EAS Performance Metrics Research Literature Review Hayes 2002
EAS Performance Metrics Research Literature Review
Read Hayes, PhD, CPP
In Sherman et al. (1997) Eck looked at five evaluations of EAS regarding shrinkage reduction. Each study compared shrinkage before the installation of EAS, and after its introduction in the test stores. Each of these studies also used a single control store to measure background trends, and to see if loss in those stores trended the same way as the stores receiving EAS. All tests indicated EAS provided some loss reduction. Shrinkage reduction in these studies was reported to vary between 28/32 percent (Bamfield 1994), to 80 percent (DiLonardo 1996). Farrington and colleagues (1993) reported even greater reductions in shoplifting losses in the two stores they examined (76 to 93 percent).
Furthermore, EAS was found to be more effective than security guards (no improvement), or store redesign (50 to 80 percent temporary improvement) (Farrington et. al. 1993). Do to practical considerations in working with retailers, sampling procedures did not provide for the desired random selection of large samples. Eck reports that only Farrington et al. (1993) reported statistical significance tests that help to determine the probability that the reported reductions were due to chance alone. This assessment of several studies appears in the U.S. Department of Justice publication Preventing Crime: What works, What doesn’t, What’s Promising. The author of the chapter on Preventing Crime at Places reviewed the scholarly literature (published research) on the use of EAS to reduce theft and loss levels (Eck, 1997). His quote regarding the five studies was a qualified support for the systems: “Shoplifting appears to be controllable by the use of EAS technology, and possibly ink tags. If more evaluations had used significance tests we could have classified EAS as ‘works.’ In the absence of this information EAS must be placed in the ‘do not know’ category. Limited evaluations of other approaches suggest that there may be alternative approaches as well” (Eck, 1997, 7-19).