Premium
The use of multilocation trials to select cultivars that are better than a control
Author(s) -
Piepho H.P.
Publication year - 1995
Publication title -
plant breeding
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.583
H-Index - 71
eISSN - 1439-0523
pISSN - 0179-9541
DOI - 10.1111/j.1439-0523.1995.tb01245.x
Subject(s) - cultivar , biology , statistical analysis , yield (engineering) , microbiology and biotechnology , statistics , control (management) , computer science , mathematics , botany , artificial intelligence , materials science , metallurgy
Multilocation trials are a common means for evaluating new cultivars or strains. Their statistical analysis is often based on multiple comparisons with one or more controls. The purpose of this paper is to demonstrate the use of the Chen‐Picket procedure for selecting cultivars that are better than a control, which does not require stringent statistical assumptions and allows conclusions about the probability that the selected subset contains all cultivars that are better than the control. This procedure appears preferable to the multiple t‐test, which is often used in the evaluation of yield trials.