Against the backdrop of greater emphasis on learner centredness and growing attention to learners' voices in recent decades, concerns have been raised over researchers' dominance in shaping research scope and findings. Whether it be quantitative, qualitative, or mixed-methods research, learners' voices are mostly absent at the methodological design stage. In this article, we discuss the imbalance between researchers' and learners' voices in L2/FL empirical studies and why this deserves our attention. We then introduce personal construct theory and delineate the repertory grid approach, with which many language researchers are unlikely to be familiar. With a step-by-step illustration, we endeavour to exemplify how an adapted version of this approach can be applied to mid/large-scale studies. The article ends with a conclusion of how this approach potentially complements and supplements the use of common quantitative (e.g., questionnaires) and qualitative (e.g., interviews) instruments, and calls for a wider adoption of this approach in L2/FL studies.