Despite thirty-four years in which both series have been collected, there is still no consensus about which of the nation’s two indicators of crime is 
better and when one is more appropriate for use than the other. This dilemma results in the lack of a definitive answer to two key policy questions: 
what is the true violent crime trend over this period and what factors affect that trend. In this paper, I take a step forward in answering these 
questions by using state-space models to extract the underlying trend from the violent crime trends of the National Crime Victimization Survey (NCVS) 
and the FBI Uniform Crime Reports (UCR). The models show that the NCVS has been more successful at representing the underlying trend. Further, 
autoregressive models comparing the two series with the estimated trend show that similar factors are associated with the NCVS and the estimated 
violent crime trend.