The purpose of the study was to develop an instrument measuring perceived driver competence in the context of the Swedish driver education system. Moreover, the purpose of this study was to evaluate the psychometric characteristics of the instrument. The development of the instrument was based on a theoretical model which consisted of two dimensions. The first dimension consisted of two constructs about perceived competence: self-efficacy and expectancy beliefs. The second dimension consisted of five aspects of driver competence that are represented in the curriculum for driver education in Sweden. Based on this model two different instruments were developed for measuring perceived driver competence. These instruments were administered at a driving test centre in Sweden for five weeks. The two different instruments were compared and the results showed that both instruments demonstrated relatively good psychometric characteristics, although the self-efficacy instrument was slightly better. Moreover, the relationship between perceived competence and performance on the theory test was much stronger for the self-efficacy instrument than for the expectancy beliefs instrument, indicating that self-efficacy is a better predictor of actual competence. Thus, the conclusion was that the selfefficacy instrument should be used in a large-scale pretest in order to administer the instrument to a larger sample that is representative for the population of test-takers.