Adaptive Selection of Counties to Control versus Intervention Groups in SW Georgia Behavioral Study

A study to assess the effectiveness of a behavioral intervention versus a no intervention control in Southwest Georgia was designed jointly by the Southwest Georgia Cancer Coalition and the Jiann-Ping Hsu College of Public Health. Control and Intervention Groups included residents in five of the 27 counties in SW GA. The 5 counties in each group were chosen to minimize differences between the two groups in terms of percent African American, Median Household Income, the number of persons per square mile, and total population, while ensuring that the counties chosen had the largest percentage African American, lowest median household income and were the most rural. Details of this ‘min-max' allocation scheme will be presented. The 'min-max' allocation scheme developed in this research selects as candidates those counties that are most underserved -- largest African American(AA) populations, lowest income (I) and most rural (R). Candidate counties are then assigned to two groups in such a way as to minimize differences between the groups in terms of these characteristics and also total population. The scheme is based on ranks and does not require sophisticated statistical methods.