RESUMEN
Our collaboration seeks to demonstrate shared interrogation by exploring the ethics of machine learning benchmarks from a socio-technical management perspective with insight from public health and ethnic studies. Benchmarks, such as ImageNet, are annotated open data sets for training algorithms. The COVID-19 pandemic reinforced the practical need for ethical information infrastructures to analyze digital and social media, especially related to medicine and race. Social media analysis that obscures Black teen mental health and ignores anti-Asian hate fails as information infrastructure. Despite inadequately handling non-dominant voices, machine learning benchmarks are the basis for analysis in operational systems. Turning to the management literature, we interrogate cross-cutting problems of benchmarks through the lens of coupling, or mutual interdependence between people, technologies, and environments. Uncoupling inequality from machine learning benchmarks may require conceptualizing the social dependencies that build structural barriers to inclusion. © 2022 IEEE Computer Society. All rights reserved.
RESUMEN
BACKGROUND: Research examining state vocational rehabilitation agency (SVRA) sponsored service patterns during the COVID-19 pandemic is needed to improve employment outcomes among multiply marginalized persons of color with disabilities (i.e., African Americans, Asian Americans, Native Americans or Alaska Natives, Latinx, and Native Hawaiians/Pacific Island¬ers). Scarce attention has been paid to examining outcome inequities in the crisis. OBJECTIVE: This study applied a stratified bootstrap data expansion approach to assess the relationship between race/ethnicity, gender, level of educational attainment at closure and em¬ployment outcomes among target group members. METHODS: National fiscal year (FY) 2019 Rehabilitation Services Administration (RSA)-911 case records (N =114,229) closed between January 20, 2020 (date of first reported COVID-19 infection in the U.S.) to June 30 2020 were extracted and re-sampled across multiple trials using bootstrap procedures to increase logistic regression model accuracy. RESULTS: The findings indicated that African Americans, Asian Americans and Native Ameri¬can or Alaska Natives were statistically significantly less likely to achieve successful employment than non-Latinx Whites. Success probabilities in the COVID-19 pandemic were 'poorest'for Na¬tive American or Alaska Native VR consumers followed by African Americans, Asian Americans, Latinx, non-Latinx Whites, and then Native Hawaiians/Pacific Islanders. African Americans and Native Americans or Alaska Natives were more often closed unsuccessful because they could not be located when compared to non-Latinx Whites. CONCLUSIONS: These findings call for new targeted SVRA service initiatives. © 2022 National Rehabilitation Association. All rights reserved.
RESUMEN
The purpose of this scoping review was to examine the literature about Coro- navirus disease 2019 (COVID-19) and health and rehabilitation implications for persons of color, including those with multiple marginalized statuses (i.e., racial/ethnic minority and disability), and map out research gaps. Several themes emerged from this review that included disproportionate burdens of infections, hospitalizations and deaths;structural racism as a driver of inequity;diminished employment opportunities, and limited data on race and disability. The findings revealed the existence of serious evidence gaps relating to multiply marginalized people of color with disabilities within the COVID-19 context. A clear need exists for more research on COVID-19 outcomes disaggregated by disability and race so that the field has a fuller understanding of the impact of the pandemic on people of color with disabilities. Additionally, minority serving institutions (e.g., HBCUs and TCUs), as experts in addressing marginalized communities' needs, should play a leadership role in equity research. © 2022 National Rehabilitation Association. All rights reserved.