The long-tailed goral (Naemorhedus caudatus) is an endangered animal species in all its habitats worldwide, including South Korea. The imbalanced sex ratio in fragmented habitats is closely associated with extinction. Therefore, sex identification using wild animal samples would be necessary. However, only a few studies have been reported about the sex identification of gorals. In this study, we thus aimed at comparing the efficiency of sex identification using various goral sample types as templates and the amelogenin (AMEL) and DEAD-box polypeptide 3 (DDX3) genes as target sequences. We extracted DNA from goral feces, tissues, and blood samples, then amplified the AMEL (SE47/SE48 and SE47/SE53 primer pairs) and DDX3 genes for sex identification, comparing the goral DDX3X and DDX3Y target sequences to those in cattle. Our results indicated that the tissue- and blood sample-derived AMEL amplicons showed an unspecific band pattern containing the sex-specific band in the case of both primer pairs we used, whereas the DDX3 amplicon showed only the sex-specific band. In the case of the feces samples, only the sex-specific band was amplified using both the AMEL and DDX3 primer pairs. However, we found that the DDX3 amplicon exhibited a clearer band pattern than the AMEL amplicon. Then, we compared the DDX3X and DDX3Y target sequences between cattle and gorals. We found 5 and 8 nucleotide differences in the DDX3X and DDX3Y sequences, respectively. In conclusion, the DDX3 gene-related sex identification of the long-tailed goral appears to be more efficient and precise than the AMEL gene-related approach. This method could be used for the sex identification of the members of the Bovidae family.