The objective of this paper is to develop intersection crash severity prediction model using ordered probit model. Motor vehicle fatalities is reported the major cause of death for people aging from 15 to 44. Moreover, intersections are recognized as the most hazardous locations on the roads. It is estimated that approximately 2.4 million intersection-related crashes occurred, representing 40 percent of all reported crashes and 21.5 percent of traffic fatalities in 2007. Furthermore, study shows that comprehensive cost of crash increases sharply with the aggregation of crash severity. Thus there is an urgent need to explore the significant factors influencing intersection crash severity. The explanatory factors in the paper include the characteristics of the environment, vehicle, and the driver. The estimated model captures the marginal effects of all important explanatory factors simultaneously. For example, the model indicates the impacts of geometry elements, alcohol/drug use, and etc on the severity of intersection crashes. Thus the results can assist engineers and government officials to have better understanding and find solutions to eliminate fatal crashes at intersections.