Estimates of vehicle miles traveled (VMT) are used extensively in transportation planning for allocating resources, estimating vehicle emissions, computing energy consumption, and assessing traffic impact. The estimates used in these applications usually come from different sources. For an objective comparison of VMT estimates from different methods, the principles and assumptions supporting the methods and the potential sources of error associated with the methods must be clearly understood. Methods of estimating VMT, including those used by the Indiana Department of Transportation (INDOT), are reviewed. Also presented is a comparison of statewide VMT estimates in Indiana from INDOT's traffic count–based method and a non–traffic-data cross-classification VMT estimation model developed for INDOT. The cross-classification model is an independent source of statewide VMT estimates in Indiana to supplement INDOT's traffic count–based estimates. The results of the comparison indicate that INDOT's traffic count–based estimates can be 10 to 20 percent higher than the estimates from the cross-classification VMT estimation model.