Using Passive Sensing to Assess the Impact of Real-Time Discrimination against Women and Underrepresented Minorities in Engineering

使用被动传感评估工程领域对女性和代表性不足的少数群体的实时歧视的影响

• 批准号: 2009977
• 负责人: Jennifer Mankoff
• 金额: $ 63.8万
• 依托单位: University of Washington
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-10-01 至 2024-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2009977&HistoricalAwards=false
• 关键词: Using; Passive; Sensing; Assess; Impact

Increasing diversity in engineering and computer science has been a goal that remains elusive. Despite significant efforts, underrepresented minorities received only 16.1% and women received only 21.9% of engineering degrees in 2018. The reasons for these low numbers are complex and multifaceted and discrimination is an important factor in why students from these groups leave engineering. The goal of this research is to develop a holistic understanding of the impact of discrimination on historically underrepresented engineering students. In this era of big data and readily available technology such as mobile phones and wearables, a comprehensive change in how data about the college student experience are collected and assessed is possible. One can now move from lab to field, connect action to behavior, and collect longitudinal data. This, in turn, makes it possible to understand bias and its impact on engineering education in new ways: By complementing self-reports with passive data collection, big data can be used to create an image of behavior while learning about specific challenges underrepresented minority and female engineering students face. The project will result in a uniquely powerful longitudinal data set, which captures real-time changes in student experiences and allows study of the impact of discrimination at scale across a variety of contexts. The project will quantify the scope, direction, and longitudinal impact on behavior and link this to long-term outcomes such as GPA and retention. This ability to connect behavior to experience in the field was lacking in past studies of discrimination. Analytic techniques capable of capturing both individual variance and looking at unequal numbers of observations, such as hierarchical linear modeling, are required due to the large sample (N=200/year) and number of variables. The data are collected at a large public university and will be most applicable to similar programs at similar institutions. The research will support policy making and intervention design in engineering programs. The ultimate goal is to diversify the pool of engineering students, which will be of direct benefit to society by increasing representation and the range of perspectives engaged in the engineering and computer science workforce.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

在工程和计算机科学中增加多样性一直是一个难以实现的目标。尽管做出了巨大努力，但在2018年，代表性不足的少数民族仅获得16.1%的工程学位，女性仅获得21.9%的工程学位。这些低数字的原因是复杂和多方面的，歧视是这些群体的学生离开工程的一个重要因素。这项研究的目标是发展的歧视对历史上代表性不足的工程专业学生的影响的整体理解。在这个大数据和移动的手机和可穿戴设备等现成技术的时代，收集和评估大学生体验数据的方式可能会发生全面变化。人们现在可以从实验室到现场，将行动与行为联系起来，并收集纵向数据。这反过来又使人们有可能以新的方式理解偏见及其对工程教育的影响：通过被动数据收集补充自我报告，大数据可用于创建行为形象，同时了解代表性不足的少数民族和女性工程专业学生面临的具体挑战。该项目将产生一个独特的强大的纵向数据集，捕捉学生经历的实时变化，并允许在各种背景下大规模研究歧视的影响。该项目将量化范围，方向和对行为的纵向影响，并将其与GPA和保留等长期结果联系起来。这种将行为与实地经验联系起来的能力在过去的歧视研究中是缺乏的。由于样本量大（N=200/年）和变量数量多，需要能够捕获个体方差并查看不等数量观测值的分析技术，如分层线性模型。这些数据是在一所大型公立大学收集的，最适用于类似机构的类似项目。该研究将支持工程项目中的政策制定和干预设计。最终目标是使工程专业的学生多样化，这将通过增加工程和计算机科学工作者的代表性和观点范围直接造福于社会。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

2021, April. Understanding health and behavioral trends of successful students through machine learning models. In International Conference on Human Interaction and Emerging Technologies (pp. 516-525). Springer, Cham.

2021 年，四月。

• DOI: 10.1007/978-3-030-74009
• 发表时间: 2021
• 期刊: International Conference on Human Interaction and Emerging Technologies
• 作者: Kim, A.;Nikseresht, F.;Dutcher, J.M.;Tumminia, M.;Villalba, D.;Cohen, S.;Creswell, K.;Creswell, D.;Dey, A.K.;Mankoff, J.
• 通讯作者: Mankoff, J.

Distress among undergraduates: Marginality, stressors and resilience resources

本科生的困境：边缘性、压力源和复原力资源

• 发表时间: 2021
• 期刊: Journal of American college health
• 影响因子: 2.4
• 作者: Nurius, P. S.

A Computational Framework for Modeling Biobehavioral Rhythms from Mobile and Wearable Data Streams

用于根据移动和可穿戴数据流对生物行为节律进行建模的计算框架

• DOI: 10.1101/2020.08.10.244020
• 发表时间: 2021
• 期刊: ACM transactions on intelligent systems and technology
• 影响因子: 5
• 作者: Yan, Runze

Anticipate and Adjust: Cultivating Access in Human-Centered Methods

预测和调整：培养以人为本的方法的可及性

• DOI: 10.1145/3491102.3501882
• 发表时间: 2022
• 期刊: CHI '22: Proceedings of the 2022 CHI Conference on Human Factors in Computing Systems
• 作者: Mack, Kelly;McDonnell, Emma;Potluri, Venkatesh;Xu, Maggie;Zabala, Jailyn;Bigham, Jeffrey;Mankoff, Jennifer;Bennett, Cynthia
• 通讯作者: Bennett, Cynthia

Leveraging Collaborative-Filtering for Personalized Behavior Modeling: A Case Study of Depression Detection among College Students

利用协同过滤进行个性化行为建模：大学生抑郁症检测案例研究

• 发表时间: 2021
• 期刊: ACM transactions on humanrobot interaction
• 作者: Xuhai Xu, Prerna Chikersal