The aim of this Diploma Thesis is to investigate the impact factors of Greek drivers’ self-declared traffic violations, which is directly related to the existing enforcement. To fulfill this object, six Binary Logistic Regression models and a Structural Equation Model (SEM) were developed using data from the second edition of the E-Survey of Road users’ Attitudes (ESRA). Driving with alcohol consumption above the legal limit, driving on a highway, but also inside and outside a residential area at a speed above the legal limit, and using a mobile phone without a headset and to access social networks while driving, have been selected as the six dependent variables of statistical models. Several statistical relationships were identified and quantified correlating these six unsafe self-declared behaviors with two latent variables and other independent variables. The covariance between the six dependent variables is positive and statistically significant revealing that drivers who engage, more frequently, in one of these risky behaviors is more likely to also engage in any of the others in combination. Lastly, in this Diploma Thesis, recommendations are provided that could enhance Greek drivers’ road safety behaviour, such as educational and training campaigns, as well as targeted and sustained interventions through enforcement increase.