Poster Number 603
See more from this Division: ASA Section: Biometry and Statistical ComputingSee more from this Session: General Biometry & Statistical Computing: II
Wednesday, October 19, 2011
Henry Gonzalez Convention Center, Hall C
Variety trial information is still most often presented in a static format whether this is in a physical or electronic print format. To create this type of variety trial information agronomists and plant breeders can choose to present the results of individual site-years or combine results across years and locations. Research has shown that variety selection based on multiple years and multiple location data is more robust then the results of a single site-year in close proximity to a individual producer’s farm or field. The reasons for which locations to combine vary from the completely arbitrary such as a political border to the delineation of an area of inference that has some biological relevance such as maturity zones. There have been several variety trial databases developed in the past such as the Illinois Variety Information Program for soybeans, the Colorado Wheat Variety Performance database that are interactive. Each of these efforts has their strengths and weaknesses. The authors set out to combine the best features these different databases and create a web-based tool that: 1) is scalable and not restricted by political border such as a state line; 2) allows the user to create his/her own area of inference; 3) allows for meaningful means comparisons by applying the rigor of statistical analysis. Focus groups with growers and agronomists were used to develop and refine the tool. The result is a web-based database that, for now, encompasses the spring wheat variety trials of Minnesota and North Dakota.
See more from this Division: ASA Section: Biometry and Statistical ComputingSee more from this Session: General Biometry & Statistical Computing: II