08/25/2015 - Laserfiche WebLink

Home Browse Search

McCleary/Rural Hotspots 159 Table 1 Crime-Related Secondary Effects of a Rural Adult Business Open Closed Log Effect 13 Property crimes 23 9.54 15 7.20 Personal crimes 3 1.24 5 2.40 Constant —3.267 —17.60 All other crimes 28 11.61 9 4.32 Open 0.475 2.06 Total crimes 54 22.39 29 13.92 e°475 1.61 Although parameter estimation requires working in the natural log metric, log- parameters are not easily interpreted. However, the exponentiated effect estimate is approximately equal to the ratio of the segments. In this instance, the value (e°475) 1.61 is interpreted as a 61% difference.The rate of total crime in Montrose was 61% higher during the 29 months that the Lion's Den was open, that is, compared to the period prior to February 2003, before the Lion's Den opened, and the period after July 2005, when it closed. This is a large, statistically significant crime-related secondary effect. Internal Validity Another set of alternative explanations involve uncontrolled threats to internal validity.The switching regime(closed-open-closed)property of the quasi-experimental design controls many of the most common threats to internal validity. Nevertheless, authorities on quasi-experimental design(Campbell&Stanley, 1966;Cook&Campbell, 1979; Shadish, Cook, & Campbell, 2002) cite maturation, history, and instrumenta- tion as the most plausible threats to the internal validity of time-series designs. The threat of maturation refers to the possibility that the effect reported in Table 1 may be due, not to the opening of the Lion's Den but to a natural trend in the village's crime rate.However,because the daily time total crime time series satisfies the simple Poisson homogeneity assumption (Feller, 1968), the maturation hypothesis is rejected. The threat of history refers to the possibility that the effect may be because of some event in the village that coincided with the opening of the Lion's Den.A search of local news media found only one significant event during the 1,662-day time series. Shortly after the Lion's Den opened, the village's only liquor-serving tavern closed permanently. However, if the tavern's closing had any effect on crime in Montrose,the expected effect would have been to reduce the crime rate during the 881 days that the Lion's Den was open.Accordingly, history is rejected as an alternative hypothesis. Instrumentation refers to the possibility that the effect may be due, not to the opening of the Lion's Den but to a coincidental change in the way that crimes are recorded in the village. If the Effingham County Sheriff stepped up the frequency of Downloaded from http;!/cjpsageput l9MIMS1LIFORNIA IRVINE on May 25,2008 ®2008 SAGE Publications.All rights rase martial use or unauthorized distribution.