Abstract: Slovenia was one of few states where numerical method of grading in the lower grades of primary school have been taken place. With latest implementation of 9-year elementary school (vs. 8 year) - which began in 1999/2000 and hasn't been fully implemented yet - a descriptive grading was introduced into first cycle (first 3 grades). While theoretically (and considering experiences from various states) descriptive grades (DG) offer more information about student' knowledge, students' parents in Slovenia do not necessarily think so. Many parents prefer the numerical method of the grading and treat DG as "not-real" grade. We examine a communicative value of the descriptive grades: what they report to the parents, how the parents understand them, how they understand written standards of knowledge and how DG affect the parents' attitude towards the school. But surely consequences of all these are of the most importance: do the changes in grading also mean the changes in education quality? (Continuation): The study is going to be completed in April 2002. The information is collected from 30 Slovenian schools (1 class per school with approximately 20 students each). The students' parents are given questionnaires and interviewed by researchers. Written and oral statements of parents are going to be analysed in different ways. First, by linguistic means: conversation analyses, pragmatics (presuppositions, implicit meaning vs. explicit meaning etc.) Second, the parent's answers are coded and then statistically analysed by few variables. Teachers and students in 3rd grade are also asked through questionnaires. Reports with DG are examined as well: their communicative value is treated in two ways: how informative is text itself and how do the parents get the message. We are especially interested in what measures are taken by parents when they found out what their child didn't know. While numeric grade doesn't tell what the student didn't know when he or she got the grade less then perfect, DG should always say exactly this, namely: what could be better. Of course, many people don't see this. Some because they don't want to and some because do not know how (Our opinion is that standards that are written on the final reports -and given to the students at the end of the school year - hold a lot of the data but are very demanding to read.) We are also interested in how DG affect home work for school; it is almost a myth in Slovenia that elementary school students are over burdened in Slovenia. However, international studies (e.g. IEA studies) do not prove it. The parents expectations (as shown from previous studies) are contradictory: children should do less home work for school but at the same time teachers who do not impose much work to be done at home are not treated as good teachers. A same "logic" works against DG which are not treated as "real" grades. With this research we would like to find out and to show why these happen.