302-1 Application of Response Surface Designs In Agricultural Research.

Poster Number 600

See more from this Division: ASA Section: Biometry and Statistical Computing
See more from this Session: General Biometry & Statistical Computing: II
Wednesday, October 19, 2011
Henry Gonzalez Convention Center, Hall C
Share |

Yared Assefa1, Juan Du1 and Scott Staggenborg2, (1)Department of Statstics, Kansas State University, Manhattan, KS
(2)Department of Agronomy, Kansas State University, Manhattan, KS
Response surface methodology (RSM) is a collection of statistical strategies that are devised to determine the conditions which optimize yield. Currently, RSM is widely applied in industries. Its application in agriculture, however, is perceived to be limited. The objectives of the present study were: (1) to investigate the current utilization of response surface designs in agricultural studies vis-a-vie their potential use, (2) to study reasons that impede application of response surface designs in agriculture, and (3) to suggest possible areas of application of response surface designs in agricultural research. Selected agricultural journals were reviewed for the type of designs reported in their articles. The potential and advantage of using response surface designs, compared to the designs they already reported is studied. Merits of adopting RSM or modified RSM are illustrated using some examples from agricultural data sets. In addition, responses of agricultural experts on their acquaintance to response surface designs were surveyed. Based on these studies, the application of response surface designs in agriculture compared to their current use was found immense. It was found out that from minimum of 33% to about 70% of recently published articles in selected journals could utilize RS designs as their methodology. Particularly, research studies that involve more than two factors, short term experiments, and those in controlled environments that aim in obtaining factor levels that optimize a certain response were found to benefit by using response surface designs. Response surface designs were also found important in early identification of factors and factor levels for a subsequent experiment in agricultural research, when budget and space are limited. Lack of exposure was found the main limiting factor for the limited utilization of RSM and response surface designs in agricultural research. Including RSM and response surface designs in the curriculum of future agricultural researchers is recommended.
See more from this Division: ASA Section: Biometry and Statistical Computing
See more from this Session: General Biometry & Statistical Computing: II
Previous Abstract | Next Abstract >>