This paper makes an analysis of stock market network connectedness by using transfer entropy method, and discovers that nodes of regions affected by the crisis become closer to each other and the total network connectedness rises during the crisis. Moreover, the closer the stock market is towards the center of the network, the more likely it is affected by the shock. Last but not the least, dynamics of total network connectedness obtained by transfer entropy method is more stable and easier to interpret, avoiding the jumping points in calculation with the VAR model. Theory of transfer entropy shows its great potential in understanding the correlation and information flow of financial markets.