At a first level, skills in designing multi-site studies in diverse environments will increasingly be required. Here, although the key research questions are clearly identified and common across sites, attention needs to be paid to how constructs are operationalized, research instruments designed, and sampling and data collection conducted at each site. The definition of product categories may, for example, differ as well as brand availability, the nature of the retail environment, or more insidiously, the socio-cultural context of consumption. Constructs or definitions used in one context are not necessarily appropriate in another. Research instruments, data collection or sampling procedures may incorporate bias, requiring reformulation or adaptation to ensure meaningful results (Craig and Douglas, 2000).