What did we find when we compared the treatment and control groups? We found that body-worn cameras had no statistically significant average effects on any of the measured outcomes.
What does this mean?
First, let’s take a look at the results of our analyses.
A group of 1,000 officers with BWCs is estimated to document 74 more uses of force in a year than officers without BWCs. However, the data are also consistent with real effect of BWCs being anywhere from a decrease of 97 documented uses of force to an increase of 244 documented uses of force per 1,000 officers, per year.
Conclusion: The presence of BWCs had no detectable effect on police use of force.
A group of 1,000 officers with BWCs is estimated to receive 57 more civilian complaints in a year than officers without BWCs. However, the data are also consistent with real effect of BWCs being anywhere from a decrease of 24 complaints to an increase of 138 complaints per 1,000 officers, per year.
Conclusion: The presence of BWCs had no detectable effect on civilian complaints filed against police.
A group of 1,000 officers with BWCs is estimated to make 127 fewer arrests for disorderly conduct in a year than officers without BWCs. However, the data are also consistent with real effect of BWCs being anywhere from a decrease of 671 arrests for disorderly conduct to an increase of 416 such arrests per 1,000 officers, per year.
Conclusion: The presence of BWCs had no detectable effect on police discretion, as measured by arrests for disorderly conduct.
We examine whether the presence of a BWC on the arresting officer has an effect on the rate of prosecutions. A group of 1,000 officers with BWCs is associated with 2,422 more prosecutions in a year than officers without BWCs. That’s our best estimate based on the data we collected. However, the data are also consistent with real effect of BWCs being anywhere from a decrease of 2,738 prosecutions to an increase of 7,582 prosecutions per 1,000 officers, per year.
Conclusion: The effect here may be meaningful—potentially change in the thousands—but there is a lot of uncertainty in the estimate. We cautiously note no detectable effect, while also reminding ourselves that access to judicial outcomes data was constrained (see About the Approach).
We examine whether the presence of a BWC on the arresting officer has an effect on case disposition. A group of 1,000 officers with BWCs is associated with 14 more cases with a disposition of “found guilty” in a year than officers without BWCs. However, the data are also consistent with real effect of BWCs being anywhere from a decrease of 26 cases with a disposition of “found guilty” to an increase of 53 cases with a disposition of “found guilty.”
Conclusion: The presence of BWCs on the arresting officer had no detectable effect on the disposition of those cases as &lquo;found guilty.”
We examine whether the presence of a BWC on the arresting officer has an effect on case disposition. A group of 1,000 officers with BWCs is associated with 16 fewer cases with a disposition of &lquo;found not guilty” in a year than officers without BWCs. However, the data are also consistent with real effect of BWCs being anywhere from a decrease of 59 cases with a disposition of “found not guilty” to an increase of 28 cases with such a disposition.
Conclusion: The presence of BWCs on the arresting officer had no detectable effect on the disposition of those cases as “found not guilty.”
We examine whether the presence of a BWC on the arresting officer has an effect on case disposition. A group of 1,000 officers with BWCs is associated with 63 more cases that were disposed with a plea entered in a year than officers without BWCs. However, the data are also consistent with real effect of BWCs being anywhere from a decrease of 629 cases disposed with plea entered to an increase of 755 cases with a such a disposition.
Conclusion: The presence of BWCs on the arresting officer had no detectable effect on the disposition of those cases with a plea entered.
We examine whether the presence of a BWC on the arresting officer has an effect on case disposition. A group of 1,000 officers with BWCs is associated with 114 fewer cases that were ultimately not pursued in a year than officers without BWCs. However, the data are also consistent with real effect of BWCs being anywhere from a decrease of 315 cases not pursued to an increase of 86 cases with a such a disposition.
Conclusion: The presence of BWCs on the arresting officer had no discernible effect on the disposition of cases as “not pursued.”