Ranking Rice (Oryza sativa L.) Genotypes Using Multi-Criteria Decision Making, Correlation and Path Coefficient Analysis

Abstract

The evaluation of selection criteria using correlation coefficients, multiple regression and path analysis was carried out for a period of two years on sixteen genotypes of rice (Oryza sativa L.).These genotypes were studied during 2008 and 2009 summer seasons at EDduim and Kosti locations in randomized complete block design with three replications per each location. The field experiment is directed to study character association; contribution of various yield influencing traits on rice for establishment of appropriate plant attributes to select and improve the grain yield, and accordingly select the most suitable genotype. Combined analysis of variance revealed highly significant effects of locations, seasons, genotypes and their interactions for most of the studied traits indicating that these genotypes are highly variable. Genotypes differed significantly in grain yield, (NERICA 4, NERICA 14, NERICA 15, YUNLU 33 and WAB-1-38-19-14-P2-HB) were higher yielding genotypes giving 3.78, 4.03, 3.24, 3.55 and 3.51 t/ha respectively.  These genotypes presented a valuable source of diversity which can be used for breeding programs. Correlation analysis in both seasons indicated that grain yield was positively and significantly correlated with plant height, number filled grains/ panicle and 1000-grain weight, while it was negatively correlated with percentage of unfilled grains/panicle. Path coefficient analysis indicated that among yield components number of filled grains/ panicle, number of panicles/m2 and 1000-grain weight showed a positive direct effect on grain yield and therefore, may be considered as selection criteria for the improvement of grain yield. Multi-objective decision-making model was employed to rank the studied genotypes according to the measured various yield influencing traits and the degree of association of each trait on yield. For determination of criteria weight this article considers the analysis of correlation that is used frequently in to quantify the degree of association between a response variable, and some explanatory variable. Consequently, we propose new weighted information criteria to be used to guide the selection of the “best” genotype based on determining correlation coefficient. As a result, compromise programming analysis is in agreement with analysis of variance and indicated that genotypes can be ranked in a descending order as: N12, N14, Y30, WAB8, WAB19, N4, Y33, Y26, N15, N17 and Y24.