Abstract LOCATIONAL ASPECTS OF ADAPTIVE REUSE: THE CASE OF NORTH CAROLINA‘S TEXTILE MILLS

This thesis examines the adaptive reuse of historic textile mills in North Carolina. The decline of the textile industry, which employed over 300,000 North Carolinians at its peak, has left scores of mills abandoned on the state‘s landscape. Many of these abandoned mills possess value that derives from a combination of their architectural splendor, their cultural significance, and their ability to reduce environmental impacts when creatively reused. Historic textile mills within North Carolina are inventoried. This unique inventory includes the precise location of 611 historic mills, their current uses, and a host of site and situational factors that might affect the probability of reuse. Reused mills are statistically contrasted against abandoned mills using logistic regression and key situational factors that could affect investment decisions. The model correctly predicts the reuse of 87 percent of abandoned mills. The key predictors of mill reuse are: local age, ethnicity, educational attainment, poverty, distance to nearest stream, and location within a historic district. Analysis of errors indicates a troublesome frequency of false negatives, i.e., reused mills that are predicted to be abandoned. This is attributed to a lack of key site variables in the logistic model, e.g., structure size, lot size, and architectural quality. The model is then used to guide selection of abandone