This paper aims to shed light on the characteristics of Chinese learners' acquisition of motion constructions by examining the motion constructions that Chinese learners choose to express motion events. To this end, the constructions 'kada' and 'ota' were analyzed in the learners' corpus. The semantic elements that consist of motion events were divided into [source], [goal], [goal, purpose of action], [path], [purpose], [manner], and [source, goal], and the syntactic representation of these elements by Chinese learners was examined. The results showed that the more frequent a typical construction is in Korean, the stronger the mapping between the meaning and the the constructions is for Chinese learners. In particular, when representing [goal], the construction 'NP-에 가다/오다' was used overwhelmingly more than the phrases 'NP-을 가다/오다' and 'NP-로 가다/오다', and this strong association led to miscoding errors due to the 'NP-에 가다/오 다' construction. However, many of these errors occurred at the beginner level and were eliminated at the intermediate and advanced levels, suggesting a developmental aspect of constructions acquisition. One pedagogical implication is that focus on form instruction is needed to help students map the cognitive elements of motion events in appropriate constructions.