Purpose

Determine whether black and Latino pedestrians are more likely to be frisked or subjected to the use of force under New York City's stop-and-frisk program.

Methods

Inverse probability weighted regression adjustment and covariate exact matching were used on the New York City Police Department's stop-and-frisk data from 2008 to 2012 to estimate the average treatment effect on the treated (i.e. the effect of being black or Latino) on four post-stop outcomes (being frisked, being subjected to any force, being subjected to weapon force, and being subjected to non-weapon force).

Results

The results show that being black or Latino is associated with being frisked and subjected to non-weapon force. Race and ethnicity are not associated with weapon force.

Conclusions

Results are contextualized in terms of threat theory, administrative and organizational policy, and implicit bias. Because black and Latino pedestrians are still more likely to be frisked and subjected to non-weapon force after matching for relevant pedestrian and stop characteristics, particular attention is paid to the role of implicit bias.

中文翻译：

---

纽约市的停顿与烦恼：估计停顿后结局中的种族差异

目的

确定黑人和拉丁裔行人是否更有可能受到纽约市的“停走走走”计划的鞭打或使用武力。

方法

使用逆概率加权回归调整和协变量精确匹配，对纽约市警察局从2008年到2012年的停药和停药数据进行了估计，以估计在四个岗位上对被治疗者的平均治疗效果（即黑人或拉丁裔的影响） -停止结果（紧张，受到任何力量，受到武器的力量以及受到非武器的力量）。

结果

结果表明，黑人或拉丁裔与isk皱和受到非武器力量有关。种族和种族与武器力量无关。

结论

根据威胁理论，行政和组织政策以及隐性偏见对结果进行了情境分析。由于黑人和拉丁裔行人在匹配了相关的行人和停车站特征后仍然更有可能被打扰并受到非武器力量的影响，因此应特别注意隐性偏见的作用。