3-WAY METHODS FOR MULTIATTRIBUTE GENOTYPE X ENVIRONMENT DATA - AN ILLUSTRATED PARTIAL SURVEY

Several ordination and clustering techniques are discussed with respect to their usefulness in analysing multiattribute genotype X environment data. The methods are briefly described and illustrated by application to data from the Australian Cotton Cultivar Trials (ACCT), a series of regional variety trials designed to investigate various cotton (Gossypium hirsutum (L.)) lines in several locations each year. Multivariate techniques applicable to three-way data are necessary to assess these lines using yield and lint-quality data. By the choice of complementary methods, it is possible to make both global and detailed statements about the relative performance of the cotton lines. These techniques can enhance the researcher's ability to make informed decisions about the genotype X environment data collected from these trials using simultaneous analysis of the attributes of interest.