Given the importance of adopting weed resistance management BMPs, it is important to develop methods to compare BMP adoption among farms and to identify factors that affect BMP adoption. Because of the relatively large number of BMPs and interactions among them, a composite index that integrates and aggregates over all practices is a necessary measure. We use data envelope analysis (DEA) to develop a measure of farmer adoption intensity for a set of interrelated BMPs. In addition, we use polychoric principal component analysis before applying the common-weight DEA method of Despotis to remove correlation among variables and transform categorical variables to continuous ones that fit better in DEA. We applied the method to survey data from soybean growers from ten states in the central and southern U.S. The empirical results suggest that most growers adopt most of the practices, but that there is room for improvement. In addition, we found a significant negative effect on BMP adoption intensity scores for growers who more highly valued the RR trait in soybeans and positive effects for growers who were concerned about herbicide resistant weeds and cost and crop safety when making herbicide decisions.