Goldilocks and the three “Ts”: Targeting, testing, and tracking for “just right” democratic policing

Abstract

Abstract: Research Summary: Police are often criticized for doing “too much” or “too little” policing in various situations. These criticisms amount to testable hypotheses about whether “less” force, or intensity, or enforcement would have been enough, or whether “more” was needed. The rise of evidence‐based policing provides a starting point for public dialogues about those hypotheses, in ways that could help to build police legitimacy. Such dialogues can be focused on the questions posed by the three “Ts”: (1) Is police action targeted in a way that is proportionate to the harm that it can prevent? (2) Has the action been tested and found effective with the kinds of targets, and their levels of harm, where it is being used? (3) Is police action tracked to ensure it is delivered in the way that has been tested, and in compliance with relevant legal requirements? In this lecture, I frame the issue as follows: Can more widespread use of better research evidence on targeting, testing, and tracking police actions, shared more clearly among the public and police, help reduce the wide range of oscillation between over‐policing and under‐policing? Policy Implications: The use of these questions in public dialogue would be especially relevant to the three biggest threats to police legitimacy in the aftermath of George Floyd's murder: (A) police killing people, (B) police stopping people, and (C) police under‐patrolling high‐crime hot spots (while over‐patrolling low‐crime areas). One result of applying the three‐Ts questions to these threats, for example, could be the end of the vast overuse of stop and search in low‐violence areas. At the same time, this approach could also lead to reductions in homicide by increasing stops in highest violence hot spots. Such changes could demonstrate how the “Goldilocks principle” for the three Ts could get policing closer to “just right” for each place and person being policed.