The objectives of this study were to identify the engineering design factors and to develop a preliminary model for red light runners among motorcyclists in Malaysia. The Random-Effect Negative Binomial (RENB) regression model has been used to determine the factors. The data collection was obtained from one-hour site observation at thirty different sites in Kajang and Serdang town, state of Selangor. In the RENB model, six independent variables were found to be significantly influencing the behaviors of the motorcyclists at signalized intersections. These variables were (i) motorcycle volume, (ii) change interval, (iii) amber time, (iv) cycle length, (v) number of signal phase and (vi) number of approach legs of intersection. Results of RENB model showed that most of the red light running violations occurred at the signalized intersections with high motorcycle volume, shorter change interval time, shorter amber time, longer cycle length, more signal phases and more approach legs.