Inbreeding can cause an important bias in the estimates of quantitative genetic parameters, which in turn can affect the estimated potential for a selection response in species that are simultaneous hermaphrodite with partial self-fertilization. In the present study, the biases caused by self-fertilization on heritability (h2) estimates for productive traits were assessed at different ages in the tropical hermaphrodite pectinid Argopecten nucleus. Self-fertilization rates were estimated in both oocytes and adults, as the percentage of zygotes formed from unfertilized oocytes and by paternity analysis using microsatellite genetic markers, respectively. Fifty sexually mature individuals were individually spawned and genotyped using microsatellite markers. The sperm of one individual used as a male was used to fertilize the oocytes of 4 other individuals, generating 10 half-sibling families (HS) and 40 full-siblings families (FS). At the age of 3.5 months, the total weight, and the length, height, and width of the shell of 60 animals per FS family were individually measured. At the age of 9 months, the same measurements were repeated in the same individuals, which were then euthanized to weigh the soft tissues. In addition, the fecundity was estimated in each individual (number of oocytes produced), and a paternity analysis was performed. No significant differences were found in the average inbreeding rates estimated in oocytes and adults. However, no significant correlation was observed in the estimates of family self-fertilization rates when using both methods. The individuals derived from crossbreeding showed higher average values for productive traits than those produced by selfing, although interactions between the genotype and the depth of culture were observed in both groups. The h2 estimates using statistical models that consider self-fertilization in adults (models II and III) were lower than those observed when self-fertilization rates were estimated in oocytes, or not included in the model. However, the genetic correlations between the traits were similar using both methods for estimating self-fertilization rates. The present results suggest that self-fertilization results in inbreeding depression and induces an overestimation of the h2 for productive traits, but it does not affect estimates of genetic correlations or genotype by environment interactions.