“…For all other evaluations, the average, largest, and smallest effects were identical. The high variance of the random effects model led to the counterintuitive finding of the smallest effect meta-analysis having a larger OR than the largest effect meta-analysis.12 One evaluation was conducted at a city hospital(Gill & Spriggs, 2005), one was conducted in a school/university setting(Lim & Wilcox, 2017), two were conducted across entire cities but were unable to be disaggregated to smaller settings(Kim, 2008;, and one was conducted in undisclosed mixed environments(Lim, Kim, Eck, & Kim, 2016).13 Some studies reported the presence of flashing lights on top of CCTV cameras. Rather than consider this a separate category, we classified these studies as "signage" given that they related to a similar causal mechanism (i.e., visible confirmation of the CCTV camera presence).…”