CHAPTER 8. Summary, discussion and conclusion 137 it investigates real programs in a real school with real students, using widely used tests that are part of the curriculum and not a simple intervention designed solely for the sake of the experiment. Ecological validity and experimental control thus seem to be incompatible by nature in long-term classroom research. Although generalizability is indeed not possible, pedagogical implications can still be well de ned by practitioners and policymakers, who can interpret the ndings against the backdrop of their own situations and determine what is relevant and meaningful to them (see also Spada, 2019). In the light of this “particularizability”, coined by Clarke (1994), the relevance of this study cannot be denied for practitioners and policymakers who might be inspired to reconsider current practice in foreign language teaching, as well as for researchers who might be inspired to replicate this study or to investigate foreign language teaching in the same direction.