Swagata “Ban” Banerjee
Alabama A&M University
Steven W. Martin
Delta Research and Extension Center, Mississippi State University

Agricultural Resource Management Survey (ARMS) data for 2003 were used to estimate two binary logit models for two definitions of genetically modified (GM) cottonseed adoption: one in which adoption of either Bacillus thuringiensis (Bt) or herbicideresistant (HR) cottonseed was considered GM cottonseed adoption, and another in which adoption of Bt, HR, or stackedgene (SG) cottonseed was considered GM cottonseed adoption. Results indicate conservation tillage did not positively affect adoption of GM cottonseed with either of these definitions, while adoption of GM cottonseed in the previous year did. Conventional, non-GM cottonseed, used as “refuge,” also did not affect these adoption decisions for the study year.

Key words: Agricultural Resource Management Survey (ARMS), binary logit model, conservation tillage, cotton, genetically modified seed, herbicide-resistant cotton, jackknife procedure, refuge cotton, stacked-gene cotton, technology adoption.