation. We will also examine issues related to privacy and safety, particularly in fragile or conflict-affected areas.
Spigel Lauren, VaxTrac, United States, Monitoring and Evaluating Development Impacts: Case Examples from Implementing a Mobile Vaccination Registry System in Different Contexts
This presentation will provide case examples and experiences with developing evaluation frameworks to assess the adoption and impact of a mobile vaccination registry system for children in Benin and Nepal. VaxTrac designs, builds, supports, and evaluates a mobile vaccine registry system aimed to increase the availability, quality and use of vaccine data. This tablet-based system was implemented in 2014. A process and outcome evaluation framework was created to capture both technology adoption and health indicators using multiple methods of data collection. The evaluation framework intends to capture the impacts of having better quality data in a shorter time frame to make decisions about vaccine waste, coverage rates, and schedule adherence as well as successes and challenges with implementing a mobile vaccination registry. Data collection methods for technology adoption and use as well as health outputs and outcomes will be shared. Challenges and accomplishments in framework development and execution will be described, with an emphasis on the importance of understanding the context within which a technology for development intervention is implemented. Findings will be presented, compared, and contrasted for each country context. Lessons learned and reflections for designing and implementing technology for development evaluation frameworks and tools will be discussed.
Chander Kumar Singh, TERI University, India, Information Driven Socio-Behavioral Change to Mitigate Arsenic Crisis
Groundwater contaminated with arsenic of natural origin poses a serious threat to the health of tens of millions of villagers across South and Southeast Asia. Testing wells for arsenic provides information that is not substitutable. Because the distribution of arsenic incidence in groundwater is difficult to predict, and varies greatly even over small distances, the safety of a well cannot be predicted without a test. A well that meets guidelines for arsenic in drinking water may be found in immediate neighborhood of a very unsafe well. There is no systematic and predictable relationship between and arsenic and well depth. At the same time, precisely because arsenic contamination varies greatly over small distances, well tests make available an effective way to avoid exposure, namely by switching to a nearby safe well. A blanket testing campaign was conducted in approx. 200 villages in Indus River Basin under PEER Science grant of USAID as well as 26 villages in Gangetic River Basin from the grant by International Growth Centre to quantify the problem of Arsenic contamination of Groundwater. A fee-based arsenic testing was offered in a randomized controlled trial in 26 villages in Bihar, India, from 2012-2015. We also tried to emulate socio-behavioral impacts by disseminating the information on well water content of Arsenic by onsite field kit testing and then using google earth posters for larger dissemination of results in the village. A survey along with a follow-up testing after two years of first wave of testing was offered. During first wave of testing 31% of households switched to Arsenic safe wells. To assess the question of sustainability, we repeated the sales offer two years after the initial campaign, at the same (nominal) sales price. We recorded significant additional demand at the time of the repeat offer, with overall coverage rising from 27% to 45%. Data limitations do not allow us to ascertain what mechanisms lie behind additional demand: wealth increases, learning, or the direct effect of repeating the offer (what one might call a ‘marketing’ or ‘nudge’ effect). However, from the vantage point of policy interest in sustainability, the reduced-form effect of making a repeat offer is highly relevant. A key finding is that a repeat offer made within two years met with significant demand. This underscores the need for a more careful assessment of experimental evidence generated with one time offers, or offers available only for a short period. Given selective recall, the question of how best to provide information to households in a way that is salient but not socially costly deserves additional attention.
Joeri Smits, ETH Zurich, Switzerland, Household Income and Financial Distress as Development Outcomes: an Experiment with Financial Diaries
Income and expenditure data, a key development outcome, is often of low quality and it does not capture the highly temporally variable and irregular nature of cash flows of households and individuals in low-income countries. This paper reports on an experiment wherein the members of microcredit groups were randomly assigned to being offered “financial diaries” to keep record of their daily cash flows. One of the aims of the experiment was to see if keeping track of the diaries changes financial behavior and outcomes, as that could reduce the validity of the diaries as a measurement instrument. Intent-to-treat effects are not found to be significantly different from zero for all outcomes, except food consumption. This effect is driven by increases in the perception of consumption, but not consumption itself. Given the cost and high dropout, the approach seems not scalable.
Jay Taneja, IBM Research – Africa, Kenya, Filling Spatial and Temporal Gaps in Development Surveys Using Night Lights
Survey data are often used to measure development outcomes, but can be imprecise, biased, and expensive. Even datasets that manage to meet stringent quality criteria often have limited statistical coverage and are not collected frequently enough to remain relevant. In this work, we develop a method to address the spatial and temporal “gaps” that result from household survey-driven datasets commonly used in development studies, such as the USAID Demographic and Health Survey. Our method employs satellite night lighting data to enable more granular and timely assessments of community development than are available from traditional datasets. We also show how pairing the night lighting data with our methods can provide deeper insight of household wealth than current methods used in census data, representing a step forward for measurement of development outcomes..