Accurate diagnosis of partial hydatidiform moles (PHMs) is crucial for improving outcomes of gestational trophoblastic neoplasia. The use of short tandem repeat (STR) polymorphism analysis to distinguish between PHM and hydropic abortuses is instrumental; however, its diagnostic power has not been comprehensively assessed. Herein, we evaluated the diagnostic efficacy of STR in differentiating between PHM and hydropic abortus, thus providing an opportunity for early measurement of human chorionic gonadotropin for PHMs. We reviewed charts of STR polymorphism analysis performed on fresh villous specimens and patient blood samples using a commercial kit for 16 loci. The genetic classification of 79 PHMs was confirmed. STR was reliable in differentiating PHMs when at least 15 loci were available. Typically, PHMs are characterized by their triploidy, including two paternal and one maternal haploid contribution. In our sample, seven PHMs lacked the three-allelic loci, requiring fluorescence in situ hybridization (FISH) analysis to investigate imbalanced biparental conceptus and single-nucleotide polymorphism array analysis to reveal cytogenetic details. Of these PHMs, two, three, and one were identified as androgenetic/biparental mosaics (diploids), monospermic diandric monogynic triploids, and a typical dispermic diandric monogynic triploid, respectively. The remaining case was monospermic origin, but its ploidy details could not be available. Therefore, STR differentiated PHM from a biparental diploid abortus in most cases. However, PHM diagnosis may be compromised when STR is used as the sole method for cases displaying distinct cytogenetic patterns lacking the three-allelic loci, including androgenetic/biparental mosaicism. Therefore, FISH should be considered to confirm the diagnosis.