SCC-CIVIC-PG Track B: Everyday Respect: Measuring & Improving Police Officer Communication During Motor Vehicle Stops

警察进行的交通拦截既普通又可能致命。该项目将使用随身摄像机的音频和视频片段来研究洛杉矶警察局（LAPD）官员和司机在这些停车期间的沟通。我们将研究警察在停车早期的沟通如何影响互动的展开，包括是否升级。我们还将研究在拦截期间如何对待社区成员是否存在种族、性别、阶级或其他差异，以及种族、残疾状况、社区背景和警官培训等因素如何相互作用，从而影响警官的沟通方式。该项目的第二阶段还包括与洛杉矶警察学院合作，根据我们的研究结果制定和实施新的培训课程，并与洛杉矶警察局宪法警务和政策办公室合作，探讨对部门政策进行修订的可能性。这项以社区为主导的研究利用了随身携带和仪表盘摄像头的镜头，以及补充停车数据、人员数据和基于社区背景的位置数据。我们超越了这一领域的现有工作，根据从洛杉矶警察局和广泛的社区成员和社区组织征求的意见，对应该衡量的“良好”沟通的各个方面（第一阶段）建立我们的沟通衡量标准。我们将雇用各种各样的人类注释人员，对我们认为对社区利益相关者重要的官员沟通的所有维度进行编码（第二阶段）。我们将使用机器学习工具，在这些人工注释的基础上进行训练，然后大规模测量沟通，然后对官员沟通的原因和后果进行统计分析，包括可能导致社区中某些群体受到不同对待的因素。该项目是对公民创新挑战项目（track b）的回应，旨在弥合基本资源与服务和社区需求之间的差距，是国家科学基金会、国土安全部和能源部之间的合作项目。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Traffic stops conducted by police officers are both mundane and potentially deadly. This project will use audio and video footage from body-worn cameras to study the communication between Los Angeles Police Department (LAPD) officers and drivers during these stops. We will examine how officers’ communication early on in a stop shapes how the interaction unfolds, including whether or not it escalates. We will also examine whether race, gender, class or other disparities exist in how community members are treated during stops, as well as how factors like race, disability status, community context, and officer training interact to shape how officers communicate. Stage 2 of this project also includes collaboration with the Los Angeles Police Academy to develop and implement new training curricula drawing on our findings and collaboration with the LAPD Office of Constitutional Policing and Policy to explore possible revisions to department policy. This community-led study draws on footage from body-worn and dashboard cameras, as well as complementary stop data, personnel data, and location-based data on community context. We move beyond existing work in this area by basing our measures of communication on views solicited from both the LAPD and a wide range of community members and community organizations regarding the various aspects of “good” communication that should be measured (Stage 1). We will employ a diverse population of human annotators to code all the dimensions of officer communication we identify as salient to community stakeholders (Stage 2). We will use machine learning tools, trained on these human annotations, to then measure communication at scale and then proceed to statistical analysis of both the causes and consequences of officer communication, including the factors that may lead some groups in the community to be treated differently than others.This project is in response to the Civic Innovation Challenge program—Track B. Bridging the gap between essential resources and services & community needs—and is a collaboration between NSF, the Department of Homeland Security, and the Department of Energy.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.