Semantic relatedness in L2 vocabulary learning: Does it really matter?
Second language (L2) textbooks often organize new vocabulary in lists of semantically related words under a common superordinate concept, such as food or family members. However, research on this topic has shown mixed results, with some studies suggesting that related lists facilitate learning, and others showing inhibiting effects. Importantly, all studies to date have been carried out in a laboratory or strictly controlled classroom setting where individual differences among students are often controlled for. Given that these differences may result in different learning gains in the authentic classroom environment compared to a controlled setting, the potential effects of semantic relatedness on vocabulary acquisition may similarly manifest differently when students are left to their own devices. This thesis reports on the first empirical study (to the author's knowledge) to test the effects of semantic relatedness on vocabulary learning in a truly authentic classroom environment. Two hundred and twelve students in beginner- and intermediate-level Spanish classes at Iowa State University were tested on their ability to translate items from one related list and one unrelated list from their course textbooks near the end of their respective units. Data were analyzed using mixed-effects logistic regression models under strict and sensitive scoring protocols.
Results indicated no evidence for a significant difference between scores on related and unrelated lists. Further regression analysis indicated a significant effect of individual lexical items on the learning outcomes, and item analyses suggested that some control over item-level characteristics may be needed to facilitate research even in the authentic classroom environment. Implications for teachers, materials developers, and researchers are discussed.