A Study On The Effect Of Chunking On College English Listening Comprehension

Listening is one of most important language skills among listening, speaking,reading, writing and translating. Although,recently,it has received more and moreattention, college students still find that listening skill is the most difficult one of all thefive basic language skills. The previous researchers both abroad and at home have madegreat contribution to the teaching and learning of listening comprehension and given theguide on the techniques and the strategies from the perspectives of cognition, cognitivepsychology, psycholinguistics, pragmatics and culture. However, there is still a problemthat students have no enough time to comprehend the input linguistic data afterrecognizing each word of an utterance as it is spoken. The corresponding studies havefound this problem is, to some degree, caused by the limited capacity of short-termmemory. In order to find out the solution, some researchers, such as Call (1985) and XuFang (2005), conduct some experiments and use chunking to make up the deficiency ofshort-term memory. Their experiments prove that the memory for clauses is the predictorof listening comprehension success. But one limitation of the previous studies is that theymerely focus on the advanced participants, not on the Chinese college students---theintermediate English learners. In Stern's (1999) opinion, researches should be carried outdifferently for the different learners at different levels of proficiency. And according toKrashen's (1980) Input Hypothesis, different learners at different levels should be givendifferent suitable input. So the present thesis studies what kind of listening input iscomprehensible for the intermediate college students and hypothesizes that the memoryfor random digits is subcomponent of primary memory, and that the memory for randomwords and for phrases are the subcomponents of speech-processing memory;thesecomponents rank in order of phrases contributing the most to an explanation of thevariance in listening scores, random words the second and the random digits the least.This thesis conducts an experiment to verify the hypotheses and answer the researchquestions, using methods of a correlation, a multiple regression and a comparison toanalyze the data. The conclusion is drawn that chunk theory plays an important role inlistening comprehension, and that for the intermediate college students the practice inmemorizing phrases can improve the scores of their listening comprehension. Theimplication of the study is that in the teaching of listening different students at differentproficiency levels should be given the different comprehensible input. For theintermediate college students the teaching of phrases is an efficient way to improvelistening comprehension.