Considerable effort has been devoted to examining the effects of transgenic Bt crops on non-target arthropods. Field studies usually compare the abundance of non-target species in Bt and non-Bt fields. Such comparisons may miss important effects. Using a spatially explicit, population dynamics model we show that the large scale planting of Bt crops can lead to population declines in non-Bt fields, making comparisons between Bt and non-Bt fields tenuous. Arthropods with a low reproductive rate, high emigration rate, and high mortality rate in Bt crop fields are most likely to experience population declines in non-Bt fields. However, these effects are only manifested when the use of Bt crops is high. Consequently, studies conducted in areas where the use of Bt crops is low may be able to detect lower arthropod abundance in Bt fields compared to non-Bt fields, but cannot determine if off-site effects will occur when the abundance of Bt crops increases. Likewise, studies conducted in areas where the use of Bt crops is high may be unable to detect negative effects of Bt crops on non-target arthropods because populations in non-Bt fields may have declined substantially. From a risk assessment perspective, being able to determine if a Bt crop will cause regional loss versus lower abundance in Bt fields with stable abundance in non-Bt fields is important. To overcome these issues we suggest a method that uses Global Positioning and Geographic Information Systems to evaluate the abundance of Bt crops surrounding sampled fields.