This study analysed the difference of firm performance between family firms and non-family firms. I define family firms in three ways: whether the controlling family are or not, whether one or more members of controlling family participate in the board of directors, whether a member of controlling family is the CEO of firm. Our results suggest that family firms have better performance except family firms managed by a member of family as a CEO. This paper contributes to the literature on the family firms to find that the various results of previous researches are attributed to different definitions.