This study aimed at confirming the factor structure of the Japanese version of Inventory of Parent and Peer Attachment (IPPA) for peers. Furthermore, it examined whether or not peer attachment is significantly influenced by father and/or mother attachment, and to see whether father and mother attachment affects mental health directly and/or indirectly via peer attachment for each gender among Japanese youth. The targets of this study were 433 (men: 149, women: 284) Japanese college students who completed every item of the Japanese version of father, mother, and peer IPPAs, as well as Clinical Outcomes in Routine Evaluation – Outcome Measure (CORE-OM) to evaluate their mental health. Confirmatory Factor Analysis (CFA) was conducted to confirm the factor structure of the Japanese version of peer IPPA. Structural Equation Modelling (SEM) was utilized to see whether or not peer attachment is significantly influenced by father and/or mother attachment, and whether or not peer attachment mediates between parental IPPA and mental health. A bi-factor model, constituted by one general factor and three specific factors forming the original IPPA developed by Armsden & Greenberg (1989), was supported by excluding one item from the specific factor Alienation. SEM revealed that the mental health of both male and female respondents was significantly influenced by peer attachment. Among males, its platform was built by attachment only to the father, whereas among females, it was built by both father and mother attachment.