A Study Of English Vocabulary Fluency Measurement And Its Validity

Vocabulary knowledge, an important indicator of language proficiency, has played a significant role in both first and second language acquisition. In recent years, studies on vocabulary can be generalized into three perspectives:vocabulary knowledge framework, vocabulary knowledge measurement and validation of vocabulary tests. As for vocabulary knowledge framework, although there might be some gaps, studies concerning vocabulary breadth and depth knowledge are complete and mature. However, research concerning another dimension of vocabulary knowledge——vocabulary fluency, is rare. With regard to vocabulary testing, Vocabulary Level Test and Controlled Productive Vocabulary Level Test are two main instruments for vocabulary breadth knowfedge. Word Association Test and Vocabulary Knowledge Scale are two most influential instruments for vocabulary depth knowledge. However, there is no widely acknowledged Vocabulary Fluency Test (VFT). In terms of validation, SPSS has been used widely in data processing. Despite continuous application of research findings into teaching practice, researchers find that learners, even English majors, still suffer from considerable difficulties in English communication when their vocabulary breadth and depth knowledge has reached a relatively high level. The author believes that this is caused by insufficient development of a vocabulary fluency knowledge. The underdevelopment of vocabulary fluency knowledge might lead to deferred information processing and outputting. Therefore, it is necessary to explore vocabulary fluency dimension in vocabulary knowledge as well. As a result, two purposes of this thesis are to explore an approach to measure vocabulary fluency knowledge and to validate the newly developed testEnlightened by the method adopted by Zhang and Lu (2013) to measure vocabulary fluency knowledge, this study used word recognition (the ability to retrieve words’ written form under the presence of words’ phonetic form) to measure vocabulary fluency knowledge. However, different from traditional form of word recognition, VFT for present study was considered and revised several times in target words selection, sentence context, and recording intervals between test items. For example, all of the target words were selected from CET-4 word list, words formulating sentence context were from the same or lower frequency level than those of target words and one second was extended in intervals between test items based on feedbacks in preliminary pilot test. The vocabulary fluency knowledge in this research referred specifically to the retrieval of words’ form in word recognition from speech. The whole study consisted of three vocabulary tests:Vocabulary Level Test, Controlled-Productive Vocabulary Level Test and VFT, among which VFT was the focus of the present study and the other two were auxiliary tests. Validation study was conducted following Hughes’(1989) framework of validity and reliability and Messick’s (1994) framework of construct validity. In order to validate the new test, 40 college freshmen were selected randomly to sit the three tests. Their scores on every test item of the three tests were collected separately. All the collected data were processed by Rasch model to validate the VFT. Semi-structured interview was conducted with 7 participants after the test to obtain participants’ reception and attitudes towards VFT. The main research questions were about the content, criterion-related, construct and face validity of the VFT.The research findings revealed the following results:1) The content of the test items were all chosen from CET-4 word lists and relevant to what learners had learned in the previous study, showing sound content validity,2) Criterion-related validation reflected the test had mediate agreement with other vocabulary measurement instruments and language tests; 3) The test displayed good construct validity in that item difficulty displayed sufficient spread without gaps in item hierarchy and were consistent with theoretical expectations, item residuals suggested that VFT measured only one meaningful construct, and the generalizability/reliability of VFT was good; 4) The majority of participants acknowledged the face validity of the present test and completed test in a way that reflected the underlying construct of vocabulary fluency knowledge based on the semi-structured interview; 5) The worst performance in VFT further confirmed the fact that insufficient attention had been paid to fluency dimension in teaching and research.The research also brought about the following implications:1) Vocabulary fluency knowledge was ignored and neglected in second language vocabulary acquisition and lagged far behind the development of vocabulary breadth knowledge. As a result, more attention should be paid to vocabulary fluency in language research and teaching; 2) Rasch model, as an objective way in test validation study, should be applied widely; 3) Learners should pay attention to their vocabulary fluency development in vocabulary learning.