The purpose of this paper is to address the focus on factors regarding the pet ownership. When we checked the difference of the income level, the number of family member, the housing condition, and the ratio of urban/rural residence between such two groups as "pet owners" and "non owners", there was a statistically significant difference on all of the four factors. When we applied a logistic model on pet ownership, we found that number of family member has a positive effect while the existence of the child of six years old or less has negative effects. However, when we split the data according to the family type, we observed that the effect of number of family member varied depending on family types. As a result, we confirmed that, especially in the group of "parents and grown-up children", the pets would have a function as a substitution of small children.