With the rapid growth of wireless sensor systems deployment, data quality has become a critical issue to the success of these applications. In this paper, we first raise the data quality problem in WSN. Then, we propose a novel metric, data consistency, to evaluate the data quality. Our consistency models consider three perspectives of consistency: temporal, numerical, and frequency, covering both individual data and data streams. Moreover, several system protocol design issues are identified to support data consistency. Finally, we propose a set of APIs to facilitate managing data consistency.