In Ghana, approximately 12% of all road traffic fatalities occur at junctions and the cause of these crashes, as assigned by the traffic police, include signal violation. Red light running place the violator and other road users at risk of road traffic crash. The main aim of the research work was to undertake baseline study to establish the current level of red light running through a direct observation survey and determine the risk factors associated with traffic light violation in the Kumasi Metropolis of Ghana. An uninterrupted road side observational survey was conducted at ten (10) signalized intersections using pro-forma checklist. A binary logit model was employed to determine the risk factors associated with traffic light violations. Overall, drivers were observed running red in 35% of all the red phases observed. From the model, factors which influence red light running include the age and gender of the driver, presence of a passenger in the vehicle, vehicle type, junction type, cycle length of the signal and queue length. There is a need for public awareness campaigns on the dangers of red light running. The education on red light violation must be accompanied by sustained enforcement of the traffic law by the traffic police to help reduce the violation of red light. Deployment of automatic red light cameras will also go a long way in ensuring enforcement at all times.