Previous studies have demonstrated that the preference for online social interaction (POSI) is one of the most important influence factors affecting problematic Internet use (PIU). Online communication promotes intimate relationships, alleviates social anxiety, and strengthens interpersonal skills. However, some other researches also show that a portion of youth are dependent on the Internet and are even addicted to the Internet, specifically because of the safety and continued interpersonal relationships that online communication provides. According to the assumptions of the Self-Worth Orientation Theory, the human is a reasonable and social (socialized) animal, and reasonable thinking of human beings pursues things’ reason. People need to construct and explain the system of the inside and outside world, which includes oneself and others, and find the self and the meaning of life; self-worth is the ultimate reason system of self-behavior. The inner and outside psychological resources could promote self-worth. And the family social economic statue (SES) is one supporting resource of self-worth system. Thus, based on the theory of Self-Worth Orientation, we hypothesized that, if a youth lives in a high SES family, his or her self-worth may be high, which may weaken the dependence of the acceptance and support from online, in turn, the effect of POSI on PIU may fade down; on the contrary, if one lives in a low family SES, his or her self-worth may be low, the acceptance and support from online interaction would be integral to make up for the lack of self-worth, thus, the effect of POSI on PIU may enhance. In sum, the family SES may moderate the effect of POSI on PIU. To explore this hypothesis, we adopted Young’s Internet addiction diagnostic questionnaire, preference online social interaction scale, and self-made family SES scale to measured 1219 adolescents and their parent. For youths, 447 were elementary school students (Mage = 10.72, SD = 1.39), 418 were middle school students (Mage = 13.04, SD = 1.23), 351 were middle school students (Mage = 16.03, SD = 0.88). For parents, 429 respondents were fathers, 729 respondents were mothers, 38 were non-parent, and 24 did not response. The date of non-parent and no response were not included in the analysis. A hierarchical regression model was used to investigate the effect of control variables (gender and youth’s age), the independent variable (POSI), the moderator variable (family SES), and interaction between POSI and family SES. The results showed that the effect of POSI on PIU was significant, and the effect was moderated by parent’s SES. Further analysis revealed that, living in high family SES, the effect of POSI on PIU was not significant (B = 0.008 SE = 0.007, t(843) = 1.24, p = .22, ?2 = .002); however, living in low family SES, teenagers’ POSI predicted PIU significantly ( B = 0.030, SE = 0.007, t(843) = 4.20, p < .001, ?2 = .020). These results support the hypothesis of Self-worth Orient Theory. This study, not only in theory level, affords a new perspective to explain the mechanism of PIU, also, it provides some new inspiration for prevention and treatment of PIU---giving more attention and offer more supports of self-worth to the adolescents who live in low family SES.