University of Michigan School of Information
Food delivery work and prediction markets: UMSI research roundup
Monday, 02/27/2023
University of Michigan School of Information faculty and PhD students are creating and sharing knowledge that helps build a better world. Here are some of their recent publications.
“Evaluating Equity and Inclusion in Cultural Heritage Grantmaking: CLIR’s Amplifying Unheard Voices Program”
Council on Library and Information Resources, February 2022.
Jesse A. Johnston, Ricardo L. Punzalan
This report summarizes a yearlong program assessment of "Amplifying Unheard Voices," a major revision of CLIR's Digitizing Hidden Collections grant program. The revision sought to expand the reach and appeal of the program to a broader range of institutions, including independent and community organizations, and to emphasize the digitization of historical materials that tell the stories of groups underrepresented in the digital historical record. Significant changes were made to the application structure, new applicant support resources were created, eligibility was expanded to Canada, and new thematic emphases and program values were added. The assessment was based on a series of qualitative data-gathering activities that included stakeholder groups and staff. Through surveys and interviews of applicants, inquirers, proposal reviewers, and staff, the authors provide a holistic view of the program, offer a series of recommendations, and identify areas for further attention.
“Automating Care: Online Food Delivery Work During the CoVID-19 Crisis in India”
FAccT’22, June 2022.
Anubha Singh, Tina Park
On March 23, 2020, the Government of India (GoI) announced one of the strictest nationwide lockdowns in the world to curb the spread of novel SARS-CoV-2, otherwise known as CoVID-19. The country came to a standstill overnight and the service industry, including small businesses and restaurants, took a massive financial hit. The unknown nature of the virus and its spread deepened anxiety among the general public, quickly turning to distrust towards any “outside” contact with goods and people.
In the hopes of (re)building consumer trust, food delivery platforms Zomato and Swiggy began providing digital solutions to exhibit care towards their customers, including: (1) sharing delivery workers’ live temperatures alongside the workers’ profile inside the app; (2) mandating the use of the controversial contact tracing app Aarogya Setu for the workers; (3) monitoring workers’ usage of masks through random selfie requests; and (4) sharing specific worker vaccination details on the app for customers to view, including vaccination date and the vaccine’s serial number.
Such invasive data gathering infrastructures to address public health threats have long focused on the surveillance of laborers, migrants, and the bodies of other marginalized communities. Framed as public health management, such biometric and health data gathering is treated as a necessary feature of caring for the well-being of the general public. However, such datafication practices - ones which primarily focus on the extraction of data from one specific community in order to mollify the concerns of another - normalizes the false perception that disease is transmitted unidirectionally: from worker to the consumer. By centering food delivery workers’ experiences during the pandemic and examining the normalization of such surveillance in the name of care and recovery, this paper aims to examine how new regimes of care are manufactured and legitimized using harmful and unethical datafication practices.
“Bones without Flesh and (Trans)Gender without Bodies: Querying Desires for Trans Historicity”
Cambridge University Press, February 2023.
In 2011, a 5,000-year-old “male” skeleton buried in a “female” way was discovered by an archaeological team just outside of modern-day Prague. This article queries the impulse to name such a discovery as evidence of transgender identity, and bodies, in an increasingly ancient past. To do so, it takes up the work of Denise Ferreira da Silva, Sylvia Wynter, and Hortense Spillers as a means to push back against the impetus to name such discoveries “transgender” in order to shore up the legitimacy of contemporary trans identity. Each of these three thinkers offers a different vantage point for rethinking such naming practices that push the reader to consider how desires to name and place “transgender” in a distant past papers over the violence of plantation slavery, global imperialism, and the Enlightenment's shift toward scientific reason. This article argues not that such anthropological discoveries should not be considered transgender, but rather that the desire for them to be, or become, transgender does not legitimate contemporary transgender identity, and may instead treat certain iterations of transness as spatially and temporally universal.
“MultiCite: Modeling realistic citations requires moving beyond the single-sentence single-label setting”
arXiv, August 2021.
Anne Lauscher, Brandon Ko, Bailey Kuehl, Sophie Johnson, David Jurgens, Arman Cohan, Kyle Lo
Citation context analysis (CCA) is an important task in natural language processing that studies how and why scholars discuss each others' work. Despite decades of study, traditional frameworks for CCA have largely relied on overly-simplistic assumptions of how authors cite, which ignore several important phenomena. For instance, scholarly papers often contain rich discussions of cited work that span multiple sentences and express multiple intents concurrently. Yet, CCA is typically approached as a single-sentence, single-label classification task, and thus existing datasets fail to capture this interesting discourse. In our work, we address this research gap by proposing a novel framework for CCA as a document-level context extraction and labeling task. We release MultiCite, a new dataset of 12,653 citation contexts from over 1,200 computational linguistics papers. Not only is it the largest collection of expert-annotated citation contexts to-date, MultiCite contains multi-sentence, multi-label citation contexts within full paper texts. Finally, we demonstrate how our dataset, while still usable for training classic CCA models, also supports the development of new types of models for CCA beyond fixed-width text classification. We release our code and dataset at this https URL.
“Work Expectations, Depressive Symptoms, and Passive Suicidal Ideation Among Older Adults: Evidence From the Health and Retirement Study”
The Gerontologist, December 2022.
Briana Mezuk, Linh Dang, David Jurgens, Jacqui Smith
Background and Objectives: Employment and work transitions (e.g., retirement) influence mental health. However, how psychosocial contexts such as anticipation and uncertainty about work transitions, irrespective of the transitions themselves, relate to mental health is unclear. This study examined the relationships of work expectations with depressive symptoms, major depression episodes (MDE), and passive suicidal ideation over a 10-year period among the “Baby Boom” cohort of the Health and Retirement Study.
Research Design and Methods: Analysis was limited to 13,247 respondents aged 53–70 observed from 2008 to 2018. Past-year depressive symptoms, MDE, and passive suicidal ideation were indexed using the Composite International Diagnostic Interview—Short Form. Expectations regarding working full-time after age 62 were assessed using a probability scale (0%–100%). Mixed-effect logistic regressions with time-varying covariates were used to assess the relationship of work expectations with mental health, accounting for demographics, health status, and functioning, and stratified by baseline employment status.
Results: At baseline, higher work expectations were inversely associated with depressive symptoms. Longitudinally, higher expectations were associated with lower odds of depressive symptoms (odds ratio [OR] = 0.93, 95% CI: 0.91, 0.94). This association was more pronounced among respondents not working at baseline (ORNot working = 0.93 vs ORWorking = 0.96). Greater uncertainty (i.e., expectations near 50%) was also inversely associated with depressive symptoms. Results were similar for past-year MDE and passive suicidal ideation.
Discussion and Implications: Expectations (overall likelihood and uncertainty), as indicators of psychosocial context, provide insight into the processes that link work transitions with depression risk.
"False Consensus, Information Theory, and Prediction Markets"
arXiv, November 2022.
Yuqing Kong, Grant Schoenebeck
We study a setting where Bayesian agents with a common prior have private information related to an event's outcome and sequentially make public announcements relating to their information. Our main result shows that when agents' private information is independent conditioning on the event's outcome whenever agents have similar beliefs about the outcome, their information is aggregated. That is, there is no false consensus.
Our main result has a short proof based on a natural information theoretic framework. A key ingredient of the framework is the equivalence between the sign of the ``interaction information'' and a super/sub-additive property of the value of people's information. This provides an intuitive interpretation and an interesting application of the interaction information, which measures the amount of information shared by three random variables.
We illustrate the power of this information theoretic framework by reproving two additional results within it: 1) that agents quickly agree when announcing (summaries of) beliefs in round robin fashion [Aaronson 2005]; and 2) results from [Chen et al 2010] on when prediction market agents should release information to maximize their payment. We also interpret the information theoretic framework and the above results in prediction markets by proving that the expected reward of revealing information is the conditional mutual information of the information revealed.
“Recordkeeping, logistics, and translation: a study of homeless services systems as infrastructure”
Archival Science, January 2023.
Pelle Tracey, Patricia Garcia, Ricardo Punzalan
Homeless services systems provide unhoused individuals access to emergency shelter, subsidized housing, and other life-sustaining resources. In this paper, we present a qualitative study that draws on the experiences of fifteen social service workers to examine how recordkeeping practices sustain homeless services systems and unite a tangled web of institutions and actors, including public housing systems, nonprofit agencies, and local governments. We address the following research questions: How is the infrastructure of homeless services sustained by recordkeeping? How are social service workers affected by increasing recordkeeping demands? In what ways do social service workers work against or ‘find the play’ in this system? To address these questions, we collected interviews and conducted artifact walkthroughs with our study participants. We analyzed the data using an infrastructural lens and found that current recordkeeping practices within homeless services systems comprise an "infrastructure of last resort" that functions logistically, prioritizing efficiency and speed. We also found that social service workers “speak back” to logistification by making the homeless services infrastructure more legible to their unhoused clients through mediation and acts of translation that help to produce better resource outcomes. Our findings show how structuring recordkeeping in ways that privilege efficiency and speed disrupts social service work and interferes with social service workers’ ability to provide care for vulnerable individuals facing life-altering and life-threatening hardships.
