Alternative methodology for Scott-Knott test

• Published in: Crop breeding and applied biotechnology 2008, Vol.8 (1), p.9-16
• Main Authors: Bhering, L.L., Cruz, C.D., Vasconcelos, E.S., Ferreira, A., Resende Jr, M.F.R.
• Format: Article
• Language: English
• Subjects: multiple comparison procedures; plant breeding; simulation; Test procedures
• ISSN: 1984-7033; 1518-7853; 1984-7033
• DOI: 10.12702/1984-7033.v08n01a02

The test proposed by Scott Knott (1974), a procedure of means grouping, is an effective alternative to performprocedures of multiple comparisons without ambiguity. This study aimed to propose a modification related to the partitioningand means grouping in the said procedure, to obtain results without ambiguity among treatments, organized in morehomogeneous groups. In the proposed methodology, treatments that did not participate in the initial group are joined for a newanalysis, which allows for a better group distribution. In a comparative study, four experiments were simulated in a randomizedcomplete block design. The first consisted of 10 and the other 3 of 100 treatments. All experiments were performed in threereplications at a significance level of 0.05 for the means grouping test. Only in the third experiment of those of 100 treatmentsthe groups formed by Scott-Knott did not differ from the methodology proposed here. The proposed methodology is consideredeffective, aiming at the identification of elite cultivar groups for recommendation.