Learning a second language is a life-long process requiring acquaintance with a repertoire of language learning strategies (LLS). Despite copious research on LLS and their role in fostering autonomous learning, few studies have examined strategies employed when trying to master specific subsystems of the target language, especially grammar. Meaningful communication in a second language hinges on implicit or highly automatized grammar knowledge, which, given the limitations of classroom-based grammar instruction, must for the most part be developed by learners themselves outside the classroom and this can best be achieved through adept use of grammar learning strategies (GLS). Our knowledge of the GLS, however, is scant, an issue that can be addressed either by designing reliable GLS measurement instruments or revalidating the few existing ones. This study adopts the latter approach by investigating the psychometric properties of Pawlak's (2018) "Grammar Learning Strategy Inventory" (GLSI) in the Chinese context. The tool was administered to 923 English majors, and the responses were subjected to exploratory and confirmatory factor analyses. We found that all seven strategy categories included in the original instrument showed acceptable composite reliability and discriminant validity. The paper concludes by suggesting further revalidation of the GLSI and proposing avenues of research that employ methodological innovations to shed further light on GLS.