The purpose of this study is to investigate the use of have + noun collocations in two sets of corpora compiled from argumentative essays written by Korean EFL learners and NS students. After extracting all the occurrences of the have + noun collocations from the two corpora, the data were analyzed in terms of the frequency of the free and restricted collocations as well as the noun collocates. Furthermore, all the collocations extracted from the Korean EFL learner corpus were evaluated in terms of acceptability, and the unacceptable occurrences were classified into eight different error types. The results of the study showed that Korean EFL learners’ frequency of the have + noun collocations in the category of free collocations was significantly higher than that of NS students. While some of the noun collocates were shared by both groups, still a number of noun collocates were idiosyncratically used by Korean EFL learners. It was also found that errors other than lexical choice took up more than two-thirds of all the errors occurring in the corpus. Implications of the findings for the teaching of have + noun collocations are presented.