The paper presents a research into the use of collocations by B1, B2 and C1 level students of English at the University of Zenica conducted by analysing student's examination papers. Collocations represent an important segment of the knowledge of a foreign language and its vocabulary. They appear in many different forms and learning them might be seen as the most difficult task of L2 learning. Difficulties in using collocations appear even at advanced levels of language learning. A preliminary study into the use of collocations by B2 level students of English showed that the students used proper lexical collocations in about 60% and proper grammatical collocations in about 50% of the cases. In view of those results, a research was conducted among B1, B2 and C1 level students of English at the English Department of the University of Zenica to investigate their use of collocations. Student translations were analysed with a focus on the different types of collocations. The translations were produced as a part of their final examination in the Contemporary English courses 1 through 8 corresponding to the different CEFR levels analysed. We assumed that the results of the students in all four years of the English studies would be consistent with the results of the preliminary study, and that there would be no significant difference between the use of grammatical and lexical collocations.

It has been noted that without the appropriate adverbial modification some passive participles are not acceptable in the attributive position in English: e.g. *a found suitcase, *a killed young man, *a built apartment block, as opposed to some passive participles that are not subject to this restriction: e.g. an abandoned village, a complicated man, the needed discipline. The assumption that identification of categorial status can be better seen at the level of larger constituents, has led us to propose some distributional diagnostic tests to confirm whether they retain their verbal force or have achieved their adjectival status in English, when used in the attributive position. Since their morphological form has not proved to be reliable for identification of their category, we have proposed to test them by classifying the types of adverbs enabling some of the passive participles to be used attributively. Given that the same adverbs can modify both, adjectives and verbs, we have focused on the classification of degree adverbs pre-modifying passive participles attributively used. The tests proposed in this paper may show that the open class degree adverbs that modify passive participles in both, attributive and predicative distribution do not fall into the single class but split up into the subclasses: degree adverbs/intensifiers and adverbs of measure/quantity modifying passive participles in attributive and predicative function, respectively. Since passive participles, too, may have different interpretation and either adjectival or verbal/passive reading when used as attributes, the open class degree adverbs are tested at the level of participial phrase used attributively within the NP. The examples were compared with the translation equivalents in BCS to see whether the same constraint affects passive attributes in BCS. Keywords: degree adverbs, intensifiers, adverbs of quantity/ measure, passive participles, complementary distribution