Identifying Dimensions of Vocabulary Knowledge in the Word Associates Test
Aaron Batty
Keio University
doi: http://dx.doi.org/10.7820/vli.v01.1.batty
Download this article (pdf)
Abstract
Depth of vocabulary knowledge (DVK) (i.e. how much a learner knows about the words he knows) is typically conceptualized as a psychologically multidimensional construct, including various forms of word knowledge. Read’s Word Associates Test (WAT) is the most common test of DVK in the literature, assessing knowledge of words’ synonyms and collocates. Despite the fact that the WAT aims to measure two dimensions of vocabulary knowledge, no studies until now have investigated whether these dimensions are psychometrically distinct. The present study seeks to fill that gap. A known-reliable-and-valid WAT developed by David Qian was administered to 530 Japanese university English majors. Confirmatory factor analysis was employed to investigate the psychometric dimensionality of the WAT. It was discovered that a bifactor model, wherein the primary explanatory factor is a vocabulary g-factor, with additional, uncorrelated factors for synonym and collocate items, demonstrated the best fit. This finding implies that although these dimensions of DVK may be somewhat distinct, they are largely subsumed by general vocabulary knowledge.
Keywords
vocabulary; depth of vocabulary knowledge; word associates test; multidimensionality; structural equation modeling.
Citation
Batty, A.(2012).Identifying dimensions of vocabulary knowledge in the word associates test. Vocabulary Learning and Instruction, 1(1), 7077. doi: 10.7820/vli.v01.1.batty