The objective of this research was to describe proportional differences across time and region in management practices among southern cotton farmers who experienced glyphosate-resistant (GR) weeds on their farms earlier than those who experienced them later and among farmers who were closest to one of four historical outbreak epicenters: Lauderdale County, TN; Macon County, GA; Edgecombe County, NC; and Terry County, TX. A mail survey was conducted with cotton farmers in 2012 from 13 southern, cotton-producing states. Survey responses on practices used by farmers were classified into three broad categories of labor, mechanical/tillage/chemical (MTC), and cultural. Proportions of respondents using practices from each category were identified by time and region; across which, proportional-difference tests were conducted. Results indicated respondents encountering GR weeds earlier were more likely than farmers who experienced them later to use the three broad-category practices (labor, 98 vs. 92%; MTC, 95 vs. 89%; and cultural, 86 vs. 76%) and specific practices, including hooded sprayers (76 vs. 58%), in-season herbicide change (83 vs. 60%), and field-border management (60 vs. 35%). Also, respondents closest to Lauderdale County were more likely than farmers closest to Edgecombe County to use broad-labor practices (99 vs. 91%) and specific practices, including hand hoeing (96 vs. 84%), hand spraying (49 vs. 31%), spot spraying (76 vs. 59%), wick applicator (13 vs. 11%), and field-border management (58 vs. 39%). Education programs on weed management can be developed and tailored according to the time and regional differences to provide effective information and communication channels to farmers.

Many studies on the adoption of precision technologies have generally used logit models to explain the adoption behavior of individuals. This study investigates factors affecting the intensity of precision agriculture technologies adopted by cotton farmers. Particular attention is given to the role of spatial yield variability on the number of precision farming technologies adopted, using a count data estimation procedure and farm-level data. Results indicate that farmers with more within-field yield variability adopted a higher number of precision agriculture technologies. Younger and better educated producers and the number of precision agriculture technologies used were significantly correlated. Finally, farmers using computers for management decisions also adopted a higher number of precision agriculture technologies.

Probit analysis identified factors that influence the adoption of precision farming technologies by Southeastern cotton farmers. Younger, more educated farmers who operated larger farms and were optimistic about the future of precision farming were most likely to adopt site-specific information technology. The probability of adopting variable-rate input application technology was higher for younger farmers who operated larger farms, owned more of the land they farmed, were more informed about the costs and benefits of precision farming, and were optimistic about the future of precision farming. Computer use was not important, possibly because custom hiring shifts the burden of computer use to agribusiness firms.

Binary logit analysis was used to identify the factors influencing adoption of Global Positioning System (GPS) guidance systems by cotton farmers in 11 Mid-south and Southeastern states. Results indicate that adoption was more likely by those who had already adopted other precision-farming practices and had used computers for farm management. In addition, younger and more affluent farmers were more likely to adopt. Farmers with larger farms and with relatively high yields were also more likely to adopt. Education was not a significant factor in a farmer's decision to adopt GPS guidance systems.

Stochastic Efficiency with Respect to a Function (SERF) is used to rank transgenic cotton technology groups and place an upper and lower bound on their value. Yield and production data from replicated plot experiments are used to build cumulative distribution functions of returns for nontransgenic, Roundup Ready, Bollgard, and stacked gene cotton cultivars. Analysis of Arkansas data indicated that the stacked gene and Roundup Ready technologies would be preferred by a large number of risk neutral and risk averse producers as long as the costs of the technology and seed are below the lower bounds calculated in this manuscript.

Personal digital assistants (PDA) and handheld global positioning systems (GPS) have become increasingly important in cotton production but little is known about their use. This research analyzed the adoption of PDA/handheld GPS devices in cotton production. A younger farmer who used a computer in farm management and had a positive perception of Extension had a greater likelihood of adopting the devices. In addition, farmers who used complementary remote sensing, plant mapping, and grid soil sampling information were more likely to use PDA/handheld GPS devices. Finally, the COTMAN in-field decision support program from Extension also positively impacted adoption.