Study
Bedford and colleagues (2021) conducted a randomized controlled trial design, clustered by police region, to assess the efficacy of police officer use of a mobile technology device, and a street check app, on street checks and reported crime incidents in Australia. The mobile technology device was implemented in Australia from April 2014 to January 2016, during which time 4,100 mobile devices were given to road police and rapid-action patrol groups. In March 2016, another 1,300 mobile devices were implemented with frontline police officers. This study included 1,054 mobile devices implemented from late April to middle July 2016. Police officers across five police regions (N = 3,547) were allocated randomly to the intervention group (n = 1,227), which was assigned the mobile technology device, or to the comparison group (n = 2,225), which was not assigned a device.
Participants in the intervention group were male (73.3 percent), held the rank of Constable (59.9 percent), and were on average 37 years old. Participants in the comparison group were male (73 percent), held the rank of Constable (56 percent), and were on average 37 years old. No significant differences were found between the two groups at baseline.
Police agency administrative data were retrieved to compare officers with and without mobile devices (including the street check app) on the number of street checks performed per month and crime incidents (driving offenses, type 2a unregistered uninsured, type 2b driving while disqualified, type 2b driving while unlicensed, public nuisance, drunk driving, drugs possession/use, assault, burglary, domestic violence, and assault/resist/evade police). For the CrimeSolutions review of this study, the outcomes of interest were the total number of street checks performed per month, the total number of driving offenses reported per month, and the number of drug possession/use instances reported per month. The study used 6 months of pre- and post-intervention data to evaluate the outcomes. All variables were recorded as the total number of instances per month, for each officer.
Several difference-in-difference (DID) tests with a negative binomial approach were performed. This analysis included time (street checks and incidents reported 6 months both pre- and post-device introduction), condition (control and intervention), and an interaction DID variable: condition × time. The study did not conduct a subgroup analysis.