The aim of this paper is to point out the issues that make the existing learner corpora difficult to use for second language acquisition and English education research, and to report on the characteristics of a new learner corpus, the Hiroshima Interlanguage Corpus (HIC), which has been compiled for conducting language acquisition research. Experimental methodologies, such as a grammatical judgment test or a picture-description task, are encouraged for use in second language acquisition research, and some kinds of language tests or questionnaires are used for English education research. In addition to these methods, learner corpus research should be adapted in order to do foreign/second language acquisition research. Although some learner corpora, such as the International Corpus of Learner English (ICLE), are available for language research, these have some difficulties as learner variables, such as particular learners’ proficiencies, are not always well described. Further, it is difficult to analyze the process of language production. This paper strives to explain the characteristics needed to compile a new, useful learner corpus for second language acquisition research.