Crowdsourcing has become a new channel for companies and organizations to collect the wisdom of crowds and reach business objectives. How to effectively motivate user participation and improve the quality of solutions has become an important issue that needs to be addressed in crowdsourcing research. While the influence of benefit factors on user participation has been widely tested, understanding of cost factors is still insufficient in extant literature. Based on social exchange theory, this paper proposed a research model to explain the impacts of benefit and cost factors on solver participation behavior as well as the moderating role of task complexity in crowdsourcing. The model will be tested using data from an online translation crowdsourcing task where solvers were invited to participate in the translation and fill out the questionnaire. This paper explores the differences in the factors which affect solvers participation intention and the quality of solutions. In addition, the role of task complexity can be found out by designing translation tasks of different task complexity and randomly assigning solvers to different tasks.