A lot of articles and publications have been published lately, warning about the problem of examination and grading knowledge. A special movement has been established for "the new culture" of examining and grading knowledge, resulting from up-to-date understanding of examination and grading knowledge. It mainly stresses:- the shift from examination of content to the aim-and-process oriented examination and grading,- formative examination of knowledge that enables learning from mistakes and encouraging meta-learning,- extending the ways and manners of examination and grading knowledge,- improving teachers' examination and grading knowledge and- cooperation of the pupils in the process if examination and grading knowledge. In the contribution we will focus on one of the segments of this new culture of examination and grading knowledge, namely the process of democratisation of knowledge grading. With an empirical research, carried out in December 2003, we were examining in what degree we could talk about the democratisation of knowledge in primary schools in the Republic of Slovenia that was supposed to be seen in:- arrangements between teachers and pupils about the content and the way of grading knowledge, - arrangements between teachers and pupils about the criteria and the grading categories for individual examination of the contents (common positioning),- arrangements between teachers and pupils about the exact dates and conditions of grading,- possibility of self- grading and common debating about teacher's grades,- possibility ob repeating ill-preformed examinations,- possibility of complaints and arbitration of grades.In the research we used descriptive and causal non-experimental method of the pedagogical research. We included 98 teachers of Slovene language (46 of them, 46,9 per cent) and mathematics (52 of them, 53,1 per cent) and 345 pupils from the seventh and eighth class of primary schools. We collected the data from teachers and pupils by using anonymous questionnaires, consisting of combination (7) of four-graded descriptive evaluation scales.The results show obvious difference between the group of teachers and the group of pupils in judging the frequency of individual characteristics of democratisation of grading knowledge during Slovene and mathematics lessons. Namely, teachers think that all of the characteristics of democratisation of grading knowledge are frequently present, while pupils believe they are not that frequent. According to this gap between their evaluations we may doubt about consistent democratisation of the process of grading knowledge in our primary schools. It is possible that teachers in their view believe that they use all of the elements of democratisation of grading knowledge, but within their own comprehension and experiences. However, they can for example tell the pupil the grade for his or her knowledge, while they are convinced that this is also a sufficient feedback about his or her knowledge, but the numerical grade is not a sufficient and accurate feedback about the quality if his or her knowledge.