Evaluations of field research on the nontarget effects of pest management, particularly the production of transgenic crops with insecticidal properties, suggest the methods used are sometimes unlikely to detect real differences among treatments. Among potential problems, abundance estimates may be scale-dependent for many arthropods, which move among experimental plots and between experimental plots and the surrounding environment. Insecticide-disturbed plots of field corn in a range of sizes in 2003 (0.03–0.53 ha; 18–72 m wide) and 2004 (0.01–0.13 ha; 9–36 m wide) were used for testing effects of scale on nontarget arthropod abundance. Possible effects of artificially isolating plots by removal of vegetation around plot borders were also studied in 2003. Community and taxon-based analyses showed abundance of foliar (aboveground) arthropods depended on plot size and isolation. While abundance of foliar arthropods was generally greater in smaller plots, isolation treatments suggested some taxa may have been either repelled or attracted to isolated plots. Levels of some epigeal (ground-dwelling) taxa were also size or isolation-dependent, but community-based analysis did not indicate a strong collective response to treatments. Recommendation of a practical but rigorous minimum plot size for nontarget studies may not be appropriate because responses to plot size varied among taxa. However, because arthropod movement into and out of plots can reduce differences between treatments, results suggest the use of small plots (width <9 m) for nontarget studies on transgenic crops generally should be avoided. Similarly, the taxon-specific effects of isolating plots should be considered when planning studies or interpreting results.