Abstract:Anonymous and aggregated statistics derived from mobile phone data have proven efficacy as a proxy for human mobility in international development work and as inputs to epidemiological modeling of the spread of infectious diseases such as COVID-19. Despite the widely accepted promise of such data for better development outcomes, challenges persist in their systematic use across countries. This is not only the case for steady-state development use cases such as in the transport or urban development sectors, but… Show more
“…CDR have also been shared with the public sector for public health and humanitarian aid applications (Milusheva et al, 2021). Access issues aside, CDR contain private and sensitive data, including phone numbers and location traces.…”
Section: Discussionmentioning
confidence: 99%
“…Our paper connects these two literatures by rigorously assessing the extent to which phone-based estimates of poverty can help with program targeting (Blumenstock, 2020). We believe the analysis will be especially relevant to the increasing number of interventions that rely on mobile money to distribute cash payments (Gentilini et al, 2020), and the growing number of contexts where mobile phone data are being made available for humanitarian purposes (Milusheva et al, 2021). For example, in just the past few years, mobile money was used to make cash transfer payments in countries including Bangladesh (Ali & May, 2021), Ghana (Karlan et al, 2021), Liberia (USAID, 2021), and Malawi (Paul et al, 2021).…”
Section: Introductionmentioning
confidence: 92%
“…For phone-based targeting, we associate no cost with the collection and analysis of phone data. While in some cases phone data may require purchase from the operator, partnerships between mobile network operators and governments for social protection and public health applications have not, to date, involved payment (Milusheva et al, 2021). The fixed cost of mobile data analysis is non-negligible but its contribution to marginal cost is close to zero as the number of screened households increases.…”
Section: Cost and Speed Calculationsmentioning
confidence: 99%
“…Second, the CDR-based method requires data from mobile network operators. Data sharing agreements with mobile network operators take at minimum a few weeks to arrange, and substantially longer in the worst case (Milusheva et al, 2021). Third, and finally, training a CDR-based poverty prediction model is expensive in terms of memory, computing power, and human capacity, and will likely take several weeks to implement.…”
Can mobile phone data improve program targeting? By combining rich survey data from a "big push" anti-poverty program in Afghanistan with detailed mobile phone logs from program beneficiaries, we study the extent to which machine learning methods can accurately differentiate ultra-poor households eligible for program benefits from ineligible households. We show that machine learning methods leveraging mobile phone data can identify ultra-poor households nearly as accurately as survey-based measures of consumption and wealth; and that combining survey-based measures with mobile phone data produces classifications more accurate than those based on a single data source.
“…CDR have also been shared with the public sector for public health and humanitarian aid applications (Milusheva et al, 2021). Access issues aside, CDR contain private and sensitive data, including phone numbers and location traces.…”
Section: Discussionmentioning
confidence: 99%
“…Our paper connects these two literatures by rigorously assessing the extent to which phone-based estimates of poverty can help with program targeting (Blumenstock, 2020). We believe the analysis will be especially relevant to the increasing number of interventions that rely on mobile money to distribute cash payments (Gentilini et al, 2020), and the growing number of contexts where mobile phone data are being made available for humanitarian purposes (Milusheva et al, 2021). For example, in just the past few years, mobile money was used to make cash transfer payments in countries including Bangladesh (Ali & May, 2021), Ghana (Karlan et al, 2021), Liberia (USAID, 2021), and Malawi (Paul et al, 2021).…”
Section: Introductionmentioning
confidence: 92%
“…For phone-based targeting, we associate no cost with the collection and analysis of phone data. While in some cases phone data may require purchase from the operator, partnerships between mobile network operators and governments for social protection and public health applications have not, to date, involved payment (Milusheva et al, 2021). The fixed cost of mobile data analysis is non-negligible but its contribution to marginal cost is close to zero as the number of screened households increases.…”
Section: Cost and Speed Calculationsmentioning
confidence: 99%
“…Second, the CDR-based method requires data from mobile network operators. Data sharing agreements with mobile network operators take at minimum a few weeks to arrange, and substantially longer in the worst case (Milusheva et al, 2021). Third, and finally, training a CDR-based poverty prediction model is expensive in terms of memory, computing power, and human capacity, and will likely take several weeks to implement.…”
Can mobile phone data improve program targeting? By combining rich survey data from a "big push" anti-poverty program in Afghanistan with detailed mobile phone logs from program beneficiaries, we study the extent to which machine learning methods can accurately differentiate ultra-poor households eligible for program benefits from ineligible households. We show that machine learning methods leveraging mobile phone data can identify ultra-poor households nearly as accurately as survey-based measures of consumption and wealth; and that combining survey-based measures with mobile phone data produces classifications more accurate than those based on a single data source.
“…This paper shows that it is possible to form a mutually beneficial collaboration between a telecom operator and a public institution, and to make use of mobility data in evidence-based policymaking without compromising applicable personal data protection law. Milusheva et al (2021) note the proven efficacy of statistics derived from aggregated, anonymized mobile phone data as a proxy for mobility. They focus on drawing lessons learned and providing recommendations to unlock systematic use, based on their experiences of implementing collaborative projects around the world.…”
Section: Gilbert Et Al (2021) Describe How Mtn Inmentioning
In this editorial, Guest Editors Richard Benjamins (Telefónica), Jeanine Vos (GSMA), and Stefaan Verhulst (Data & Policy Editor-in-Chief) draw insights from a set of peer-reviewed, open access articles in a Data & Policy special collection dedicated to the use of Telco Big Data Analytics for COVID-19.
Agent-based models frequently make use of scaling techniques to render the simulated samples of population more tractable. The degree to which this scaling has implications for model forecasts, however, has yet to be explored; in particular, no research on the spatial implications of this has been done. This work presents a simulation of the spread of Covid-19 among districts in Zimbabwe and assesses the extent to which results vary relative to the samples upon which they are based. It is determined that in particular, different geographical dynamics of the spread of disease are associated with varying population sizes, with implications for others seeking to use scaled populations in their research.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.