Browsing by Subject "Racial disparities"
Now showing 1 - 2 of 2
- Results Per Page
- Sort Options
Item Disparate exposure to fine particulate air pollution in formerly redlined cities : Chicago, Dallas, and Fort Worth(2022-05-11) Kane, Clare Ennis MacLise; Olmstead, Sheila M.Fine particulate matter (PM₂.₅) pollution is the largest environmental health risk in the United States and globally (GBD, 2019). The leading sources of PM₂.₅ pollution in the United States are fossil-fuel combustion sources like power generation and residential energy use. People of color are disproportionately exposed to PM₂.₅ pollution and have higher rates of asthma, which is known to be triggered by PM₂.₅ exposure. This thesis evaluates satellite PM₂.₅ pollution in three formerly Home Owner’s Loan Corporation (HOLC) “redlined” cities (Chicago, Dallas, and Fort Worth) to determine if historic housing policies that have perpetuated residential segregation contribute to current disparities in PM₂.₅ pollution exposure. Results suggest that residents currently living in historically low-grade HOLC neighborhoods in Chicago are exposed to significantly higher levels of PM₂.₅ pollution than high-grade HOLC neighborhoods. Although results for Dallas-Fort Worth are not statistically significant, a positive relationship between increase in HOLC grade and PM₂.₅ concentrations was found. Additionally, formerly low-grade HOLC neighborhoods had significantly higher asthma rates in 2017 than high-grade HOLC areas in all three cities. All three cities also have qualitative examples of citizens who are residing in formerly redlined neighborhoods, experiencing high concentrations of PM₂.₅ pollution from surrounding industry, and experiencing poor health outcomes. These findings further support efforts by communities of color to understand energy equity and advocate for environmental justice policies in their neighborhoods as well as the Environmental Protection Agency’s (EPA) goal to understand the air quality concerns in overburdened communities and the health impacts these have on residentsItem The language of devaluation : how linguistic markers communicate and perpetuate social inequities(2023-02-08) Brandler, Serena; Pennebaker, James W.; Swann, William B; Griffin, Zenzi M; Jordan, Kayden NSystemic social inequities can be formed and maintained through explicitly degrading language such as hate speech and slurs but how might humans accomplish this implicitly? Previous research has pointed to language as a potential vehicle for implicit bias, but it is not yet known what specific set of linguistic markers might underlie devaluing language regardless of context. In two studies using large, real-world data, the current project explored the ways that grammatical, structural and contextual features of language communicate the relative worth of social groups. Using varied methods in natural language processing including dictionary-based approaches, topic modeling, named entity recognition, and parts of speech tagger, the current work gives promising candidates for devaluing language. In Study 1, text analyses were conducted on 18,264 missing persons reports from a large national database. Results indicated that non-White missing persons reports contained fewer words, less detail, less affect and were more likely to feature active (i.e., implying personal fault) rather than passive language. Similarly, non-White missing persons reports were more likely to feature words like “runaway” rather than “disappeared” or “missing,” again implying that the victim played an active role in their disappearance. Black missing persons reports in particular also feature less personal pronoun use. In Study 2, 199,941 tweets discussing 30 mass shootings in the US were analyzed to investigate if the language candidates found in Study 1 could be replicated in a different context. Specifically, the study examined the ways that language may be differentially impacted depending on the race of the victims under discussion. Effects for word count, detail and passivity were replicated while contrary findings in affect and personal pronoun use point to the potential impact of context and individual differences (e.g., political affiliation) in buffering or augmenting devaluing language. The implications for systemic disparities in multiple domains are discussed using language as a lens.