Newsline Newsline Newsline Newsline Newsline Newsline Newsline
NTEP Seeks Better
Data Analysis
In an interest to improve the acquisition and analysis of NTEP variety trial data, the NTEP Policy Committee recently approved five statistical proposals for initiation. Funding is provided by a grant from the United States Golf Association (USGA) Turfgrass and Environmental Research Committee.
All projects will use data from the 1990 National Kentucky Bluegrass Test (Med./High Maint.) and the 1990 National Perennial Ryegrass Test. Following is a summary of the five projects:
1. "Analysis of Regional Turfgrass Performance Trials: Interpretation of Genotype by Environment Interactions"
Investigators:
Dr. Scott Ebdon, Dr. Hugh Gauch, University of Massachusetts
Summary:
This project will investigate the performance of varieties at various locations and will answer some of the questions concerning why varieties perform well at some locations and not others. This may lead to a logical grouping or regional analysis of data.
2. "Statistical Evaluations of Turfgrass Variety Trials"
Dr. Ramon Littell, Dr. Albert Dudeck, University of Florida
Summary:
This project will study the appropriateness of current NTEP testing and analysis techniques including the rating scale, the LSD statistic and other statistical methods, field trial layout and design, etc.
3. "Evaluation of `Quality Rating Scale', Developing a New Scale and ANOVA Assumption Fulfillment"
Investigators:
Dr. Gary Krause, Dr. Suleiman Bughrara, University of Missouri
Summary:
Drs. Krause and Bughrara will scrutinize the 1-9 rating scale to determine how it fits into proper statistical crtiteria. Alternative rating scales will also be investigated.
Investigators:
Dr. Elizabeth Guertal, Dr. Edzard van Santen, Auburn University
Summary:
Large field trials may be hampered by non-uniform soils, disease pressure and/or ratings. Nearest Neighbor Analysis compares each plot with the two neighboring plots to determine the uniformity of data collected. Stability Analysis provides an indication of variety performance across differing locations. This will show if a particular turfgrass is a high-performing or low performing variety.
5. "Evaluation of Turf Quality Ratings in NTEP Experiments with Ordinal Data Techniques"
Investigators:
Dr. Oliver Schabenberger, Dr. John Rogers, Michigan State University
Summary:
This project will study the current NTEP statistical analysis techniques and compare the results to a method developed by Dr. Schabenberger. (Project will be moved with Dr. Schabenberger to Virginia Tech University)
Thanks again to the USGA for their generous support!!
Newsline Newsline Newsline Newsline Newsline Newsline Newsline Newsline
Page 2