Our study on perceptions of corruption consist of a large scale population survey (2,256 citizens and 352 police officers). The results of this quantitative survey show that there exists a significant difference between occupational status groups as well as between police departments in the perception of corruption. Specialized anti-corruption officers are more severe but also more selective in judging the situations compared to other police services. Furthermore, high occupational status respondents and police officers have more perceptions of corruption in common than lower status groups. These findings were refined on the basis of qualitative focus groups with police respondents. It revealed that although the perceptions of corruption of police officers are relative stable over time, the societal context as well as individual elements determines the link between their perceptions and their behavior in the field. The results empirically support the theoretical model introduced by Peters and Welch. They suggested that any corrupt act is judged according to situational characteristics into following four dimensions: the public official involved, the favour provided by the public official, the payoff gained by the official and the donor of the payoff. However, our study results demonstrate that the use of situational characteristics alone cannot give us an explanation for variations in perceptions of corruption. A direct comparison between different occupational status groups – which are potential subgroups where tolerance for corruption is learned – and the police organisation working against crime provide genuine empirical data. This contrasting analysis of perceptions of corruption within these groups empirically supports our thesis that the process of ‘differential occupational association’ influences people’s perceptions of corruption. This theoretical framework is based on the work of Sutherland. The core of this PhD thesis consists of four articles. The first article is dedicated to the question of the ...