Based on a sample of individuals who committed speeding offence(s) during a particular period, say 1995 – 2005, we exploit the panel nature of the data to account for individual specific unobserved factors that can plague our estimates. We choose a base year, say 2000, identify individuals who had (or had not) committed speeding offences in that year, and follow both groups of individuals throughout the sample period. A difference–in–difference estimation technique is employed to compare the tendency and frequency of speeding offences committed by different group of drivers.
