More Like this

1. Credit Scores That Prioritize Customer Welfare: Theory and Evidence from Nigeria International Conference on Learning Representations (ICLR 2023) Workshop on Practical Machine Learning for Developing Countries, Kigali, Rwanda.

   Amaya, S.R.; Bj orkegren, D.; Blumenstock, J.E.; Nair, S. (December 2023, ICLR 2023 Workshop on Practical Machine Learning for Developing Countries)

   Poverty maps derived from satellite imagery are increasingly used to inform high-stakes policy decisions, such as the allocation of humanitarian aid and the distribution of government resources. Such poverty maps are typically constructed by training machine learning algorithms on a relatively modest amount of “ground truth” data from surveys, and then predicting poverty levels in areas where imagery exists but surveys do not. Using survey and satellite data from ten countries, this paper investigates disparities in representation, systematic biases in prediction errors, and fairness concerns in satellite-based poverty mapping across urban and rural lines, and shows how these phenomena affect the validity of policies based on predicted maps. Our findings highlight the importance of careful error and bias analysis before using satellite-based poverty maps in real-world policy decisions. 

   more » « less
2. Fairness and representation in satellite-based poverty maps: Evidence of urban-rural disparities and their impacts on downstream policy

   https://doi.org/10.24963/ijcai.2023/653  

   Aiken, Emily; Rolf, Ester; Blumenstock, Joshua E. (December 2023, 32nd International Joint Conference on Artificial Intelligence (IJCAI 2023))

   Poverty maps derived from satellite imagery are increasingly used to inform high-stakes policy decisions, such as the allocation of humanitarian aid and the distribution of government resources. Such poverty maps are typically constructed by training machine learning algorithms on a relatively modest amount of “ground truth” data from surveys, and then predicting poverty levels in areas where imagery exists but surveys do not. Using survey and satellite data from ten countries, this paper investigates disparities in representation, systematic biases in prediction errors, and fairness concerns in satellite-based poverty mapping across urban and rural lines, and shows how these phenomena affect the validity of policies based on predicted maps. Our findings highlight the importance of careful error and bias analysis before using satellite-based poverty maps in real-world policy decisions. 

   more » « less
3. Distributing a survey using Every Door Direct Mail in an ideal use case

   https://doi.org/10.1177/20597991251329754  

   White, Jacob_C; Bessette, Douglas_L (April 2025, Methodological Innovations)

   While paper mail-based surveys avoid much of the risk of bots and fraudulent data, they suffer from lower response rates and ever-inflating material and logistical costs. In response, there is a nascent, but growing literature investigating a lower cost, explicitly anonymous, mail-based survey distribution method called Every Door Direct Mail (EDDM). This study contributes to this growing body of literature by using EDDM to disseminate a sequential mixed-mode census-style survey that meets best use-case recommendations per past research. We make several design alterations to elicit higher response rates including using an outer envelope and cash incentive. The survey, distributed near large-scale solar developments in three urban Michigan communities (~1554 households), was geographically based, targeted a specific and limited population, and covered the potentially sensitive topic of local solar development, which may have also led to a higher response rate. The survey achieved an overall response rate of 10.2% with 158 complete surveys returned, demonstrating this work’s usefulness, use case, and flexibility. 

   more » « less
4. Leveraging the strengths of citizen science and structured surveys to achieve scalable inference on population size

   https://doi.org/10.1111/1365-2664.14512  

   Stillman, Andrew N.; Howell, Paige E.; Zimmerman, Guthrie S.; Bjerre, Emily R.; Millsap, Brian A.; Robinson, Orin J.; Fink, Daniel; Stuber, Erica F.; Ruiz‐Gutierrez, Viviana (October 2023, Journal of Applied Ecology)

   Abstract Population size is a key metric for management and policy decisions, yet wildlife monitoring programmes are often limited by the spatial and temporal scope of surveys. In these cases, citizen science data may provide complementary information at higher resolution and greater extent.We present a case study demonstrating how data from the eBird citizen science programme can be combined with regional monitoring efforts by the US Fish and Wildlife Service to produce high‐resolution estimates of golden eagle abundance. We developed a model that uses aerial survey data from the western United States to calibrate high‐resolution annual estimates of relative abundance from eBird. Using this model, we compared regional population size estimates based on the calibrated eBird information with those based on aerial survey data alone.Population size estimates based on the calibrated eBird information had strong correspondence to estimates from aerial survey data in two out of four regions, and population trajectories based on the two approaches showed high correlations.We demonstrate how the combination of citizen science data and targeted surveys can be used to (a) increase the spatial resolution of population size estimates, (b) extend the spatial extent of inference and (c) predict population size beyond the temporal period of surveys. Findings based on this case study can be used to refine policy metrics used by the US Fish and Wildlife Service and inform permitting regulations (e.g. mortality/harm associated with wind energy development).Policy implications: Our results demonstrate the ability of citizen science data to complement targeted monitoring programmes and improve the efficacy of decision frameworks that require information on population size or trajectory. After validating citizen science data against survey‐based benchmarks, agencies can harness strengths of citizen science data to supplement information needs and increase the resolution and extent of population size predictions. 

   more » « less
5. Research analytics capabilities (RAC) survey: development, validation and revision using the Rasch model

   https://doi.org/10.1108/JARHE-12-2023-0578  

   Robershaw, Katherine L; Xiao, Min; Wallett, Erin; Wolf, Baron G (July 2024, Journal of Applied Research in Higher Education)

   PurposeThe research enterprise within higher education is becoming more competitive as funding agencies require more collaborative research projects, higher-level of accountability and competition for limited resources. As a result, research analytics has emerged as a field, like many other areas within higher education to act as a data-informed unit to better understand how research institutions can effectively grow their research strategy. This is a new and emerging field within higher education. Design/methodology/approachAs businesses and other industries are embracing recent advances in data technologies such as cloud computing and big data analytic tools to inform decision making, research administration in higher education is seeing a potential in incorporating advanced data analytics to improve day-to-day operations and strategic advancement in institutional research. This paper documents the development of a survey measuring research administrators’ perspectives on how higher education and other research institutions perceive the use of data and analytics within the research administration functions. The survey development process started with composing a literature review on recent developments in data analytics within the research administration in the higher education domain, from which major components of data analytics in research administration were conceptualized and identified. This was followed by an item matrix mapping the evidence from literature with corresponding, newly drafted survey items. After revising the initial survey based on suggestions from a panel of subject matter experts to review, a pilot study was conducted using the revised survey instrument and validated by employing the Rasch measurement analysis. FindingsAfter revising the survey based on suggestions from the subject matter experts, a pilot study was conducted using the revised survey instrument. The resultant survey instrument consists of six dimensions and 36 survey items with an establishment of reasonable item fit, item separation and reliability. This survey protocol is useful for higher educational institutions to gauge research administrators’ perceptions of the culture of data analytics use in the workplace. Suggestions for future revisions and potential use of the survey were made. Originality/valueVery limited scholarly work has been published on this topic. The use of data-informed and data-driven approaches with in research strategy within higher education is an emerging field of study and practice. 

   more » « less