| There are several gaps in the existing literature on trust and social capital. First, the methods of analyzing the trust structure and comparing different types of trust need to be improved. Second, few studies have explored the regional difference of trust, especially generalized trust, although China has dramatic diversities geographically, culturally and economically. Third, it still hasn't been convincingly explored that whether there is a relationship, if any, linear or nonlinear, between social capital and trust in the literature. Fourth, it is still unknown that whether the institutional resources will condition the relationship between social capital and trust.;In order to fill these gaps, this research applies a new data analyzing method to study the trust structure in urban China, and compare the differences among its components. This research also carefully scrutinizes different trust theories and social capital theories, advocates the resource view of social capital, and proposes that social capital measured in two different dimensions will affect the level of generalized trust, conditioned by several institutional resources.;The results show that Chaxugeju does exist in the respondents' trust in natural persons. Specifically, the trust placed in natural persons is lower than in institutions. Chaxugeju is more remarkable in interpersonal trust than in institutional trust. In terms of regional variations, the respondents in Southern China show higher level of generalized trust than their counterparts in Northern China. In addition, the rank-order from high to low in the level of generalized trust is the East, the Middle, and the West.;In this study, the concept of social capital is measured from two dimensions. One is social capital embedded in discussion network (by the name generator), the other is social capital embedded in New Year greeting network (by the position generator). The data show that the urban residents' discussion networks are relatively large, resourceful, dense and highly heterogeneous. On the other hand, the New Year greeting networks have the following features: containing more relatives, more interaction among members, smaller size, and less embedded yet unequally distributed resources.;The results confirm the hypothesis that social capital affects the generalized trust. Specifically, social capital embedded in discussion network negatively affects trust while social capital embedded in New Year greeting network positively influences trust. Overall, social capital has positive impact on the generalized trust in a nonlinear way. Finally, the results confirm that the relationship between social capital and trust is contingent on several institutional factors like Hukou, housing, and Danwei.;The results are not only meaningful academically; they also help understand current China and help shape the policy making in relevant fields. It gives clearer and more definite description on differences among types of trust in current urban China, which shed lights on the mechanisms behind social actions taken by citizens, and provides references for improving social policy. The findings on regional differences of generalized trust permit local government to adopt different policies to supervise economy and social action regionally, instead of using unified standard. The decomposition of social capital into two dimensional social capital provides a new perspective to understand social network in China. Such understanding makes agents construct and utilize their resources efficiently to fulfill their action purposes. Finally, the analysis on moderating effects of structural resources draws a clearer and more complete picture on the relation between social capital and trust and on the relation between individual resources and institutional structures. |
