[en] There is a growing need for an accurate measure of genetic relatedness in plant breeding applications due to crucial issues like varietal identification and intellectual property protection. This study aimed to: (1) calculate the coefficients of coancestry of selected rice entries using pedigree and DNA marker data; and, (2) compare the pedigree- and the marker-based estimates of coefficients of coancestry. Eight modern varieties and their 25 initial, intermediate, and immediate progenitors were chosen for the study. The DNA assay generated a total of 559 alleles from 119 SSR loci, with an average of 4.7 alleles per locus and an average polymorphism information content of 0.59. The pedigree-based coefficient of coancestry (fij) and marker-based coefficients of coancestry (non relatives method, Nfij; and tabular method, Tfij) were calculated in all pairwise genotype combinations. The mean values for fij, Nfij, and Tfij, computed among the 33 entries were 0.0649, 0.1502, and 0.0452, respectively. Both fij and Tfij were lower than Nfij because of the assumption that initial progenitors are unrelated. The discrepancies between fij and Tfij would be due to random drift and human selection occurring during inbred line development, the effect of which is to shift the parental genomic contributions from expected Mendelian ratios. Tfij yielded a higher correlation with fij and also had a lower proportion of genotype combinations that deviated ??0.10 from fij. The correlation of Nfij was low to fij but was high to Tfij, and the proportions of genotype combinations wherein Nfij deviated ??0.10 from fij were comparable to its deviations from Tfij. In terms of suitability to cluster analysis, fij was the poorest while Nfij and Tfij fared well. The Tfij dendrogram agreed more closely to the fij dendrogram than did the Nfij dendrogram. Tfij is, thus, more accurate to use when pedigree information is not a problem; Nfij is more meaningful to use otherwise. The highly significant correlation of the marker estimates to fij indicate their validity as measures of genetic relatedness, while their immunity to assumptions of the pedigree method and their appropriateness to classification imply their superiority. If data from molecular markers are to be useful for varietal identification and protection, they have to be translated to some meaningful measurement, especially when the number of both marker variants and genotypes under comparison are increased. A measure to directly quantify the degree of similarity, thus, becomes important and indispensable. This study has demonstrated that marker based estimates of coefficient of coancestry can address this need in rice