A Comparative Study of Collocations in a Native Corpus and a Learner Corpus of Spanish

AbstractThis study proposes a method for evaluating the written production of Spanish collocations. We begin by asking if the native speaker model is the appropriate one for learners. In order to answer this question we undertook the annotation of collocations in two parallel corpora, one by native speakers, and another by learners. Once both corpora were annotated, the collocational richness of learners and native speakers were compared. In order to measure collocational richness, four parameters were established (density, variety, sophistication and number of errors). Our results show that learners do, in fact, use collocations, but their choices lack the variety, sophistication and correction exhibited by native speakers