This paper through the collection of Korean students' actual corpora, analyzes the students' use of modern Chinese classifier, found that students using the classifiers have several characteristics as follows: (1) The use of classifiers are few in number, only a few dozen, far less than the total number of measure words in modern Chinese, the student of primary level and advanced level students in the selection of quantifiers, focused on the classifiers and concrete noun collocation, abstract nouns are not sensitive to; (2) Students universal widely use quantifier "pan", which is used of high frequency, relative frequency is very high; (3) In the example of leakage using classifiers, nearly half were missed using classifiers between the demonstratives pronoun "this" (or "that") and the noun; (4) When students use of classifiers, they do not know what classifier should be used according to the specific language environment, and do not know the correct position of the classifiers, especially the verbal classifiers. Overall the investigation results, the accuracy of senior level students using the classifier was generally better than that of primary level students, but the results also show that senior level students in the learning of new knowledge while forgetting of previously learning, so it prompt teachers, in the teaching, should summarize classifiers using rules, let the student not only studying new classifiers but also reviewing used classifiers. Because the author collected corpus analysis level is limited, finite, cannot say the results of these analyses covers Korean students using quantifiers everything, but I believe that it is a good try, to a certain extent, can reflect the Korean students of modern Chinese classifiers and errors, for future teaching to provide some reference.