Semantic clustering as opposed to thematic clustering in vocabulary learning

Abstract: The presentation of vocabulary is an important factor in second and foreign language learning. It is assumed that the most advantageous way of doing so is by means of semantic clustering. However, as is often the case in this field, this option has not been empirically demonstrated to be so. Tinkham (1997) proposes that thematic clustering may be a more effective means of presentation and compares the two using artificial vocabulary items. His findings confirmed the hypothesis underlying his proposal. As the findings of no single study can be assumed to be proof of the validity of a hypothesis, multiple replication studies are needed to confirm or disconfirm the findings. It is with this purpose in mind that we have replicated Tinkham's study. Our findings do not support those of Tinkham. This variance is discussed in the narrow context of the two studies and in the wider context of research in applied linguistics