CSDE Trainees Seminars

Rebecca Rebbe – “Child Protection Reports and Placements of Infants Prenatally Exposed to Substances”

Presented alongside Joseph Mienko, Emily Brown, and Ali Rowhani-Rahbar

Abstract: Research has indicated that substance exposure during pregnancy is dangerous often affecting infants’ health. But it is unclear how hospitals and CPS respond to these infants. This study examines differences in hospital and CPS system initial responses within the neonatal period by type of prenatal substance exposure and race. Children born in Washington State from 2006-2013 whose birth records indicated prenatal substance exposure through ICD-9 codes were included in the study (n=12,276). Birth records were linked with CPS records. The initial response by the hospital and CPS was examined using multinomial logit models with interactions of substance exposure and race. Predicted probabilities were then simulated to enhance interpretability. Overall, we found that few infants prenatally exposed to substances were removed from their parents initially, but exposure types were treated differently. We found that minority substance-exposed infants were not reported to CPS or placed out-of-home more than white infants.

Rebecca earned her MSW at Boston College and her master’s in education in Risk & Prevention from the Harvard University.  She has a bachelor’s degree in social work from the University of Portland and has served as an AmeriCorps member working with adolescents. Rebecca has over 7 years post-MSW experience working with families involved with the child welfare system. She worked as an in-home family crisis clinician in Boston before working for the Washington State public child welfare system in Seattle as an ongoing case worker, a CPS investigator, and a permanency and adoptions supervisor with adolescents.  Rebecca studies the responses to child maltreatment through systems, laws, and policies.

---

Lee Fiorio – “Understanding Patterns of Human Mobility at Different Time Scales”

Presented alongside Emilio Zagheni, Guy Abel, Johnathan Hill, Gabriel Pestre, Emmanuel Letouze, and Jixuan Cai

Abstract: Recent decades have seen an explosion in the quantity of behavioral data generated by human interaction with digital devices. A growing body of literature has focused on the value and potential pitfalls of leveraging these “digital trace data”‘ to analyze social processes including human migration and mobility, but blind spots remain. One challenge, well known to migration scholars, is to standardize and compare different kinds of movements across different time and geographic scales. In this paper, we develop a methodology for parsing the population-level migration signal from individual-level point-in-time data using flexible time-scales. We propose a stochastic model for simulating patterns in digital trace data and test it against three datasets: geo-tagged Tweets and Gowalla check-ins in the U.S.; cell phone call detail records in Senegal. Similar patterns observed across all three empirical datasets demonstrate the utility of our approach for studying migration via digital trace data.

Lee is a graduate student in the department of geography and CSDE fellow. His work focuses on neighborhood change and migration in the US context with an emphasis on methodology and data visualization.

---

Connor Gilroy – “Extending the Demography of Sexuality With Digital Trace Data”

Presented alongside RidhiKashyap

Abstract: A core challenge in the demography of sexuality is accurate and detailed measurement of sexual preferences in the population. We approach this challenge using digital trace data from Facebook’s advertising platform to obtain aggregate estimates for sexuality in the United States, stratified by age, gender, geography, and relationship status. We find our data to be a good predictor of variation in other sources of sexual minority prevalence, Gallup and the American Community Survey. We then go beyond these sources to characterize our population in terms of age, gender, and bisexual or same-gender interest. We find a higher willingness to disclose sexuality among younger cohorts than older, and among heterosexual men than women. Conversely, we find large numbers of young bisexual women. Ultimately, 4.2 million or 2% of Facebook users in the United States disclose non-heterosexual interests, encompassing a large proportion of lesbian, gay, and bisexual (LGB) individuals in the US.

Connor is a graduate student in the Department of Sociology, and has research interests at the intersection of Sociology, Demography, and Data Science. As part of his fellowship, Connor will be studying the relationship between online and offline LGBTQ communities. The project is intended to combine online and offline data to understand the impact of the development of online communities on offline behavior. It is an important topic that is tied to the well-being of populations, which is a central research theme at CSDE. Connor’s CSDE Affiliate mentors are Kate Stovel and Emilio Zagheni.

---