高中学生英语语法能力与英语水平相关分析 75页

万方数据华北电力大学硕士学位论文References⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．61AppendixIRevisedOralImitationTest⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯65AppendixIIWrittenMetalinguisticTest⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．．67AppendixIIITheQuestionnaire⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯74PaperPublishedinthePeriodofMAEducation．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．75Acknowledgement⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯76V 万方数据华北电力大学硕{：学位论文目录英文摘要⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯OD·O01⋯⋯⋯U0．．．．QO．．．．⋯⋯⋯⋯l中文摘要⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯06．00⋯⋯⋯⋯⋯I|表格目录⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯Ix图标目录⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯X缩略语表⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯Xl第1章引言⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯11．1研究背景⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯一11．2研究目的与意义⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．．21．3论文结构⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯。3第2章文献综述⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．4211语法能力⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯42．2显性知识和隐性知识⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．82．3语法能力与外语水平的相关国内外研究⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯13第3章理论框架⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．193．1语法的内涵与分类⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯193．2显性知识和隐性知识操作的理论框架⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯203．2．1区分隐性知识和显性知识的方法⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯203．2_2隐性知识和显性知识的测量方法⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯223．3：语水平的构建⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．．24第4章调查方法⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．274．1研究设计⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯274．1．1研究问题⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯一274．1．2研究对象⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯一284．1．3研究J：具⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．．284．2数据l发集⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯34第5章研究结果分析及讨论⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯365．1学生的屁件语法知识和隐性语法知识⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯36 万方数据华北fb力大学硕士学位论文5．2学生的二语水平⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯455．3学生的显性语法知识和隐性语法知识与英语水平相关性⋯⋯⋯⋯⋯⋯⋯495．4研究结果讨论⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．53第6章结论⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯586．1本研究的主要发现⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯一586．2启示⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．596．3研究的局限性及建议⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．59参考文献⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯61附录一：口头诱导测试题⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯65附录二：元语言知识测试题⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯一681附录三：辅助问卷⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．74攻读硕士学位期间发表的论文⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯75致谢⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯76V 万方数据华北I【_l力人学硕Ij学位论文ListofTablesTable3一lKeyCharacteristicsofImplicitandExplicitKnowledge⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯21Table3-2AnalysisoftheFiveTestsMeasuringExplicitandImplicitKnowledge．．．．23Table3-3TestforEnglishGeneralProficiency⋯．．．⋯⋯．．．．⋯⋯．．．．．．．．．．．．．．．．⋯．．．．．⋯⋯．．．．．．．．．．．．26Table4-1TestingGrammaticalStructures．．．．．．．．．．．．．⋯⋯．．⋯⋯．．⋯⋯⋯．⋯⋯．．．．．⋯．⋯．．．⋯．．．．．．．．．29Table5·-1DescriptiveStatisticAllocation⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．⋯．．37Table5-2DescriptiveStatisticsofMeanandStandardDeviationacrossSchoolsandGrades⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．38Table5-3Kruskal—WallisTestforGongqingNo．1MiddleSch001．．．．．．．．．．．．．．．．．．．．．．．．．．．．．41Table5—4StatisticsofImplicitKnowledgeamongThreeGradesonRyan’STest⋯⋯42Table5—5ResultsofExplicitKnowledgeamongThreeGrades012Ryan’STest⋯⋯⋯43Table5-6DetailedDescriptionofExplicitandhnplicitScoreonEachSpecificGrammaticalltem⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯43Table5．7ScoreofFourPartsAllocation⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．．45qable5-8GeneralDistributionofMeanScoreofEveryPart．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．46Table5-9ACorrelationAnalysisofExplicitandImplicitKnowledgeandCourseAchievementinSpearmanRank．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．，49Table5—10SpearmanAnalysisforGradeOne⋯．⋯．⋯．⋯．．．．⋯⋯．．⋯．．．．．⋯⋯⋯．．．．⋯⋯．．．．．．51Table5—11SpearmanAnalysisforGradeTwo⋯⋯．．．．．．⋯．⋯⋯．．．．．⋯⋯．⋯．．．⋯⋯．．⋯⋯⋯⋯51Table5—12SpearmanAnalysisforGradeThree⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．⋯⋯52Table5-13AnOverallPictureoftheSpearmanCorrelationAnalysisbetweenExplicitandImplicitKnowledgeandtheFourPartsacrossThreeGrades．．．．．．．．．．．．．53Table5-14StatisticsLayoutofQuestionnaire⋯．．⋯．．．．⋯⋯⋯．．⋯．．⋯⋯．⋯⋯．．⋯⋯⋯．⋯⋯．．54X 万方数据华北f乜力人学硕L学位论文ListofFigure2-1CommunicativeCompetence⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．6Figure2-2ModelofLarsen-Freeman．．．．⋯．．．．．．．．．．⋯．．．⋯．．．⋯．．．⋯．．．．．．．．．．⋯．．．⋯．．．⋯．．．⋯．．．．．．．．．⋯．7Figure2-3TheNon—interface．StrongInterfaceandWeakInterface．．，⋯．．．⋯．．．⋯．，．，．，，，．．13Figure5—1MeanScoreAllocationforSchool⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．39Figure5-2MeanScoreAllocationforGongqingNO．1acrossGrades．．．．．．．．．．．．．．．．．．．．．．40Figure5-3MeanScoreAllocationforChangjiuPrivateacrossGrades⋯⋯⋯⋯⋯⋯⋯38Figure5-4MeanScoreAllocationforJiangyiTown．．．．⋯．．⋯．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．41Figure5-5OverallDescriptionfromthePerspectiveofSchool⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯47Figure5-6TotalMeanScoreDistributionacrossThreeGradesinGongqingNO．1MiddleSch001．．⋯⋯．．⋯．．⋯．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．．。。．．．．．。．．．。．。。．，。。。．．。。．。．．．47Figure5-7TotalMeanScoreDistributionacrossThreeGradesinChangjiuPrivateMiddleSchool⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．48Figure5-8TotalMeanScoreDistributionacrossThreeGradesinJiangyiTownMiddleSchool⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯．⋯⋯．⋯⋯．．⋯．．z18× 万方数据华北电力大学硕士学位论文LlL2SLAGREUGJTTGJTTOFELListofAbbreviationsFirstLanguageSecondLanguageSecondLanguageAcquisitionGradeRecordExaminationUntimedGrammaticalJudgmentTestTimedGrammaticalJudgmentTestofEnglishasaForeignLanguageX 万方数据华北电力人学硕!L学位论文Chapter1Introduction1．1ResearchBackgroundItisacknowledgedthatgrammaractsasagreatpartinsecondlanguageacquisitionandlearning．Inviewoflanguagesystem，itisthefundamentalelementoflanguage．Fromthepointoffunction，ithasadispensablerelationshipwitheveryaspectofourlanguageuse，suchaslistening，reading,andwriting．Moreimportantly,itmostlydependsonourgrammarwhetherthesentenceswesaidcanbeacceptedornot．Fromourlearningexperience，welearnalanguagefromitsvocabulary,grammar,sentenceandthentotext．SowecanseegrammaractsasameanstohelpUSmasteralanguage．However,teachinggrammarisstillcontroversial．WhentheGrammarTranslationMethodwasprevailing，learnersattachedgreatattentionongrammar,buttheproblemwasthattheireffectwasfarfromsatisfaction．Teachersinclassjustshowedstudentsthegrammarrules，butneglectedtheenhancementoftheircommunicativeskills．Therefore，studentsmayhavealargeamountofgrammarknowledgebutwithoutanyabilitytocommunicatewith．Itistruewi廿1theclassroomteachingnow．Asentenceheardinanintensiveclasscarlexemplifythis．‘"Mymothercomef绝Idoingmyhomework．”Heknowsthemeaningbutstillcarl’tconveyincorrectsentencestructure．Theirgrammarknowledgecannotbeintemalizedintotheirgrammaticalcompetence．SoitprovedtobetessefficientGradually,thegrammartranslationmethodwasreplacedbycommunicativeteachingmethod．Teacherstaughtstudentsgrammarthroughdialogue．BasedOntheirpractice，studentslearngrammar．Astimewentby,learnersCanspeakEnglish，butmostofthemdon’tknowwhythestructureshouldbelikethisSinceexplicitknowledgeandimplicitknowledgewasfirstputforward，manycomprehensiveanddepthstudies(Ellis，2002；Erlam，2006)havebeenconductedinsecondlanguageacquisitionarea．However,thesestudiespaidgreatattentiononthe， 万方数据华北电力大学坝一l：学位论文rolethesetwotypesofknowledgeplayed．Forexample，howtomeasureexplicitknowledgeandimplicitknowledge，howexplicitknowledgetransferintoimplicitknowledgeandwhetherimplicitknowledgeisrelatedtoEnglishproficiency．InChina，inthepastyears，someresearchers(Gu，2005；Dai，2005)haveconductedaseriesofstudiesandaccumulatedabundantexperienceandachievements．However,theymainlyfocusontheories．Bydiscussingforegoingstudies，wecanseethattherearequitefewstudiesonlearner’Sgrammaticalcompetencefromthepointofimplicitknowledgeandexplicitknowledge．Besides，thosestudiesmostlyfocusedoneitherEnglishmajorundergraduatesorNon-Englishmajorundergraduates．Studiescenteredonmiddleschoolstudentsarerare．Furthermore，studiesfocusedonthecorrelationofgrammaticalcompetenceandgeneralEnglishproficiencyofmiddleschoolstudentsfromexplicitandimplicitknowledgearerarer．So，itisworthdoingthistopic1．2ResearchPurposeandSignificanceThisstudytakesseniormiddleschoolstudentsassubjectsandemployedcourseachievementtestasthetooltomeasuretheIearner’SEnglishproficiency,thepurposeofwhichistotrytoinvestigatetheexplicitandimplicitgrammaticalknowledgeofhi曲middleschoolstudentsandthecorrelationbetweenthesetwotypesofknowledgeandtheirEnglishproficiency．Comparedwithdifferentgrades．itisintendedtorevealwhat’Sthedifferenceamonggrades．Ithelpstomakeclearofwhichstructuresareeasierforstudentstograspinacquiringprocesscomparedwithothersaswell．Itisalsosupposedtoanalyzethecorrelationbetweensecondlanguagelearners’implicitandexplicitgrammaticalknowledgeandtheirEnglishproficiencyofeachpartofthecourseachievement．Itisexpectedtohelpteachersbeawareofstudents’grammaticalknowledgecompositionandhaveabetterunderstandingofstudents’grammaticalcompetence．Theyknowhowtotakeeffectivemeasurestoteachgrammar．Andthensomesuggestionsonhowtospeeduptheinternalizationof，secondlanguageknowledgeareputforward．2 万方数据华北电力大学硕士学位论文1．3LayoutoftheThesisThepapercontainsaltogethersixchapters．Thisisthebeginningofthethesisandthefollowingchaptersareorganizedasbelow．Chapter2presentsaseriesofelaboratedresearchesongrammaticalcompetencebothathomeandabroad，andintroducessomeconstituentsofsecondlanguageknowledge．Inthispart，severalissuesconcerningexplicitandimplicitknowledgearetalkedaboutaswell．Itendswithareviewofcurrentresearchesongrammaticalcompetenceintermsofexplicitandimplicitknowledgebothathomeandabroad．Chapter3addressesthetheoreticalframeworkappliedinthisstudy．Chapter4describestheinstrumentandmethodusedinthisresearchaswellasthesubjectstakingpartinthetest．Theresearchdesignandthedetailedresearchproceduresarealsoinvolved．Chapter5reportsthespecificdataanalysesfromeverysinglepointunderthehelpofspecializedstatisticalsoftwareChapter6firstlyshowsthemajorfindings，andthenidentifiesthelimitationsofthethesisandfinally,putsforwardsomesuggestionsforfurther伊ammarteachingandlearning．3 万方数据华北电力大学硕士学位论文Chapter2LiteratureReviewInordertodiscussthecorrelationbetweenexplicitknowledgeandimplicitknowledgeandtheirEnglishproficiency,wehavetoknowaboutsomebasicconcepts，suchasgrammaticalcompetency,explicitandimplicitknowledgeandtheirrelationship．Inthischapter,theseissuesarediscussedandbasedonthediscussion，thetheoreticalframework，gradually,isformed．Threemaincontentsareincludedinthischapter．Thefirstpartconsidershowgrammaticalcompetencehasbeenaddressedbypreviousoutstandingscholars(Chomsky,1965；Hymes，1971)．Thesecondreviewssecondlanguageknowledgeandtherelationshipbetweentwotypesofknowledge．Thethirdpartcriticallyintroducesthesplendidprogressachievedinsecondlanguageacquisitiononsecondlanguageknowledgeandsecondlanguageproficiencybothhomeandabroad．2．1GrammaticalCompetenceSinceAmericanlinguistChomsky(1965)firstproposedtheterminology“grammaticalcompetence”，manyaresearcher(Hymes，1971；Canale&Swain，t980)worksonthisareaandgivevariousdefinitionsunderdifferentmodels．Nowlet’Shaveabriefreview．ThetermfirstcamefromChomsky(1965)whoisanAmericanlinguist．Inhisopinion，thegrammaticalcompetencejustreferstothegrammaticalknowledgeoftheidealuserwhileneglectingtheactualuseunderdifferentcontexts．HeheldthebeliefthattoknowalanguageistoknowitsgrammarThatistosay,knowingalanguagemeansheorshehastheabilitytounderstandandcreatesgrammaticallycorrectsentences．Chomsky’S(1965)definitionofGrammaticalCompetencewasquestionedbylinguists(ttymes，1971)atthattimeforitonlyincorporatestherulesofgrammarbutwithoutanyconsiderationofactualappropriateness．Somepeopleacceptedhisideaofcompetenceandperformance，butfurtherextendedhisdefinitionofcompetence，whichreferstoboththegrammarrulesandtheabilitytouseit4 万方数据华北电力人学硕!}：学位论文appropriately．Foritscriticism，laterthecommunicativecompetencewasprevailing．ttymes(1971)putforward“grammaticalcompetence”onthebasisofChomsky(1965)toaccountforboththeimplicitknowledgeoflanguageandtheabilitytouseit．Hetakesthesocialculturalfactorsintoaccount．Hethinksthatgrammaticalcompetenceisonlyoneaspectthataffectspeople’Scommunicativecompetence．Thecombinationofthisfactorwithotherfactorstogethercanbeusedtodecidewhethertheutterancesthepeoplespeakcanbeacceptedornotincommunication．Specificallyspeaking，accordingtoHymes(1971)，hiscommunicativecompetenceismadeupoffourcomponents．(1)．Possibility-一whether(andtowhatdegree)somethingisformallypossible，i．e．theabilitytoproducegrammaticalsentences；(2)．Feasibility⋯whether(andtowhatdegree)somethingisfeasible，i．e．theabilitytoproducesentenceswhichcanbedecodedbythehumanbrain(3)。Appropriateness⋯whether(andtowhatextent)somethingisappropriate，i．etheabilitytousecorrectformsoflanguageinaspecificsocio．culturalcontext(4)．Performance⋯whether(andtowhatdegree)somethingisinfactdoneHymes’(1971)researchhasafeaturethathisconcentrationisshiftedfromthesystemoflanguageitselftotheactualuse．Widdowson(1989)foughtagainstHymes’(1971)pointofview．Hebelievesthatcompetencealtogetherconsistsoftwoparts--knowledgeandability．ThefourparametersHymes(1971)mentionedcanbeclassifiedintogrammaticalcompetenceandpragmaticcompetence，inwhichgrammaticalcompetencereferstotheparameterofpossibilityandotherthreeparametersbelongtopragmaticcompetence．Inhisopinion，theparameterofpossibilityisleamer’Sabilitytocreategrammaticalsentences，whichisthesametoChomsky’S(1965)idea．Thus，grammaticalcompetenceisregardedasthebasicrequirementofcommunicativecompetence．Inthefollowingyears，Hymes’(1971)ideaWaSinheritedandfurtherenhancedbyotherlinguists，suchasCa，naleandSwain5 华北电力人学硕士学位论文《1980)，Canale(1983)andBachman(1990)CanaleandSwain(1980)putforwardamorecomplicatedtheoreticalmodelaboutcommunicativecompetence．Firstlythreearethreedimensions，andlateritisfurtherextendedintofourcomponentsbyCanale(1983)，includinggrammaticalcompetence，sociolinguisticcompetence，discoursecompetenceandstrategiccompetence．(Seethechartbelow)Figure2-1：Communicativezkq20160118CompetenceFromthefigure，wecanfindoutgrammaticalcompetenceistreatedasonedispensablepartofcommunicativecompetence．Inhispointofview,grammaticalcompetencereferstothemasteryoflanguagesystem，namelyphonology，vocabulary,spelling，wordformation．(Canale&Swain，1980)Itpaysgreatattentiontotheknowledgeandskillsneededtounderstandandconveyappropriatelythemeaningoflanguage．(Canale，1983：7)BasedonCanaleandSwain’S(1980)framework，Bachman(1990)proposesalittlemoresophisticatedanddetailedmodelaboutcommunicativecompetence．Thecontentsofitcontributetoatreediagram．Inhisopinion，communicativecompetenceconsistsofthreepans：languagecompetence，strategiccompetenceandphysiologicalmechanisms．Thelanguagecompetenceissubdividedintotwosmallscaleelements，includingorganizationalcompetenceandpragmaticcompetence．TheorganizationalandtextualcompetencecompromisesgrammaticalcompetencecompetenceTherefore，inBachman’S(1990)model，grammaticalcompetenceismore6万方数据 华北电力大学硕士学位论文detailedanditmainlydealtwiththecompetencyinmorphology,vocabulary,syntaxandgraphology．Inanotherword，grammaticalcompetenceinvolvesthedimensionoflanguagebothinformandstIuchlre．DifferentvoiceongrammaticalcompetenceisheardinthesuccessorperiodLarsen-Freeman(1991)pointsoutthatonlythreedimensionsareexistedingrammaticalcompetence，whichareform，meaninganduse．Inhiseyes，thethreecomponentshaveamutualconnectionwitheachother．Itismeaninglesstoputgrammarinaseparatesituation．Thus，ifoneelementchanges，itwillaffecttheothertworemainingelements．Thisinterrelatedrelationshowsthatthesystemisholistic．Alargenumberoflinguists(Celce—Murcia，2002；Yule，2002；He&Chen，2004)agreewithLarsenFreeman(1991)．Forexample，He&Chen(2004)thinkthatgrammaticalcompetenceiscomposedofphonology,morpheme，syntacticstructures，semanticmeaningandpragmaticusezkq20160118Figure2—2：ModelofLarsen—FreemanThroughoutthereviewofthedefinitionsofgrammaticalcompetencefromtheabovementionedlinguists，wecanseeitisstillcontroversialonthisitem．Theirdefinitionsvaryfromonetoeachotherandlittlecensushasreached．However,byanalyzingthecontentofthosedefinitionsongrammaticalcompetenceindepth，we锄figureoutallofthedefinitionshavesomethingincommon．Therearetwobasic，elementsincorporatedingrammaticalcompetence，whicharethegraspedrulesof7万方数据 华北电力大学硕_上学位论文languageandaccuratelyandappropriatelyuseoflanguageinagivencontext．Thefirstpartwerefertoexplicitgrammaticalknowledge，andthelatter,werefertoimplicitgrammaticalknowledge，whicharehighlyconsideredinthispaperSincetheauthorexploresthecorrelationbetweengrammaticalcompetencyandtheirEnglishproficiency,andbasedonthediscussionongrammaticalcompetence，inthepaper,theauthorwouldliketoexplaingrammaticalcompetencyfromexplicitknowledgeandimplicitknowledge．So，itisquiteurgentlyneededtodiscussthesecondpartofthischapter-·-·-explicitknowledgeandimplicitknowledge2．2ExplicitKnowledgeandImplicitKnowledgeSincethetwoterms“explicitknowledge”“implicitknowledge"’werefirstcareabouttheroleproposed，someresearchers(Bialystok，1979；Krashen，1981)，whoofeachknowledgeplaysinsecondlanguageacquisition，begantodoalotofworkontellingfromthesetwoconcepts．Krashen(1981)usetheterm“acquisition’’andzkq20160118knowledge’’inhisMonitor“learning：’toreferto“implicitknowledge”and“explicitModel．Inhisidea，theimplicitknowledgeisthatthetargetedlanguageactivatedbylearnersinthegivensituation．Bialystok(1979)holdsthebeliefthatthecontentsof．thesetwotypesofknowledgearequitesimilar,andtheybothdirectlypointstotheandSOon．Afterthat，moreworktargetedlanguage，includingvocabulary,grammarhasbeendoneondistinguishingthesetwoitemsfromdifferentdimensions．Intheabriefreviewonthedefinitionsandtheirfollowingpart，theauthorwillgiveinterrelationship．2．2．1ImplicitKnowledgeEllis(2004)indicatesNativespeakers’implicitknowledgemightberule。basedareintuitivetoorformulaicknowledgc．Whenchildrenlearntheirfirstlanguage，theymasterthefeaturesandrules．Thisknowledgeareincidentallyacquiredandstoredintheirmindvividly；Thenativespeakersorcompetentlearnersuselanguagewithoutawarenessofactivatingthisknowledge；instead，theyCanapplythisknowledgeto8万方数据 华北电力大学硕I‘学位论文makeendlessgrammaticallycorrectsentencesspontaneously．Thatistosay,whentheyspeak，theyarenotawareofsearchingtherelatedknowledgetoexpresstheirideas．SothisknowledgecannotbenoticedeasilyandcannotbetakenouttoanalyzeKrashen(1983)showsthatimplicitknowledgeCanalsobelearnt．Butitisnotthesamewayaswelearntexplicitknowledge．Inhisbook，hecallsthisprocessofpickingup“acquisition”．Thisacquisitionisfromthenaturalenvironmentandisautomaticallyinherited．NotgreateffortsarerequiredduringthelearningprocessSchmidt(2001)holdsthebeliefthatimplicitknowledgeiswhatlearnershaveintheirmind，buttheycannotusewordstoexpress．Sincethecannotimplicitknowledgebeanalyzedandnoticed，SOitistheexplicitknowledgethatistheobjectiveofourlearning．2．2．2ExplicitKnowledgeEllis(2004)pointsoutthatexplicitknowledgeistheknowledgethatlearnersuseconsciously．Thedetaileddefinitionzkqcanbe20160118seenbelowExplicitL2knowledgeisthedeclarativeand妒已nanomalousknowledgeofthephonological"lexical,grammatical,pragmatic,andsociocriticalfeaturesofanL2togetherwiththemetalanguageforlabelingthisknowledge．Itisheldconscious&and括learnableandverbalizable．Itistypicallyaccessedthroughcontrolle@rocessingwhenL2learnersexperiencesomekindoflinguisticdifficultyintheuseoftheL2．LearnersvaryinthebreadthanddepthoftheirL2explicitknowledge．(Ellis，2004：244—245)Asislistedabove，weknowexplicitknowledgeisthedescribedknowledge．Itisverbalizedandlearnersusethisknowledgeconsciously．Whenwelearnalanguage，theruleswelearntistheexplicitknowledge．Nativespeakersarenotintendedtospeakcorrectlybuttheycanspeakgrammaticallycorrectsentencesandcanjudgewhetherasentenceisgrammaticallycorrectwithoutconsciousnessandgreatefforts9万方数据 华北Ii_!-力大学硕，I二学位论文withbutnotHowever,inthissituation，wesaythelearnersequipimplicitknowledgetheexplicitknowledge．Becausetheyjustgiveouttheirjudgmentbyintuitive．Theyjustknowthesentenceiswrong,buttheydon’tknowwhyitiswrong．Ifthelearnersknowthesentenceisgrammaticallywrongandcanexplainthereasonwhyitisincorrect，thenwecansaythelearnershavemasteredtheexplicitknowledgeBialystok(1994)believesthattheexplicitknowledgelearningiswithoutlimitationwhiletheimplicitknowledgelearningisaffectedbyage．Explicitknowledgeandimplicitknowledgehavenodifferenceswhetherundertheenvironmentofsecondlanguageorforeignlanguages．Forthisaspect，explicitknowledgepresentssomefeaturesoflanguage．Learnersjustobtainthelanguage，butcannotapplywhattheylearntothecorrectactualuse．Thisknowledgecannotensurethatlearnersmodifymistakesautomaticallyintheprocessoftalkingorspeaking．Justaswccanseeinourdailylife，somestudentsarecalled‘‘highscoresandlowabilities”．Theyaregoodatgranmlaticalrules，butdon’texcelinoutputtasks．Whenzkq20160118toexplicitknowledge．Therefore，inthetheymeetdifficulties，theymostlyappeallanguageoutput，thepoorerthelanguageability,themoretheyturntoexplicitknowledge．(Paradis，1994)2．2．3TheRelationshipbetweenExplicitandImplicitKnowledgeForfour-yeardevelopment，ithasbeenheatedlydiscussedwhetherexplicitknowledgeandimplicitknowledgecaninterferencebetweeneachotherandthreeinfluentialmodelshavebeentakenshaped，whicharenoninterfaceposition，thestronginterfacepositionandtheweakinterfaceposition．2．2．3．1TheNoninterfacePositionAlargenumberoflinguists(Ellis，1993．2004．2006；Hulstijn，J．&Hulstijn，W1984；Krashen，1981．1983；Paradis，1994)thinkthesetwotypesofknowledgeareindependent．Theyareopposedtotheideathatexplicitknowledgecandirectlyconveatoimplicitknowledgeandthepossibilitythatimplicitknowledgetransforms10万方数据 华北LU力大学颂士学位论文intoexplicitknowledge．Evidencesareshowedonthefollowingthreeparts：(1)Thesetwotypesofknowledgereferstotwodifferentlearningsystem．(2)Thesetwotypesofknowledgearestoredindifferentplacesinourbrain．(3)Thesetwotypesofknowledgeareindifferentprocess，inwhichexplicitknowledgeisthecontrolledprocesswhiletheexplicitknowledgeistheautomaticprocess．Krashen(1981)stronglyadherestothispoint．Hethinksexplicitknowledgecanonlyplayapartinsupervisingtheoutputwhichcomesfromdirectlyimplicitknowledge．However,Paradis(1994)arebitteragainstthisnoninterfaceposition．Hewhathehaslearntinandshowsthatlearnerswho1camtwoquotesneurologylanguageswilllosetheirabilitytospeaktheirmothertonguewhentheyalwayscommunicateinL2．Healsoarguesthatpracticefunctionslittleintransformingexplicitknowledgeintoimplicitknowledge．Buthedoesn’tshowanyevidence．JustlikeKrashen(1981)，Ellis(2004，2006)alsoagreesthatthesetwokindsofknowledgearedichotomouszkq201601182．2．3．2．TheStrongInterfacePositionSomeresearches(DeKeyser,2003；Dienes&Perner,1999)，however,believethetwotypesofknowledgeascontinuous．Theyclaimthatnotonlyexplicitknowledgebegotfromimplicitknowledgebutalsoexplicitknowledgecallbedirectlytransformedintoimplicitknowledgebymeansofpractice．FrenschandRlinger(2003)putfbrwardthatwhatourlearners’firstcontactistheinputofrulesandthroughourpractice，itconvertstoimplicitknowledgeandfinallywegettheresult．Inthecourseofpractice，learnersformthehypothesisandcorrectthewronghypothesisandatlastlearnthestructuresorrules．Itshouldbenotedthatexplicitknowledgecannotbeadaptedintoimplicitknowledgeintuitivelyjustviapractice．Inthisprocess，learnershavetopaygreatattentiononincorrectnessoftheirpe晌珊anceandthenformtherighthypothesisEllis(2005)doesn’tagreewiththepointthatexplicitknowledgeandimplicitknowledgecanbemutuallyconverted．Butheemphasizestheimportanceofpractice11万方数据 万方数据华北电力大学硕士学位论文andsumsupfourfunctionsofit．(1)Practicecanhelplearnerspossessthedeclarativeknowledgeeasily．(2)Practicecalldoalottomakestudentsformimplicitknowledge(3)Practicecanhelpchunkingofimplicitknowledge．(4)Practicecanhelptoactivatetherelatedknowledgewhileproduction．Asaresult，wecanseepracticeplaysarolebetweenexplicitknowledgeandexplicitknowledge2．2．3．3TheⅥ，eakInterfacePositionTheweakinterfacepositionactsasallintermediatestatebetweenstronginterfacepositionandnoninterfaceposition．Ithassomedifferencefromthetwo．Itbaresthecontentthatexplicitknowledgeandimplicitknowledgearedifferent，butitalsoagreeswiththeideathatexplicitknowledgecanbetransformedintoimplicitknowledge，However，thistransformationisnotwithoutcondition．Ellis(2005)istherepresentative．Ellis(2005)reviewsthethreeexistingversionsofweakinterfaceposition．ttefindsoutthesethreeversionshaveincommonthatexplicitknowledgeislikelytoconvertintoimplicitknowledgeinsomedegree．Restrictionsarecomposedonhowandwhenitwillhappen．ThefirstversionmaintainsthatexplicitknowledgeCanIransformintoimplicitknowledgethroughpracticeonlywhenthelearnerisequippeddevelopmentallytoacquirethelinguisticform．Thesecondversionarguesthatexplicitknowledgecontributescircuitouslytotheacquisitionofimplicitknowledgebyraisingdiscriminatoryattentiontocertainforms(Ellis，1993)．ThelastversionheldbySchmidt&Frota(1986)andSharwood—Smith(1981)putsforwardatheorythatlearnerscangenerateoutputbyapplyingtheirexplicitknowledge，whichservesasinputtotheirimplicitlearningprocessestheotherwayround。TypeA(a)TheNon—interfaceposition————————————’practice——————————◆TypeBExplicit／declarativeExplicit／proceduralTypeC——————————◆communicative————————————+TypeD。Implicit／declarativelanguageuseImplicit／procedural 万方数据华北电力人学硕士学位论文(b)TheStrongInterfacePosition(C)TheweakInterfacePositionTypeA风脚ir锄nwgrammarinstructionTypeJccommunicativeTypeBExplicit／proceduralTypeDImplicit／declarativelanguageuse／practiceimplicit／proceduralFigure2·3：TheNon—interface．StrongInterfaceandWeakInterfacePositionThefigureaboveshowsclearlythecomparisonofthreepositions．Innoninterfaceposition，thereisnowayofconvertingexplicitknowledgedirectlyintoimplicitknowledge．Bypractice，onlyexplicit／declarativeknowledgeCalltransformintoexplicit／proceduralknowledge．Instronginterfaceposition，throughpractice，explicit／declarativeknowledgecanautomaticallyconverttoimplicit／proceduralknowledge．Accordingtothefigureshowinginweakposition，explicitdeclarativeknowledgemaydevelopintoimplicit／declarativeknowledgeontheconditionsthatlearnersarefullygraspedtoincludethenewknowledgeintotheirinterlanguagelearningorganisms2．3GrammaticalCompetencefromthePerspectiveofExplicitandImplicitKnowledgeInrecentyears，there’Sheateddiscussionandstudiesongrammaticalcompetencefromtheperspectiveofexplicitknowledgeandimplicitknowledge．Inthefollowingpart，theauthorintroducessomerelatedstudiesbothhomeandabroad． 万方数据华北电力大学硕．J：学位论文23．1StudiesAbroadManystudieshavebeenlauncheddeeplytoinvestigatethelearners’linguisticcompetenceintermsoftherolethatexplicitandimplicitknowledgeplayinthecourseoflanguageusesincetheearly1980sBialystok(1979)conductsafamousempiricalresearch．Inhisstudy,thesubjectsare317studentswhoseL2isFrench，andtheresultsshowedmatundertimepressureandmorerelaxedconditions，L2learnerswithdifferentproficiencylevelsdidthesamewellingrammaticaljudgmenttest；but，whentheyarcrequiredtoanswerwheretheproblemisorwhattheruleofthesentenceisviolated．Basedonthedatacollectedfromhisresearch，Bialystokformsthehypothesethatlearnersmakegrammaticalityjudgmentsaccordingtotheirimplicitknowledgeandwhenmoredetailedquestionsareasked，theyprefertoturntoexplicitknowledgeforhelpHulstijn，J．＆Hulstijn，W．(1984)findoutthat，inastoryretellingtask，learnerswithDutchasaL2doworseonwordorderruleswhentheydon’thavegraspedexplicitknowledgeoftheserulesthantheyhave．Scott(1989)launchesaresearchtotestifytheeffectofgrammarinstructionseparatelyusingexplicitandimplicitways．ThesubjectsofhisstudyarestudentsfromadvancedFrenchclass．TheexplicitlearningwayhappenswhentheteachermanagetoteachacertainFrenchrule,studentsareopenedtotherulesfirstbeforeexamplesareshowntothemtohelpthemabsorbtherules．Theyfirsthaveaccesstoastorycontainingthetargetedgrammarrules，andthentheycarefullylistentothestoryagainandagain．Theresultsindicatethat，inthegrammartestsoforalandwrittenforms，thoughstudentsintheimplicitlearninggrouphavebeenopenedtothestorymanytimesmorethantheexplicitlearninggroup，thestudentsintheexplicitlearninggroupfargobetterthantheimplicitlearninggroupGreenandHecht(1992)conductedalarge—scaleempiricalresearchaboutthelearners’linguisticcompetencefromthepointofexplicitandimplicitknowledge．Thesubjectsofthisexperimentare300German-speakingEnglishlearnersfromGermansecondaryschools，50Englishnativespeakersandauniversitywithexposureto14 万方数据华北I乜力人学硕上学位论文formalinstructionvaryingfrom3to12years．Researchersshowsentenceswithdiversifiedgrammaticalerrorsandallofthesubjectsaredemandedtocorrectthesentencesandtelltheresearcheswhattheruletheyhavebeenviolated．ThestudyresultsarethattheEnglishnativespeakerscancorrectmoreerrors(96％)thantheGermanEFLleamers(78％)，however,theEFLlearners(46呦excelslightlyinexplainingerrorsthanthenativespeakers(42％)．Thisindicatesthatthelearnersabilitytoexplaintherulesisfarbehindtheirabilitytocorrecttheerrorsandtheymorerelyonimplicitknowledgetomodifysomeerrors．Thereislowconnectionbetweenexplicitknowledge(ruleknowledge)andimplicitknowledge(theabilitytocorrect)．TheresultsalsogiveevidencetoprovethatthenativespeakershaveextensiverelianceonimplicitknowledgeHartandEllis(1998)applyanumberoftests(grammaticalityjudgment，oralproduction，metalinguisticknowledge，SecondaryLevelEnglishProficiencyTest，theTestofEnglishasaForeignLanguage，)toinvestigatethisissue，andtwofactorsarefoundwhichcanbetranslatedasexplicitandimplicit．Theirresultsshowthatthesetwofactorsaredifficulttointerprettoo，but，because，astheymakeclearthemselves，noneofthosetestsispuremeasuresofeitherexplicitorimplicitknowledgeMoreover,sinceonlyonestructure(verbcomplements)isusedinthisstudy,itoccurstobeacasewhereitisverydifficulttoformarule．whichmakesexplicitknowledgeatamuchdisadvantage；andthegeneralizabilityoftheresultsisofmanyquestions．MacroryandStone(2000)findthat，studentsstarttohaveafairlygoodmasterofthemorphologyoftheFrenchpresentperfecttenseinthattheyareabletoprovideexplicitrulestatementsorgethi811marksonadiscrete—-pointgap--fillingtestafterfourorfiveyearsofFrenchstudyinaBritishsecondaryschool，but，inspontaneousdiscourse，theyareinclinedtoomittheauxiliarymostofthetime，exceptinlargelyformulaieutterance．UsingaseriesofteststhatarcintendedtomeasureexplicitandimplicitL2grammaticalknowledgeofseventeengrammaticalstructures；Ellis(2005)studiestherelationbetweenlearningdifficultyandthetwotypesofknowledgeare．Inhisstudy,theresultsshowth，atstructuresthataredifficultintermsofexplicit 万方数据华北f乜力大学硕士学位论文knowledgearealwayseasyintermsofimplicitknowledgeandsometimesviceversaandthat，overall，thereisnocorrelationbetweentherankordersofdifficultyofseventeengrammaticalstructuresforthetwotypesofknowledge．Acorrelationanalysisdemonstratesthatthestructureschangeastowhetheritisimplicitorexplicitknowledgeofthemthatisquiterelatedtothemeasureofgenerallanguageproficiency．AregressionanalysisprovesthatbothtypesofknowledgeCanusetopredictgenerallanguageproficiency．2．3．2StudiesatHomeOnL2grammaticalcompetence，therelevantresearchesathomefromtheperspectiveofexplicitknowledgeandimplicitknowledgemostlyemphasizeontheoricsZhang(2004)dealswithEnglishgrammarteaching．Hestatesbrieflythatduetothehi曲pressurecomingfromexaminations，inclass，explicitknowledgealwaysbecomesthemainconstituentofEnglishgrammarotherthanimplicitknowledge，andmoreover,thesignificanceofdevelopingstudents’implicitgrammaticalcompetencewasoftenignored．Ontheotherhand，communicativelanguageteachingattachesgreatimportanceonthepracticaluseoflanguagewhichresultsinthedisregardofgrammarrulesandmuchlessattentiononimplicitinstruction．Astwoofthemostbasiclearningmechanisms，implicitandexplicitknowledgelearningpossesownfeaturesandcharacteristicsandwithdifferentoperatingqualificationsandascendency．Hence，theytwocan’tbeseparatedfromonetoanotherduringtheprocessoflearningandgreaterattentionoughttobepaidonthecooperationofthetwotypesofknowledgeforeffectiveEnglishclassroomgrammarteaching．Dai(2005)discoversthattheactualapplicationofthetwoconceptions”explicit”，’’implicit”andseveralotherrelatedconceptsdogreatharmandhadgreatimpactontheempiricalstudydesignanddatacollectionandanalysis，whichcausedmutuallycontradictoryresults．，Underthein—depthanalysisoftherelationshipbetweenexplicitandimplicit 万方数据华北电力大学硕一卜学位论文knowledge，Gu(2005)hadfaithintheideathattherewereintersectsontheirretrievalpathsbetweenthesetwotypesofknowledge．Theyarenottwocompletelydifferentkindsofknowledge．Sun(2007a)believesthatdeclarativeknowledgelaysthebasisforChineselanguagelearners’perceiveofEnglishandisofmuchimportancetothefurtherdevelopmentoftheirlanguageawareness，culturecognitionaswellastheirlinguisticcompetence．Yang，L．andYang，Z．(2007)furtherconductedaninvestigationontheanalysisofthefunctionofexplicitknowledgeinthecourseofgrammarlearningandtheyholdthebeliefthatitwasexplicitknowledgethatlaysthefoundationforsecondlanguagelearning．Nonetheless，nearlyalltheSLAtheoriesaremainlyfocusontheexplanationofhowstudents’implicitknowledgedeveloped．(Ellis，2006)Theobjectiveofmasteringexplicitknowledgeistopavethewayforthewellcommandofimplicitknowledge．Implicitknowledgefunctionsalotduringthesessionofsecondlanguagelearning．Inhisanotherbook，hetalksaboutthefunctionsofproceduralanddeclarativeknowledgeonthelearningprocesslanguageleamers．(Sun，2007b)Du(2009)analyzesthecorrelationbetweengrammaticalknowledgeandcompetenceundertheprinciplesofexplicitandimplicitlearningtheory．Sheconstructsanactualgrammaticalteachingmodel．Inthismodel，shetakestheinternalizationofgrammaticalknowledgeandtheenhancementofgrammaticalcompetencetogether．Zeng(2009)launchesanempiricalresearchontheuseofimplicitknowledgeofEnglishmajorstudents．TheresultshowsthatimplicitgrammaticalknowledgehaspositivecorrelationtotheirEnglishproficiency．ZengandBai(2010)，taking92Englishmajorstudentsasthesubjectsandadoptingoralelicitedimitationtestasthetool，trytotestifthereiscertainrelationshipbetweenstudents’performanceandtheirEnglishproficiency．Thescoresdemonstratethatstudents’generalimplicitknowledgeisfarfromsatisfactionandtherearesignificantdifferencesamonggroups．Du(2010)discussesthekeyreasonthatimpededtheenhancementofstudents’grammaticalcompetencefromtheperspectiveofrelationshipbetweengrarmnatical 万方数据华北电力大学硕士学位论文knowledgeandcompetence．Ontheotherhand，sheanalyzestheteachingpatternsofGrammar．Attheendofthethesis，shepointsouttheimplicitlearninganditsimplicationsongrammarteachingandlearning．Becauseoftheparticularrequirementsoflearners’competencewithindifferentgrades，sheagainconductsaresearchonclearlydefiningtheconceptsfirstlyandthendiscussingtheissueofthegrammarcompetencyconstructionZhangandZhao(2011)detectthatthecontributiondegreeofthesetwotypesofknowledgedevotedtoreadingcomprehensionrespectivelywiththesubjectsofstudentswhoarefightingfortheirmasterdegree．Theresultshowsthatimplicitknowledgecontributesgreatertoreadingcomprehensioncomparedtoexplicitknowledge．TheresultoftheirgrammaticalcompetenceismainlyderivedfromtheirscoresOilseveralspecificstructuresorlanguagepatterns．DuetothelackoftheresearchesonEnglishmajorstudents，inacertainsense，thisstudyisquitemeaningfulforitdoesagreathelptosecondlanguagelearnersininternalizingexplicitknowledgeandpromotingtheirgrammarcompetence．WiththepurposeofenhancinggrammaticalcompetenceofEnglishlearnersandrecommendingsomepositiveandconstructiveproposalstoEnglishteachingandlearning,thisstudytakesalargeamountofstudentsassubjectswhoarefromauniversitywithdifferentgradesandaspecialcross—sectionmethodisapplied，whichaimstomakeclearofpresentstateanddistinctcharacteristicoftheirgrammaticalcompetenceforeachgrade．Throughthedetailedreviewonempiricalstudyabove，wecandrawtheconclusionthatnocensushasbeenreachedonthecorrelationbetweenexplicitandimplicitknowledgeandtheirEnglishproficiency．Besides，Mostofthestudiesdiscusstheproblemsontheoreticallevel．MoreoveLintermsofthesubjectstakingpartintheexperiments，universitystudentsaretheirfirstchoice．Fewerresearchescenteredonmiddleschoolstudents．Therefore，empiricalstudiesaredesirable．Andempiricalstudiesonmiddleschoolstudentsarcmoreappreciable．Thispaperisfinattempttomeetthedeficiencybyinvestigatingthecorrelationbetweengrammaticalcompetenceandsecondlanguageproficiencyamongthemiddleschoolstudents． 万方数据华北IU力人学硕士学位论文Chapter3TheoreticalFrameworkThischapterlaysoutatentativetheoreticalframeworkofGrammar,explicitknowledgeandimplicitknowledge，measurementsforsecondlanguageproficiency3．1TheConnotationandClassificationofGrammarManyscholarshavedivergentviewsondefinitionofgrammar,anduntilnow,theydidn’tcometoanagreement．RadfordthinksGrammarreferstotwodifferentbutinterrelatedlanguageknowledges，whicharemorphologyandsyntax．Itistostudytheinterpretationofwords，phrases，sentencestructuresandrules．Fromkin(1988)believesgrammarcontainsvoice，vocabulary⋯thesmallestunitofsemantic，andsetofrules．Namely,itincludesphonetics，lexicology,semanticsandsyntax，etc．Lyon(SitefromStem，1983)regardedgrammarasadescriptionofsetofrules．Hethinksofgrammargenerallyasthewaystocombinelanguagegrammarwordsintoalargemeaningfulunit．“LinguisticDictionary’’inFrencheditionshowsthefollowingfourkindsofdefinitions：1)Ageneraldescriptionoflanguage，mainlyreferstoatheoreticaldescriptionofitsorganizationalprinciples，includingphonemics，syntax，lexicologyandsemantics．Thissenseofgrammarisgrammarpatterns．2)Aterminologyincontrastwithphonemics，mainlystudyingthechangeofmorphemeanditscombinationrules．3)Onlyadescriptiontothegrammaticalmorphemes，suchasarticles，conjunctionsandpositions，notincludingthegeneralrulesofmorphemeandtheircombination．4)Inthetransformationalgrammar,grammarreferstopeople：Sideallanguageskills，namelytheabilitytounderstandtherelationamongthemeaningsofcertainwords．Fromthegrammarteachingsystemandlonglastinggrammarteachingbattle，wecallreachthecensusthatthegenerallanguageteachersarcinlinewiththesecondedition．ThefirstdefinitioniSthe1atestunderstandingofgrammarformedwiththe，developmentofmodemlinguistics．Thethirdandfourthtypesofdefinitionaretoo29 万方数据华北电力人学硕士学位论文narrowtoaccept．Therefore，inthispaper,wediscussgrammarmainlyfromthefirstandsecondsenseHistorically,grammarisdividedintoprescriptivegrammaranddescriptivegrammarbecauseoftheirdifferentguidelines．Prescriptivegrammarisalsocalledtraditionalgrammar,withtheaimtoradicalandeliminatingfallacy．Itignoresthechangeofthelanguagephenomenon．Basedonlanguagefacts，descriptivegrammarrecordsthelanguagefactsobjectivelyandthensimplyforrnulatestherulesillustratinggeneratedsentences．Fromthedescriptivemethod，itisclassifiedintocontrastiveandcomparativegranlmaLFromthepurpose，itismadeupofpedagogicalgrammar,referencegrammarandlinguisticgrammar．Inlinguisticgrammar,thereexistscasegrammarandstratificationalgrammarThereisnodoubtthatdifferentunderstandingofgrammar,variedknowledgeoftheessenceoflanguageandvariousviewstowardsforeignlanguageteaching，leadstovariousclassificationofgrammar．Inthispaper,theauthorlikestoinvestigatedescriptivegrammarforitisquitereachabletohighschoolstudentsInthispaper,withregardtogrammar,theauthorselects17grammaticalstructuresbasedonaseriesofcriteria．ThespecificdescriptionofthemwillbediscussedonChapter4．3．2TheoperationalizationofExplicitandImplicitKnowledge3．2．1TheDistinctionbetweenExplicitKnowledgeandImplicitKnowledgeTherearealtogethersevenfactorstotellimplicitknowledgefromexplicitknowledge，namelyawareness，typeofknowledge，systematicityandcertaintyofsecondlanguageknowledge，accessibilityofknowledge，useofsecondlanguage，self-reportandleamability．(Ellis，2004．2005)(Seethesheetbelow)Krashen(1981)，pointsoutthatimplicitknowledgeistreatedasnonverbalizableproceduralknowledge20 万方数据华北电力大学硕卜学位论文oflanguagethatisunconscious，unanalyzedandautomatized，andhencedoesgreatgoodtospontaneouslanguageperformance．ThisvariedkindofknowledgecanbeacquiredbynativespeakersorL2learnersinnaturalisticatmosphere．Explicitknowledge，however,isinclinedtobelearnedratherthanacquired，andisconscious，declarative，verbalizable，andaccessibleonlybycontrolledprocessingwhenL2learnersexperiencelinguisticdifficultyintheuseofL2．Itmaybeappliedonlyinsituationswherethereistimeavailableformonitoringtheoutput．Concerningthenatureofrepresentationofimplicitknowledge，therewasaheateddebateonwhetheritis。memory—based’or‘rule-based’．(Bialystok，1990)Itisnowgenerallyrecognizedthatimplicitknowledgeconsistsofbothformulaiclinguisticcomponentsbasedonmemoryandabstractandstructuredrule．(Ellis，1993)HanandEllis(1998)claimedthatadistinctionneedstobemadebetweenexplicitknowledgeasanalyzedknowledge(1alowledgeaboutL2itemsandstructuresofwhichleamersareawarealthoughnotnecessarilyfullyconscious)andasmetalanguage(thelanguageusedtoanalyzeordescribealanguage)．Ellis(2005)furtherproposedthatthetwocomponentsofexplicitknowledge--analyzedknowledgeandmetalanguagecanbeseparatelymeasuredbyUntimedGJTandmetalinguisticknowledgetest．Table3-1：KeyCharacteristicsofImplicitandExplicitKnowledgeCharacteristicsImplicitknowledgeExplicitknowledgeAwarenessIntuitiveawarenessofConsciousawarenessoflinguisticnormslinguisticnormsTypesofknowledgeProceduralknowledgeofDeclarativeknowledgeofrulesfragmentsgrammaticalrulesandfragmentsSystematicityVariablebutsystematicAnomalousandknowledgeinconsistentknowledgeAccessibilityAccesstoknowledgebyAccesstoknowledgebymeansofautomaticmeansofcontrolled 万方数据华北电力大学硕士学位论文processingAccesstoknowledgeAccesstoknowledgeUseofL2knowledgeduringfluentperformanceduringplanningdifficultySelf-reportNonverbalizableVerbalizableLearnabilityPotentiallyonlywithinAnyagecriticalperiodAsforthesystematicityandcertaintyofL2knowledge，weshouldbearinmindthattheoutcomesofthepsychometricstudythatEllis(2005)conducteddidnotprovideevidencetosupportthehypothesesconcerningthesetwoaspects．Itcouldbepossiblethatitistheconstructvalidityofthefiveteststhatisproblematic，orthespecificationsoftheaspectsintermsofimplicitandexplicitknowledgeareinneedofsubstantiation．3．2．2TheoperationalizationofExplicitKnowledgeandImplicitKnowledgeItisnotonlynecessarybutalsoimportanttoproposemethodstomeasurethetwotypesofknowledgesincewehavementionedthesetwokindsofknowledgeabove．Nowlet’SreviewsomemethodsthathavebeenusedinsecondlanguageacquisitionVariousmethodsareusedbylinguists(Doughty,2003；Han&Ellis，1998)totestexplicitandimplicitknowledge．Doughty(2003)givesapictureofmeasuresthataretypicallyemployed．Therearefourkindsofmethodsincludedaltogether：firstly,constrained，constructedresponse；secondlNselectedresponses；thirdly,metalinguisticjudgmentresponses；andthelastoneisfreeresponses．Thesefourmeasuresrevealtwocharacteristics．OneiStotestwhetherformiSinvolvedincomprehensionoroutputthesecondlanguage，theotheristoexaminetowhatextentthatlanguageuseiscontrolledorfreeHanandEllis(1998)adoptthefourinstrumentstomeasurethetwotypesofknowledgeusedinverbcomplementstructures．Theyare：(1)AnOralProductionTest；22 万方数据华北电力人学硕I：学位论文(2)AnUntimedGrammaticalityJudgmentTest(UGJT)；(3)ATimedGrammaticalityJudgmentTest(TGJT)；(4)Aninterview．TheOralProductionTestandtheTimedGrammaticalityJudgmentTestareconductedwithintimelimitation，whichareemployedtoexaminetheimplicitknowledge．TheUntimedGrammaticalityJudgmentTestandallinterviewaredesignedtotestexplicitknowledgeErlam(2006)suggeststhatElicitedOralImitationmayu．setomeasuretheimplicitknowledgebecauseitsconditionsseemtobecontrolled．Comparedwithfree-languageproductiontest，thistestismorelikelytobeeasilyoperatedandevaluated．Itisalsopossibleforresearcherstomakeclearofeachtargetedspecificlanguagestructures．ZengandBat(2010)employmanywaystoteststudents，namelyoraltest，oralelicitedimitationtestandtimedgrammaticalityjudgmenttest．Throughthesetests，theyattempttoanalyzetheadvantagesanddisadvantagesofeachmethodAtlast，theyfindoutitisoralelicitedimitationtestthatisaneffectivewaytomeasurestudents’implicitknowledgeEllis(2004，2005)emphasizestheimportanceofdevelopingvalidandreliablemethodstomeasureimplicitandexplicitknowledge．Hegazesatthepreviousmethodsusedinsecondlanguageacquisitionresearchesanddoesquitealargenumberofexperimentsandcarefullyevaluatestheproblemsexistinginthosemethodologicalmethods．(Green&Hecht，1992；Hu，2002)Noticingtheproblemsofinsufficientreliabilityandvalidity,hetrieshisbesttoworkoutrelativelycapablewaystomeasurethosetwotypesofknowledgeseparately．Methodscanbeseenclearlyonthetablebelow．Table3—2：AnalysisoftheFiveTestsMeasuringExplicitandImplicitKnowledgeTestTypeofmeasureTypeofknowledgemeasuredElicitedOralImitationConstrained，constructedImplicitknowledgeTestResponse—productionOralNarrativeTestFreeproductionImplicitknowledgeTimedGrammaticalityMetalinguisticjudgment，Implicitknowledge23 万方数据华北l乜力人学顺J1学位论文UntimedGrarnmaticalityMetalinguisticjudgrnentExplicitknowledgeJudgmentTestMetalinguisticSelectedresponsesExplicitknowledgeKnowledgeTest幸TakenfromEllis，2009However,atthesametime，hepointsoutthatthetwokeyfactorsusedtospecifyeithertypeofknowledgeseparatelyarethepresenceorabsenceoflimitationontimeandthetypeoftasks．Itisacknowledgedthatcommunicativetasksundertimepressurethatarerequiredtoproducemeaning—focusedoralutterancesaremoreinclinedtourgetheextractionofimplicitknowledge．Inretum，writingtasksthatallocatesufficienttimefortask—takerstosupervisetheirlanguageproductionarelikelytoelicitlearners’explicitknowledge．Onethingthatlinguistsshouldbearinmindisthatitisimpossibletodesignteststhatcallelicitpureexplicitknowledge，asnomatterwhattheinstrumentis，learnerswilloften，inonewayoranother,turntotheirimplicitknowledgeforhelpandthengiveresponseaccordingly．(Ellis，2004)Fromthefore-goinganalysisabove，wecanfigureout，intermsofhowtooperationalizethetwotypesofknowledge，researchorshaven’treachagreementyet．Nomatterwhatmethodsweadopt，weshouldbearinmind，testsmustbedesignedseparatelytomeasureeachtimeofknowledge．ThoughEllis’S(2004)theoryhassomeflaw,hestillsprovidesUSwithadirectiononhowtomeasurethesetwotypesofknowledgeseparately．Uptonow,itistestableandpractical．Inthispaper,consideringtheactualcondition，theauthorpreferstotakesomeoftheapproachesEllisused，notcoveringallofthemandalsohavingsomeslightrevisions．Moredetailedinformationwillbeintroducedinsubsequentsections3．3ConstructsofSecondLanguageProficiencyOneofthemaintopicsofthispaperistotestifytowhatextentstudents’explicitandimplicitknowledgeiscorrelatedtotheirsecondlanguageproficiency．Inthe，followingpart，theauthorisgoingtobrieflyconsiderthemeasurementsand24 万方数据华北电力人学硕士学位论文operationalizationappliedinpreviousSLAresearchesandthenspontaneouslydrawsaconclusionabouttheconstructsofsecondlanguageproficiencyinthisthesisWaysdifferinassessingofsecondlanguageproficiency．Incurrentmodelsbuiltonlanguagecircle，mostofthemarebasedonlanguagecompetenceorcommunicativecompetence．(Bachman，1990)Butnotthesame，inSLAresearches，secondlanguageproficiencywasmeasuredinvariousaspects，suchaslearners’ownershipofwordsandgrammar(Roehr,2008b)，theabilitytoappropriatelyuselanguageunderrealcontext，theircomprehensivecompetencyinlistening，speaking，readingand,writingInreviewofthoseapproachespublished，HananiaandShikhani(1986)classifiedproficiencymeasurementsintofourtypes．Thatisin—housemeasures，impressionisticjudgment，standardizedtestscoresandinstitutionalstatus．Specificallyspeaking，in—housemeasuresarethosemethodsdevelopedforasalientpurposeandusedforsolestudy,Impressionisticjudgmentreferstoassessmentmadebyresearchersorself-evaluationbylearners．Standardizedtestisdesignedandadministratedofficiallyandscoredinstrictprinciples．Forexample，GRE，TOFEL，whicharewell-knownexamsa11overtheworld．HananiaandShikhani(1986)stronglybelievedthat：Clozeisconsideredanintegrativeratherthanad括crete-pointtestbecauseitdrawsatonceontheoverallgrammatical,semantic,andrhetoricalknowledgeofthelanguage．Toreconstructthetextualmessage,studentshavetounderstandkeyideasandperceiveinterrelationshipswithinastretchofcontinuousdiscourse，andtheyhavetoproducerathe,"thansimplyrecognize，anappropriateworddoreachblank．The&cusofthetaskinvoh,edismorecommunicativethanformalinnature,anditisthereforeconsideredtoreflectapersonjabilitytofiznctioninthelanguage，p．99)Tosumup，intermsofhowtomeasureorevaluatesecondlanguageproficiency,，agreatmanyvariablesshouldbetakenintoconsideration．Noneoftheconstructsis25 万方数据华北电力大学硕二I：学位论文completelyperfect．Eachhasbothlimitationsandadvantages．GreatattentionshouldbepaidtowhilemeasuresecondlanguageproficiencyConsideringtheparticularityofhighschool，conveniencetojudgeandscore，inthispaper,itadoptedacademicproficiencytestastooltoscalethelevelwhichholdattheendoftheschoolyear．Inhighschool，thereisBOspecificclassasinuniversityTheyhavenoseparateclassesnamedListening，Speaking,andIntensiveReading．Onlyoneitemexists，calledEnglishClass．ItcoversalltheroleswhichareneededaboutEnglish．So，takingmoneyandtimeintoconsideration，inGradeone，twoandthree，theauthortakesthefinaltestasmeasurement．Fourpartsareincorporated⋯vocabulary&Grammar,ReadingComprehensionandWriting．Thedetailedinformationisjustasthebelowtablelists．Exceptthosethreeparts，l"dalsoliketoaddClozeintothisProficiencytesttoteststudents’comprehensiveproficiency．Test，developedbyFotos(1991)，isusedHisdozeisoriginatedfromanintermediateEnglishbook．Thecontentofthetestiscomposedofthreeparagraphs．Therearea11together400wordswith20blanksinside．Theseblanksaresystematicallyspacedblanks，actingasgrammaticalfunctionsTable3—3：TestforEnglishGeneralProficiencyGradeAcademicProficiencyTestVocabulary&OneClozeReadingComprehensionWritingGrammarVocabulary&TwoClozeReadingComprehensionWritingGrammarVocabulary&1]}1reeClozeReadingComprehensionWritingGrammar26 万方数据华北电力人学硕士学位论文Chapter4MethodologyThispartoutlinesthedesignofthepaper,whichincludesthespecificdescriptionsofthesubjects，instrumentsaswellastheproceduresofdatacollectionanddataanalysis4．1ResearchDesignInthissection，thepapergivesabriefintroductiontotheresearchdesign，whichiscomposedoftheresearchquestions，thesubjectsandtheinstrumentswemainlyemployedwiththeaimtoachievethethesis’Sgoal4．1．1StudyQuestionsThispapermainlyfocusesonthecorrelationbetweengrammaticalcompetenceandsecondlanguageproficiencyofhighmiddleschoolstudents．Basedontheresultsshownamongthreegrades，theauthortriestofindoutthedifferencesandsimilarities，andfurtherputsforwardssomesuggestionstoEnglishteachingandlearninginmiddlesch001．Inordertoachievethegoal，theauthormustfirstsolvetwokeyproblems，oneisthegrammaticalcompetenceofhighmiddleschoolstudents；theotheroneisthesecondlanguageproficiency．Therefore，thequestionsdesignedinthispaperinvolvetwoaspects．Detailsdisplayasbellow(1)Whatisthehi。ghmiddleschoolstudents’grammaticalcompetenceinthreegradesfromtheaspectsofexplicitknowledgeandimplicitknowledge?(2)WhatisthehJl曲middleschoolstudents’secondlanguageproficiencyinthreegradesregardingtheircourseachievementtest?(3)Towhatextentisimplicitgrammaticalcompetencetotheirsecondlanguageproficiency?(4)Towhatextentisexplicitgrammaticalcompetencetotheirsecondlanguageproficiency?， 万方数据华北【U力大学硕：i：学位论文4．1．2SubjectsThequestionnairesaredeliveredinthreeschoolsacrossGongqingCity-一一GongqingNo．1SeniorMiddleSchool，ChangjiuPrivateandJiangyiTown．Thenatureofthesethreeschoolsarenotthesame，norwiththeirteachingstandardsAmongthem，GongqingNo．1MiddleSchoolandJiangyitownMiddleSchoolarepublicsecondarysch001．Buttheyarelocatedindifferentregions．GongqingNo．1representstheurban，whileJiangyiTownisonbehalfoftherural．ChangjiuPrivatestandsforanotherpole．Altogether,atotalof600pupilsjoinedtheexperimentwitheachschool200studentsonaverageandtheyareaskedtocompletethe2testsandaquestionnaire．Theparticipantsarerandomlyselectedcoveringthreegrades，with75inGradeone，65inGradetwoand60inGradethree．AllofthemhaveaccesstoEnglishwhentheyareinGradethreeinPrimarySch001．TheyhavebeenstudyingEnglishfor7-10years．fi．e．4yearsinPrimarySchool，3yearsinJuniorMiddleSchooland1-3yearsinSeniorMiddleSch001)．BasicgrammaticalknowledgewasexposedtothemandtheyallhaveabasicunderstandingofEnglishgrammar．Thetestsweredeliveredtothemattheendofthesemesterlastsummer,whenthefinaleX锄COmes4．1．3ResearchInstrumentsThispaperadoptedthemethodsanddevicesderivedfromasetoftestsEllis(2005)employedbefore．Fromtheliteraturereview,wecansee，inhisresearch，hetestsimplicitknowledgethreetimesbythreedifferentmethods。-_-__。·TheTimedGJT,theOralNarrativeTestandTheOralImitmionTest，whileexplicitknowledgeismeasuredtwicewiththeMetalinguisticKnowledgeTestandtheUntimedTest．Inthisresearch，onlytwotestsareused—-．---TheTimedGJTandtheMetalinguisticKnowledgeTest．Thereasonwhyhediscardsthosetwoothermethodsreliesonthedifficultyintakingthemintoeffectiveinreality．Freelanguageoutputishardtocommandandrate．Thus，it，willbringaboutmanyproblemstotheexperiment．Inaddition，thereisnoguarantee28 万方数据华北电力人学硕I：学位论文thatthetargetstructurescanbegeneratedideally．TheauthorcannotperfectlypredictwhichstructuretheparticipantswillelicitBearinmindthatEllis’Stestswercdesignedtotestforeignstudents，SOwhentransferredtoChinacontext，minorrevisionisurgentlyneeded．Thenextpartdrawstheexhaustivepictureofthecontentsandscorestandardofthetests"restcontent一17GrammaticaIStructureInthispaper,17grammaticalstructureswerechosentoanalyzestudents’grammaticalcompetence．Accordingtothepreviousresearch，thesestructureswerenotselectedrandomly,butundertheguidanceofasetofprinciplesAtthefirstplace，thesetargetlanguagestructuresarealllearntinmiddleschoolThatistosay,thesestructuresareallcoveredinthetextbooks．TheyarecompulsoryitemsinMiddleSchoolstudents’teachingsyllabusSecondly,theseselectedstructuresappeartobeproblematictolearnersduringtheirprocessofacquisition．Andtheyareshowntolearnersindifferentperiods，standingforearlyandlateacquisitionsequence．(Burr&Kiparsky,1972)Thirdly,thesestructurescoveralargerangeofEnglishlevelaccordingtowhentheyareintroducedtolearners．Advancedlevels，upperintermediateaswellaslowerintermediateareallincorporatedLastbutnotleast，thesestructuresincorporatebothsyntacticandmorphologicaliternsinessence．Undertheguidanceofthesecriteria，theauthorconsultsalotofgrammarbooksandtextbooksnOWusinginMiddleSch001．BasedonthedepthchatwithsomeseniorEnglishteacherswhohavemuchexperienceinteaching，theauthorfinallymakeshisdeterminationtotakethese17grammaticalfeaturesastestcontents．Thefollowingtablehaslainouttheselectioncriteria．Underthedirectionofthesecriteria，theauthorfinallydesignatedthecxaminationpapersTable4-1：TestingGrammaticalStructuresTypeStructureExamplesoflearnerserrorAcquisitionsequence29 万方数据华北电力大学硕士学位论文VerbLilysaysshewantsbuyingaEarl"ySComplementsnewpencil．RegularpastJuliabuyacarthedaybeforeIntermediateMtenseyesterday．TheywillgoHongkong，aren"tQuestionTagsLateSthey?Yes／N0DoesJimlikesoranges?IntermediateMquestmnsModalVerbIshouldtogohomeBOW．EarlyMUnrealIfhehadknownhisphoneLateSconditionalsnumber，hewillcallhim．ShehasbeeninBeijingsince2SinceandForIntermediateSyearsIndefiniteArticleHehadthegreattimeyesterdayLateMBetweenI980and1990，theErgativeVerbspopulationofBeijingwasLateSdecreased．LilyisstilllivinginhisgrandpaPossessive．SLateMhomePlural—STomatethreeapplejustnow．EarlyMKevingotoschoolbybikeThirdperson—SLateMeveryday．ThecakemyfatherbuysitisRelativeclausesLateSbad．Embedded1wanttoknowwhatcanIdoforLateSquestionyou。DativeFatherexplainedSamtheLateSalternationat／sWer．MyapplesaremoresweeterthanComparativesLateSyours．AdyerbplacementShespeaksverywellJapanese．LateSNote：SstandsforSyntactic；MstandsforMorphologicalOralImitation1沁stThispaperiscomposedofabatteryofaltogether34statements，includingbothungrammaticalandgrammaticalsentencesofthetargetstructures．Andeachoccupies 万方数据华北电力大学坝一i-学位论文half．ThesesentenceswerelookedonastheperfecttoolsbyEllis(2005)．Thesentenceswereshowntostudentsintheformofanaudiorecord．Thentheparticipantswereaskedtodecidewhethertheyagreeornotwithwhattherecordersaid．Thisstepwassettocallonstudent’Sattentiononmeaning．Aftertheyhaveunderstoodthebasicmeaningofeachstatement，theywererequiredtorepeatorallywhattheyheardincorrectEnglishexpressionandtheauthorsoundrecordedwhattheparticipantssaidtimelywithoutonewordomittedBeforeconductingthistest，directionsweredeliveredtosubjectsinadvance．Atfirst，thecontentwasbroadcastedbyailaudiorecordwithEnglish．AndthentheauthorexplainedthemitembyiteminChineseincaseofanymisunderstandsduringthetest．Fromthebeginningtotheend，theauthorwasinpresenceandmanipulatedthewholetestingprocessSOastoguaranteethateverysubjecthasdonethefirststepbeforetheymovedtothenextpart．Whenthistestcametoanend，aninterviewwasfollowed．Theinterviewwasalsorecordedvividly．Eachtesttakerwasdemandedtodepicttheirconceptiononthetest．Thisfollow—upisalsoapartofthetest，whichwasdesignedtoinspecttheirawarenessofthegrammaticalstructuresandfurtherprovideevidenceforUSforfurtherdiscussingtherelationshipbetweenconsciousnessandcompetence．Theanswerswereanalyzedbyrecognizingmandatoryoccasionsinusingthetargetlanguagestructures．Examinee’Sfailtosimulateasentencetotallyortoretelltheentiresentenceinsuchsentenceformationthattheydidnotcreateacompulsorymandatorycontextforthetargetstructurewasconsideredasacircumvent．Ineachimitatingsentence，1minutewasgiventotheparticipantonlyifheprovidedthetargetstructureheheardcorrectly．Otherwise，0hemaygetregardlessofwhetherheisintendedtogivebutsupplyUSwithafalseansweratlastorheisattemptedtoavoidthetargetstructure．Scoreswererewrittenintheformofapercentage．DetailedinstructiontotheOralImitationTestCanbeturnedupinErlam(2006)．MetalinguisticKnowledgeTest31 万方数据华北电力人学硕士学位论文Ellis(2004)pointsoutthatmetalinguisticknowledgeisquiterelevanttothedescriptionoflanguage．Therefore，explicitknowledgelargelyincludesmetalinguisticlanguageandCO-developswithit．Moreover,thelabelofdescriptivelanguageprovidedbymetalinguisticlanguagefacilitatestheunderstandingoftheconceptionoflanguageandtheverbalexpressionofexplicitknowledge．However,theexplicitknowledgedescribedisnotequaltometalinguisticlanguage，whilethemetalinguisticlanguageorthelabeloflanguageistheinstrumenttodepictexplicitknowledgeandnotequaltoexplicitknowledgeitself．Gombert(1992)itemedthemetalinguisticlanguageasreflectiveattitudetowardslanguageasanobjectandtheprocessingofexplicitknowledgeistheproductionofthereflectiveattitude．Inthispaper,metalinguisticknowledgetestwasoriginatedfromEllis．Thereareseventeengrammaticalsentencestructures．Ineachsentence,theauthorhasunderlinedtheerrorpart．TheparticipantswereaskedtoexplainwhattypesofrulesthesentencebreaksandtheywererequiredtowritedownthereasonsinEnglishinthevacantspaceandfurthertocorrecttheminrightform．Iftheparticipantsdon’trecognizetherulethesentencedisobeys，theywereallowedtomissitandthenmoveontothefollowingquestions．Eachsentencehastwoscoreaspects，eithercorrectorincorrect．Theitemwithoutanswersisregardedasincorrectandwouldnotgiveanyscore．Thereareseventeenstructuresandeachstructurehastwosentences，SOthistesthasthirty-fourscoresina11．Fromthetablelistedabove,wecansee，these17grammaticalsentenceswerealsocategorizedintotwotypes-_--··syntacticandmorphological．Sointhispart，thescoreofsyntacticandmorphologicaliscalculatedrespectivelyInsomepreviousstudies，variousotherwaysareprovidedtoexplainrules．Forexample，provideparticipantswithseveralchoicestochoosefromandthendemandedthemtotranslateitintoChinese．Onthisaspect，theauthorfirmlystickstotheideathataskingparticipantstoexplaintherulesinEnglishismoreacceptable．Inthisway,itismoreapproachabletocheckstudents’overallcompetencyofexplicitgrammaticalknowledge．Afterall，expressingideasinEnglishisquitemoreauitablecompared 万方数据华北IU力人学硕士学位论文withconveyinginChineseTestofEnglishProficiencyInordertodiscussthecorrelationbetweenExplicit&implicitgrammaticalknowledgeandtheirEnglishproficiencyofMiddleSchoolstudents，theauthorattemptstousepartofcourseachievementsastooltoevaluatestudents’proficiencyCourseachievementistocheckstudents’Englishproficiencyperiodically．Generallyspeaking，inmiddleschool，courseachievementhasstructuredmodule．Listening，Grammar&Vocabulary,Cloze，ReadingandWriting．Inthispaper,Referringtocourseachievementtemplate，giventhelimitedfundingandcapacity,theauthorjustinvolvesfouraspectswhichCanhighlyservestothethesis-····-GrammarandVocabulary,Cloze，ReadingandWriting．Thatistosay,GradeOne，mainlyfocusesonGrammarandVocabulary,Cloze，ReadingandWriting．GradeTwoandThreearealsothesefourpartsappearingintheircourseachievementtest．StudentQuestionnaireAquestionnairewasusedasasupplementtothetest，whichattemptedtoinquireintothelearners’learningstrategyandhabits，Itwilldohelptothefollowingdiscussionpartandsupporttheanalysisofstudents’explicitandimplicitknowledgeaswellastheirEnglishproficiencyThequestionnairewascompositeoftwopartsinthestructure．BeforetickingtheanswersaboutPartII，subjectsareexpcctedto611intheirpersonalinformationfirst，whicharequestionslikeGrade，School，Gender．Thesebasicpersonalinformationarealltothepointtothepaperanalysis．Part11contains9questions．whicharemainlydesignedtoinvestigatelearners’attitude，strategyandhabitsofEnglishlearning．These9questionsinnaturebelongtOself-report．TheyareinheritedbutwithslightchangesfromthequestionnaireoriginatedbyagroupofresearchersinBeijingForeignLanguageUniversity．Beloweachquestion，fourchoicesareprovidedforparticipantstochoos，efrom．Thesubjectsarerequiredtoassessthemselvesfirstand 万方数据华北电力大学硕士学位论文thenselectthemostsuitableonethatfitthemcompletelyaccordingtotheirrealsituation．Inordertoalleviatetheanxietyandobtainthemostrealisticdata,subjectsaretoldinadvancethattheirdataisonlyusedforpaperwritingandguaranteenotexposuretoothers．Participantsareencouragedtosupplytheauthenticandhonestdataastheanswershavenofightorwrong,goodorbad．Everytypeofanswersarewelcomeandacceptablebecauseofindividualdifference4．2DataCollectionThesetwotestsandaquestionnairewereshowedinasequencelikethis：TheOralImitationTest，thequestionnaire，theMetalinguisticKnowledgeTest．TheOralImitationTestwasconductedinalisteningandspeakingclassinlanguagelab．Eachstudenthasaseparateseatequippedwithatape，whichcanrecordsoundforresearchuse．Theylistenedtothematerialsonceandwrotedowntheanswerstothestatementontheassignedanswersheet．Andthen，movedontothesecondstep，repeatedthesentencesorallyandrecordedwhattheysaidbythemselvesunderthesupervisorofateachera．ssistant．AquestionnairewassentoutaftertheOralImitationTest．ThemetalinguisticknowledgetestwascarriedoutinanintensiveclassStudentsareaskedtofillouttheblanksunderthedirections．Notimelimitstothistestinprinciple，butbecausethereisonly45minutesforeachclassinMiddleSchool，SOthetestisexpectedtofinishin45minutes．Basedonthepre—survey,45minutesisquiteenoughfordoingthepaper．SoitisnottheproblemaffectingourresultsInthelightofthepreviousdiscussiononliteraturereviewpart，wecangettheinformationthattheOralImitationTestcanbewidelyusedtoinvestigatetheimplicitknowledgeofaleamer,andtheexplicitgrammaticalknowledgeofanEnglishlearnerCanbescaledbytheUntimcdMetalinguisticTest．AlongwithasetofscoresofEnglishproficiencytestachievedbyfourparts，thesethreetypesofSCOreSwereallgoneintotheSPSSsoftware，SPSSsoftwarewasaspecializedtoolinanalyzingstatistics．TheDescriptivefunctionwasemployedtocalculatethedataofthetests．，Cronbach’SAlphawasintroducedtodetectthereliabilityofthosethreemeasures．To 万方数据华北电力大学硕I．学位论文solvethecoreproblemofthepaperissue，non-parametrictestswerecarriedouttohaveacorrelationanalysisofbothimplicit＆explicitgrammaticalknowledgeandEnglishproficiencyrespectively．TosurveythecorrelationbetweenMiddleSchoolstudents’generalexplicitandimplicitgrammaticalknowledgeofthetargetstructuresandtheircomprehensiveEnglishproficiency,andtherelationshipbetweentheirspecificgrammaticalknowledgeofthetargetstructuresandtheirEnglishproficiency,theauthorconductedabatteryofSpearmanCorrelationsanalysis．Above—mentionedanalysismethodswerethemainlyadopted．Besidesthosetools，ahostofadditionaltoolsinSPSSwerealsoimplementedtotestifywhethertheconclusionsummarizedbeforerightornot．Also，inordertoseewhetherthetargetstructuresvariedfromgradesandschools，anumberofRyan’SandKruskal—Walliswereperformed．Indiscussionpart，theresultsofthequestionnairewereaddedSOastobesteadthethesistoexplaintheproblemandprovidecompellingevidenceinstrengtheningtheresults35 万方数据华北电力大学硕士学位论文Chapter5DataAnalysisandDiscussionThischapterintendedtorevealtheupshotofthethesisfromthethreeaspects：firstandforemost，itdisplayedtheoutcomeofthedescriptivestatisticsofthetwotests．Afterwards，itaddressedtheresultswithregardtothecorrelationbetweenexplicitandimplicitknowledgeandgeneralEnglishproficiencybasedontheSpearmanRankCorrelationAnalysisandKruskal—WallisTestanalysis．Lastly,itdepictedtheachievementspertainingtothecorrelationbetweenexplicitandimplicitgrammaticalknowledgeofspecifictargetfeatureandthecomprehensiveEnglishproficiencyderivedfromthethreetestdata．5．1ExplicitandImplicitGrammaticalKnowledge5．1．1TheResultsoftheTestsTable5—1showsthestandarddeviationandmeanintheformofpercentageaccuracyoftheOralImitationTestandtheUntimedMetalinguisticdonebythetotalof600studentsinGongqingcovetingthreeMiddleSchoolswithdifferentlevels．Fromthetable，wecanfigureoutthat，judgingfromthewholeMiddleSchool，thestudentshadamuchhigherscoreontheUntimedMetalinguisticTestwitha0．617accuracypercentage；however,theyscoredlowerontheOralImitationTest，ofwhichtheaccuracypercentageonlyreachedO．514andhadamuchlargerstandarddeviationThismeadstheirexplicitgrammaticalknowledgeleveloutweighstheirimplicitknowledgelevel．Butseenfromtheperspectiveofstandarddeviation，wecanclearlyseethattheirexplicitgrammaticalknowledgewasnotbalancedamongstudentsfromdifferentschoolsindifferentgrade．Comparedwithexplicitknowledge，onthewhole，implicitgrammaticalknowledgewasrelativelysteadyforwhichthestandarddeviationofthe／OralImitationTestwas9．568，muchlowerthanthatofUntimedMetalinguistictest36 万方数据华北电力人学硕L学位论文Table5—1：DescriptiveStatisticAllocationStandardDeviationTestNMeanAccuracyPercentageOralImitation60014．6500．5149．568TestUntimedMetalinguistic6009．220O．61716．125Test5．1．2Correlation：TheLearners’ExplicitandImplicitKnowledgeTable5-2displaysthedescriptivestatisticsofmeanandstandarddeviationofthesubjects’resultsamongdifferentschoolsacrossdifferentgradesunderthetwotypesofmeasurements．Fromtheperspectiveofschoollevel，wecanclearlyseethatthemeanscoreofbothimplicitandexplicitknowledgeofthethreetestedschoolsarequitehigh．Allofthemhavereached．0．65orabove．Itisnoticeablethatthemeanscoresofexplicitknowledgeexceedmuchmorethanthatofimplicitknowledge．Thesethreetestedschool，regardlessoftheirlocationinurbanorrural，themeanscoresofexplicitknowledgehasmeasuredupO．75．whichsurpasses0．1tothescoreofimplicitknowledge．Andthechangeratekeepsinabalanceonthewholethoughschooldifferencesexisting．However,thedifferencesexistandstillcannotbeignoredduetothevariousnaturesoftheschoolsandtheirdifferentteachingleveraswellasdifferentteachingidea，methodsandfacilityavailable．Fromthetable，subtledisparityamongschoolsCanbeawareof．ThescoresofGongqingNo．1MiddleSchooloverpassthoseothertwoschoolsnomatteronimplicitknowledgeOFOilexplicitknowledgeIntermsofthegradelevel，ineachschoolacrossthreegrades，itisacknowledgedthatexplicitgrammaticalknowledgescoreshigherthanthatofimplicitknowledgewhethertheyareinthefirstgrade，inthesecondgradeorinthethirdgrade．Andasgradeincreasing，theirscoresonthesebothaspectsareincreasedpaceinpace．Let’StakeGongqingNo．1MiddleSchoolasallexample．Thetotalscoreofimplicit37 万方数据华北电力入学硕十学位论文knowledgereachesO．7001，andthemeanscoreforeverygradeismorethan0．68．Thetotalforexplicitknowledgeismuchhigher,whicharrivesatalevelofO．7785andforeachgrade，thescoregoestoabove0．75．Thetotalscoreofimplicitknowledgestandarddeviationis0．7455，whichisalittlebitlowerthanthatofexplicitknowledgeThisprovidesproofsonEllis’Shypothesisthatimplicitknowledgetestwillproducesystematicresponsescomparedwithexplicitknowledgetest．Table5—2：DescriptiveStatisticsofMeanandStandardDeviationacrossSchoolsandGfadesImplicitKnowledgeExplicitKnowedgeschoolGradeNMeanSDNMeanSD175．6824．0738475．7513．08354265．6923．0589565．7836．07124GongqingNo．1360．7283．0779560_8215．06017Total200．7001．07455200．7785．07543175．6513．0773675．7215．08469265．6625．0602365．7433．07225ChangjiuPrivate360．6875．0825360．7876．06374Total200．6750．07892200，7532．07796175．6458．0758475．7034，08514265．6514．0635765．7321．07365JiangyiTown360．6788．0845860．7548．06658Total200．6535．08015200．7501．07993Indetails，figure5-1to5-4onbelowcharthasanelaborateddescriptiononevolutionmodeofstudents’explicitandimplicitscoresamongdifferentschoolsandwithindifferentgradelevelsFromfigurefive-one，wecangettheideathatboththeimplicitknowledgescoreandtheexplicitknowledgescoreshowadownwardtrendfromGongqingNo．1MiddleSchooltoJiangyiTown．Buttheimplicitknowledgescoredecreasefarmorethanexplicitknowledgescore．Specificallyspeaking,forthesethreeschools，there’Sgreatdistanceoftheirimplicitknowledgebetweeneachtwoschoolssincethe，decreaserateishigher．However,theirexplicitknowledgegapdoesn’tappeartobeSO 万方数据华北电力大学硕士学位论文large，especiallyforthegapbetweenChangjiuPrivateandJiangyiTown．Whoshouldberesponsibleforthisdifference，theteachers，thestudentsthemselvesorthepublic?Allofthesefactorsshouldbetakenintoconsiderationandtheyallcontributetothisresult．^^V．葫0．78。⋯“?≮，0，76’—睡—一mm*———。。瓣}0．74}0．720。7⋯、、％‰}0，530蓐5、一j．＼。i画；⋯一一0，64；0蔫2O、巷磊嚣爨霎鼋1n墓No童CbanNiuPritateJiang；"i翔≮￥秘—◇一im0“e琏Mean小数酾ei{MeanFigure5—1：MeanScoreAllocationforSchoolFigure5-2to5-4presentsthedifferencesamongeachgradeineachsch001．Ascanbeseen，foreachschool，themeanpointofexplicitknowledgeisfarbeyondtheirmeanscoreofimplicitknowledge，Andasgradeincreasing,theirperformanceonbothtwosidesimproved．Theonlydifferenceresidesintheirincreaserate．Allthesephenomenaindicatethatthestudentshaveagoodgraspofgrammaritems，whichisbetterthanthecommandofimplicitknowledge．Theyjustmemorizedtherulesthemselvesbuthaven’tconvertedthemintotheirownknowledgestoredintheirsub．consciousness．Whentheyarerequiredtousethesepatternsinutterance，theyalwaysfirstsearchontheirmemoryaboutstructuresandthenselectoneofthemostsuitableone．whichwastestimeandinmostoccasionresultsinfailure．A1lofthesestatisticscanbeusedtoaccountfortheepidemicalphenomenainourcurrentsociety．StudentshavelearuedEnglishforseveralyears．Theyexcelintakingexamsandcarldoquitewell．Butwhentheyareaskedtodeliverastoryorcommunicatewithpeopleinreal39 万方数据华北电力大学硕士学位论文context，theproblemsappearhereandthere．Theyshowfairlylimitedabilitiesintalkingwithcorrectform．Figure5-2：MeanScoreAllocationforGongqingNO．1acrossGradesFigure5—3：MeanScoreAllocationforChangjiuPrivateacrossGrades 万方数据华北lU力大学坝十学位论文⋯⋯“⋯÷⋯一⋯7—盼}mpIicitMean{—戮一ExphckFigure5—4：MeanScoreAllocationforJiangyiTownTable5—3comestheresultsofKruskal—WallisTest，whichmainlydealswithliOn—parametricanalysisofthedata．Duetothemuchsimilarityonthewholeamongschoolsoftheperformanceonexplicitknowledgetestandimplicitknowledgetestandwithlimitedtimeandenergytocompletethetask，here，theauthorjusttakesGongqingNo．1MiddleSchooltohaveaKruskal—WallisTest．Fromthetable，wecanknowaboutthatthedifferencesignificanceofimplicitknowledgeisat0．02levels，andthedifferencesignificanceofexplicitknowledgeisaround0．06levels．Whatthesestatisticsdeclaresisthatthereexistquitealotofsignificantvariancesofthestudents’resultsamonggradesonbothexplicitknowledgeandimplicitknowledge．Table5—3：Kruskal—WallisTestforGongqingNo．1MiddleSchoolExplicitKnowledgeImplicitKnowledgeChi—Square8．81511．415df3Asynp．Sig．0．06O．02Withtheaimtohaveafurtherinvestigationonlocatingthedifferencesamongthesethreegrades，theauthorthenputthedatacollectedintoSPSSforRyanTest，41 万方数据华北电力人学硕上学位论文Table5—4andtable5-5giveanoveralldescriptionoftheoutcome．Table5-4displaysthestatisticsofimplicitknowledgeamongthreegrades．Aswecanseefromthetable，inColumnZ，thescoreofthatshowsgreatsignificanceamongthreegrades．especiallyGradeone—GradethreeandGradetwo．GradethreeIntermsofthedifferencelocationaboutexplicitknowledge，wecangettheresultfromTable5-5．InTable5-5，theZscoresvariesfrom．1．6013to．3．2190betweenthecomparisonsonGradeOne-GradeTwo，GradeOne-GradeThree．andGradeTwo·GradeThree．Thesignificancesofthosetwocomparisons，i．e．Grade2-Grade3，Grade1一Grade3havereached2．5above．Anditisworthytobenotedthatthefigureswegetareallwithnegativesignbeforenumberswhetheronimplicitknowledgeresultsoronexplicitresults．Thismeansthestudents’grammaticalknowledgeimproveswiththegrowthoftheirgradeCombinedwithTable5-4andTable5-5，wecanknowthatthestudentsinthreegradescanbeclassifiedintotwodifferentgroupsaccordingtotheircomparisonresults．GradeOneandGradeTwocanputtogetherandtheyareinlowerlevel．GradeThreeclassifiesintoanothergroup，thehigherlevelgroupastheyareequippedwithhighperformanceonbothexplicitknowledgeandimplicitknowledge．ThismayhassomerelationshiptothereinforcementoftheCollegeEntranceExaminationinGradeThree．ThisresultsCanbeappliedtoothertwoschoolswhichhaven’tdonethisRyan’STestbecauseonpreviousanalysis，thesethreeschoolshasnearlythesametrendonthewholeTable5-4：StatisticsofImplicitKnowledgeamongThreeGradesOilRyan’STestGrade(A)Grade(B)MeanDifference(A-B1d．1ZZeri￡2．0．03011—1．5920土2．12l——3，0．06542—3．5514士2．3823．0．0511l一2．7630士2．1242 万方数据华北电力大学硕上学位论文Table5-5：ResultsofExplicitKnowledgeamongThreeGradesonRyan’STestGrade(A)Grade(B)MeanDifference(A—B)d—lZZerit2—0．03351一1．6013土2．1213—0．06182—3．2190土2-3823—0．04461—2．5840士2．125．1．3Comparison：SpecificGrammaticalStructuresForthepurposeofknowingwelloftheexplicitandimplicitknowledgeofeachspecificfeature，theauthortookthemeanscoreofeachiteminOralImitationtestandtheUntimedMetalinguisticTestintoconsideration．ThedatawassummedupinTable5-6listedbelow．Table5-6hasdescribedclearlythemearlscoreofimplicitknowledgeandexplicitknowledgetoeachitemandthedifferencesexistineachitemStepintodeep，wecanseefi-omoverall，themeanscoreforexplicitknowledgeisfarhigherthanthescoreofimplicitknowledge．BecausemostofthefightcolumnDifferencemeanispositive．Thedifferencemaysomewaycontributetothemeasuredtools，WecanstillfigureouttheStandardDeviationofeachitemappearstobevaried，notonlyonexplicitknowledgebutalsoonimplicitknowledge．ThefigureofSDshowscomparativedifferences．Thisindicatesthereismuchvarianceamongeachgrammaticalstructure．Allthestructuresstudentsleamtarenotinabalancedstate．Table5-6：DetailedDescriptionofExplicitandImplicitScoreonEachSpecificGrammaticalItemExplicitImplicitDifferenceMeanSDMearlSDMCanVerbComplementsO．8150．3270．6170．4530．198ComparativesO．6250．4130．5510．4860．074Embeddedquestion0．6170．5020．5380．4890．079ErgativeVerbs0．23l0．4170．4050．476．0．174Dativealternation0．6280．4010．5140．4920．1144B 万方数据华北电力大学硕十学位论文IndefiniteArticle0．6570．4710．5020．4760．155ModalVerb0．7230．4730．47l0．4850．252Plural．SO．7180．4610，4240．4880．294Adverbplacement0．6350．4710．5730．5010，062Possessive．S0．8250．4110．4560．4680．369QuestionTags0．2860．4460．5910．491—0．305Regularpasttense0．7840．4230．6470．4570．137Relativeclauses0．7950．4470．50l0．4710．294SinceandFor0．5230，4540．5010．4670．022Thirdperson-SO．6180．4130．4120．4830．206Unrealconditionals0．5890．4880．7080．476—0．119Yes／Noquestions0．8040．3970．4760．4910．328Oneofthemainissuesthisthesissharedgreatefforttodowithistofindoutwhat’Sthecompositionstateofparticipants’explicitknowledgeandimplicitknowledge，especiallytoeachindividualitem．Re—scanningtheTable6，wecansuccessfullyobtaintheinformationforbothtwoparts．BasedonTable5-6，intermsofexplicitknowledge，whatstructuresstudentsdoquitewellisPossessive—S，Yes／Noquestions，VerbComplements，RegularpasttenseandRelativeclauses．Themeanscoreforthesestructuresreachesnearly0．8．However,also，therearesomestructuresseemedtobealittlebitdifficultforMiddleSchoolstudents，suchasquestiontags(themeanscoreisonly0．286)andtheErgativeVerbs(themeanscoreforitisjust0．231)Asforimplicitknowledge，structureslikeUnrealconditionals，regularpasttenseseemtobeeasytothetesttakers．Themeanscoreforthesestructureshasachievedaround0．65．Nevertheless，therearesomestructuresintested17grammaticalstructuresdifficulttoparticipants，forexample，thirdperson-Sandplural—S．StudentsonlygetlowerscoreforthesesentencespatternsSeenthelastcolumninTable5-6，wecanknowthedifferenceofexplicit 万方数据华北电力大学坝Ij学位论文knowledgeandimplicitknowledgeintuitivelyfromthenumberchange．FromthescoretheygetinexplicitknowledgeandexplicitknowledgeaswellastheDifferenceMeaninthelastcolumn，wecandrawtheresultthatcertainstructureshavenodifferenceindifficultyinexplicitandimplicitknowledge．Thescoreforthesestructuresvariesquitesmall．Theyhavegottenparticularthesamepointforthesetwokindsofknowledge．WecanfindexamplesasSinceandFor，Comparativesetc．Butsomestructureshavearemarkabledisparityonthescore．Structuresseemtobeeasyinexplicitknowledgebuttobedifficultinimplicitknowledge．ThesestructuresarelikePossessive-S，Yes／Noquestions，etc．Itisnoticeablethattherearesolnestructuresthedifferencemeanappearingtobenegative．Itmeansthescoreforexplicitknowledgeislessthantheygotforimplicitknowledge．Fromthetested17grammaticalstructures，thephenomenaonlyworkedonthreestructures．Thisindicates，amongstructures，therearedistinctvariancesindifficultyofthestructures5．2Learner’SL2proficiencyAspreviousmentioned，theauthorconductedthetestinthreeschoolsaltogether9grades．Becauseofquitesimilarityonscoresbetweenschools，inordertosavetimeandenergy,here，theauthortakesoneoftheschoolsasanexampletoshowtheresults．5．2．1TheCourseAchievement11estThereare600studentsinalltakingpartinthetest．200comefromeachschoolIneveryschool，75studentsarefromGradeOne，65studentsarefromGradeTwoand60studentsarefromGradeThree．ThetotalscoreforeachcourseachievcmentiS150Butbecause，ontheonehand，thereissomedifficultyinmeasuringlistening；Ol"ltheotherhand，listeningisoutofthediscussion．SothetotalscorefortheauthorbeingconcernedisVocabularyandGrammar,Cloze，ReadingandWritingfourpafts，whichaltogetheraccounts120．DetailedallocationcanseeTable5—711able5—7：ScoreofFourPartsAllocation45 万方数据华北电力人学坝一f．学位论文Vocabulary&GrammarClozeReadingWritingTotal1530402511013．70％27．30％36．40％22．60％100％5．2．2TheOverallEnglishProficiencyThetestheldonJuly1st，2014．Acrossthreeschools，altogether600studentsjoinedintheexamination．Collectingthedataforthefourparts，vocabularyandgrammar,cloze，readingandwriting，thentheauthorcomestotherawresultsastable9displayed．Thedetaileddatawasputintocomputerforprocessingforcorrelationanalysis．Aboutthiscontent，itwillbeelabomteddiscussedonthefollowingpartsFromtable5-8plusfigure5-5，wecanclearlyknowthestudents’overallEnglishproficiencyisincreasedasgradegrowth．Allthreeschoolsexperiencealmostthesameincreaserate．ThedifferencevariesSOlittlethatwecanignoreinacertaindegree．Theyhavegottenthemeanscoreallabove70，whichisfarbeyondthequalifiedlineforthesefourparts．ThisindicatesallthestudentstestedbytheauthorhavearrivedatthequalifiedEnglishproficiencyFigure5-6tofigure5—8presentthespecificallocationacrossgradeinsideeachsch001．ThemeanscorefromGradeOnetoGradeTwoisinacomparativelowrateincontrastwiththegrowthfromGradeTwotoGradeThree．Thisappliestothethreeschoolstested．Proofscanbefoundinfigure5-6．5·7and5-8Table5-8：GeneralDistributionofMeanScoreofEveryPartMeanMcallMearlMeanMcanVocabularySchoolGradeGlozeReadingWritingTotal&Gralniuarl1022．5321882．5GongqingNO．1292134208431224342090ChangjiuPrivate1872l321677 万方数据华北电力大学硕士学位论文2919．53220．58l31122．53220．5861719．5281670．5JiangyiTown281830187431022．53219．584——^～一www’^一——w“，⋯^一—^⋯，≥『¨一-毒三亭嘻参蚕『_??I『_≥。黔⋯∞懋黧黧竺篓篓i篓⋯。⋯⋯⋯⋯一￡???～。÷。⋯。÷一1“。j”。。。?一⋯～”?⋯一‘”一1笺m⋯w⋯～?V¨¨"||wV””⋯¨|。。⋯、S““自—“鑫0拜嚣《{n￡No+i—泓￡n,ingjiuPrivate一糍一ll§建霉毋TownFigure5-5：OverallDescriptionfromthePerspectiveofSchool|92一；～v—l霉0～8鑫{86一蚤080了SFigure5-6：TotalMeanscoreDistributionacrossThreeGradesinGongqingNO．147 万方数据华北电力大学硕士学位论文MiddleSchoolFigure5—7：TotalMeanScoreDistributionacrossThreeGradesinChangiiuPrivateMiddleSchoolFigure5·8：TotalMeanScoreDistributionacrossThreeGradesinJiangyiTownMiddleSchool 万方数据牛北Ib力人学预jj学位论文5．3CorrelationbetweenExplicitandImplicitKnowledgeandEnglishProficiency5．3．1CorrelationbetweenExplicitandImplicitKnowledgeandCourseAchievementsAspreviousdiscussed，themeanscoreoftheOralImitationTestwasusedtomeasureparticipants’implicitknowledge，andthestatisticsoftheUntimedMetalinguisticTestswasaddedupasawaytoestimatetheexplicitknowledgeofthelearners．Intermsoflearners’Englishproficieney,theauthoradoptedtheachievementtestasthemainmeans．Inthisthesis，itwasdealtwithtwice．Firstly,themeansoffourpartswasregardedasawholeSOastostudythecorrelationbetweengeneralexplicitknowledgeandimplicitknowledgeandoverallEnglishproficiency．Aftercompletingthis，eachpartwasdiscussedrespectively．OnlyinthiswaycantheauthorachievethegoalofmakingathoroughinquiryontheroleeachkindofknowledgeplayedineachsectionTable5-9：ACorrelationAnalysisofExplicitandlmplicitKnowledgeandCourseAchievementinSpearmanRankNCourseAchievementGradelexplicitknowledge75R=0．415(p-=0．024)Implicitknowledge75R=0．376(p=0．018)Grade2explicitknowledge65R=0．408(p=0，017)Implicitknowledge65R=0．398(p=0，014)Grade3explicitknowledge60R=0．401①=0．012)Implicitknowledge60R=0．468(p=0．009)Noted，inthistable，Rrepresentsthecorrelationmodulus．ThelargerRvalueis，thegreaterassociationtheyhave．Pstandsforthepossibilityandsimilarityofthedatareceivedincontrastwiththerealstatistics．IfthevalueofPiSunder0．05．thismeans49 万方数据华北}乜力大学硕士学位论文thedataisvalidandtheresultsareofgreatsignificanceTable5-9bringsittolightthatthecorrelationbetweenexplicitknowledgeandimplicitknowledgeandtheircourseachievementacrossgradesinGongqingNo．1MiddleSch001．Fromthetable，wecallnotethatinGradeOne，thecorrelationmodulusofexplicitknowledgeis0．415，whilethecorrelationmodulusofimplicitknowledgeis0．376．AndthePvaluesofbothtwoitemsareO．024and0．018．whichhaveattainedthesignificancelevelboth．Thisatteststhestatisticsarevalidandmanifeststhatbothimplicitknowledgeandexplicitknowledgecontributeconsiderabletothetest．However,deeplyobserved，itisclearedthatcorrelationmodulusvarieslittleintwotypesofcorrelation，withO．039varies．Therefore，inthissense，genermlyspeaking，inexaminationprocess，thestudentsinGradeOneappealmoretoexplicitknowledgeratherthanimplicitknowledge．InGradeTwoandGradeThree，thecasesarenearlythesame．TheresultsdiscussedaboutinGradeOnecanbeappliedtothembothingeneral．ButwhentheyareinGradeThree，implicitknowledgeoverweighsexplicitknowledge．Theybegintotumtoimplicitknowledgeforhelpwhenencounteringproblems5．3．2CorrelationbetweenExplicitandImplicitKnowledgeandCourseAchievementsofEachPartAsweknown，thetestedthreegradestookthecourseachievementtestatthesametime，onJuly1“，2014．ThoughtheauthorfocustheattentionontheSalTlepartsforthreegrades，whatshouldwebearinmindisthat，theytookthreekindsoftestingpapersduetotheirgradelevel．Specifically,thecontentsforfourpansarenotthesame．Soitisurgentlyneededtodocorrelationanalysistothreegradesoneachpartinapurposetoseethecorrelationbetweenexplicit&implicitknowledgeandcachpartperformance．Table5—10comestheSpearmananalysisforGradeOne．Fromthetable，wecanseethatexplicitknowledgehasastrongrelationshipwithVocabularyandGrammarpart．Thecorrelationmodulusis0．488，whicharrivesatthe0．05significancelevel50 万方数据华北电力大学硕十学位论文ItalsohasasignificantrelationshipwithMeanscore．Noothersignificancelevelhasbeenreachedwithotherparts．Asforimplicitknowledge，itisquitesurprisingtofindoutthatnostrongcorrelationtofourpartsinGradehasbeenfound．Onthewhole，thisshowsingradeone，studentstakesexaminationrelyingmoreontheirexplicitknowledge．Thereasonmayresideintheirlackofsufficientimplicitknowledge．Moreover,theawarenessofexplicitknowledgehasnotbeenformedinreality．Table5—10：SpearmanAnalysisforGradeOneVocabularyandClozeReadingWritingMeanGraminarexplicit．488(+1．336"301．298．415(半)knowledgeImplicit．367．311．254．236．376knowledge4meansthesigmificancelevelwithin0．054木meansthesignificancelevelwithin0．01Table5-11discussesthecorrelationbetweenfourpartsscoreandexplicitandimplicitknowledge．Fromthetable，wecailgettheideathatexplicitknowledgehasaquitestrongsignificancewithVocabularyandGrammar,Writing，withthefigureO．501and0．512respectively．TheyareallwithinthesignificancelevelofO．01．Asformeanscoreforimplicitknowledge，itmarkshighcorrelationsignificancewiththreeproficiencyperformance，VocabularyandGrammar,ReadingandWriting．AllofthembelongtotheO．05significancelevelSeenfromthetable，itisinterestingtorealizethatthereisnostrongcorrelationbetweenbothexplicitandimplicitknowledgeandCloze．ButbothexplicitandimplicitknowledgeshowspowerrelationtoVocabularyandGrammarandwriting．IntermsofVocabulary＆GrammarandWriting．explicitknowledgeexperiencesmorepowerfulthanimplicitknowledge．Table5—1I：SpearmanAnalysisforGradeTwoVocabulary&ClozeReadingWritingMeanGrammarexplicitknowledge50l(料)．379．412．512(++、．408(+15l 万方数据华北电力大学硕士学位论文韦meansthesignificance1evelwithinO．05料meansthesignificancelevelwithin0．0lTable5-12liststhecorrelationbetweenfourpartsscoreandexplicitandimplicitknowledgeofGradeThree．AsweCallsee，theexplicitknowledgestillfunctionsalotinVocabularyandGrammarpart．NotlikeinGradeTwo，thecorrelationmodulusisnotashigherasinGradeTwo．InGradeThree，explicitknowledgeplaysagreatroleinwriting(0．512)asthesignificancerategetstheO．01level．Withregardtoimplicitknowledge，itcontributesmoretoCloze(O．447)andReading(0．526)．Especiallytoreading，thesignificanceratereachesthe0．叭level．Ithasquitestrongsignificancecorrelationtothemeanofthecourseachievementasexplicitknowledgedoes．TheyarebothwithintheO．01significancelevel．Table5·12：SpearmanAnalysisforGradeThreeVocabulary&ClozeReadingWritingMeanGrammarexplicit0．464(+10．3470．316O．512(++、0．401(料)knowledgeImplicitO．3150．447(+)O．526(4幸、0．3580．468(++、knowledge4meansthesignificance1evelwithin0．0544meansthesignificancelevelwithinO．01WhatwehavementionedaboveistheSpearmananalysisofthecorrelationbetweenexplicit&implicitknowledgeandeachpartofthecourseachievementacrossgradesseparately．Now,let’Shaveasummaryandputthemtogetherandmakeapanoramaofthe“map”．ThenwhatwegainedistheTable5—13．ThistablegivesUSanoveralldescriptionoftheSpearmananalysisofthecorrelationbetweenexplicit&implicitknowledgeandeachpartofthetestaswellasthemeanscoreofthewhole．FromtheTable5—13．wecanknowexplicitknowledgehasastrongcorrelationwithVocabularyandGrammarandWriting，andthisphenomenonismoreobviousinGradeTwo．ImplicitknowledgecontributestoReadingstronglyinGradeThreewhilenosignificantrelationtoReadinginGradeTwo．Inanoverallpointofview,wecanfigureoutthattheircorrelationsignificanceturnstobestrongerandstrongerasgrade52 万方数据华北电力人学硕士学位论文growth．Asgradeincreasing，implicitknowledgefunctionsmoreinexams．MoreandmorestudentsformthehabitatofconsultingtheirimplicitknowledgeforhelpTable5-13：AnOverallPictureoftheSpearmanCorrelationAnalysisbetweenExplicitandImplicitKnowledgeandtheFourPartsacrossThreeGradesVocabularyClozeReadingWritingMeanandGrom2,nlarexplicit．488(+1．336．301．298．415(术)Gradeknowledge1Implicit．367-311．254．236．376knowledgeexplicit，501(料)．379．412．512(。8、．408GradeknowledgeImplicitknowledge．387(4)．311．343(41．336(+1．398explicitGradeknowledge，464(+1．347．316．512(料)．40l(料)Implicit．315．447(+)．526(++1．358．468(4+、knowledge宰meansthesignificancelevelwithin0．054丰meansthesignificancelevelwithin0．015．4Discussion5．4．1Learner’sGrammaticalCompetenceInapurposetomakeclearofthestudentsgrammaticalcompetenceonexplicitknowledgeandimplicitknowledge，inthisstudy,theauthorconductedtwotestsformeasuringeachitemamong600studentsfromGongqingCity．Asthosetestshavebeenprovedeffectivebyresearchers(Ellis，2005)before，here，twoofthemappliedtostudentsdirectly．Theresultsalsoonceagainconfirmedthereliabilityofthemeasurements．AsanalyzedinChapter5section2，theaccuracypercentageofOralImitationtestsisonly0．514．TheaccuracypercentageofUntimedMetalinguisticTestreaches0．617，whichpassed60％．Thismeansthemajorityofthestudentshavegraspedacertainamountofexplicitknowledge，andtheirexplicitgrammaticalcompetenceisfarbeyondtheirimplicitgrammaticalcompetence．Seenfromthe53 万方数据华北I【l力大学硕一L学位论文perspectiveofstandarddeviation，theresultsofOralImitationtestis9，568，withUntimedMetalinguisticTestsscoring16．125．Thesefiguresmanifeststhatamongthetest—takersinthispaper,explicitknowledgeofthernappearstohavelargerdifferencesincontrasttoimplicitknowledge．Specificallyspeaking，thescoreofexplicitknowledgeshowsgreatervariances．Someofthestudentshadgotahighmarkwhileothersgotquitelowmarks．Scoresamongstudentsisinstarkcontrast．Thedifferentresultsofthesetwotypesofknowledgemayprovideevidencetothefactthatexplicitknowledgecompetenceandimplicitcompetencedoesn’tdevelopinabalance．There’Smuchgapbetweenthemandaconsiderableeffortshouldbedonetointernalizetheexplicitknowledge．Inordertoaccountforthedifferences，thequestionnairewassentout．Thequestionnaireincludestwoparts，withoneofthemdescribingthepersonalinformationandtheotherlistingquestionsrelatedtoEnglishstudyandlearningComparedwiththeobjectivetest，thisisasubjectiveself-evaluationabouttheirSeniorMiddleSchoolstudy．TheresultsareasbelowTable5—14：StatisticsLayoutofQuestionnaireNeverSeldoraSorrletimesOflenAlwaysTotal1．ThinkinguseEnglishtoanswer66753597030600teacher’Squestioninclass2．ActivelytakingpartinEnglishactivity4718823910917600organizedbyteachersorstudents3．Watchingoriginalmoviesorreading992571368028600originalEnglishworks4．Positivelyopening2011636l4035600mouthinClass5．Havegreatinterestsandenthusiasmon1361521996053600English6．Teacherspaying69671031671哆4600greatattentionto 万方数据华北fb力人学硕士学位论文dealingwithGrammar7．BefondofGrammar8895172110135600Class8．Teachersorganizingvariouskindsofactivitiesforstudents1671261539l63600topracticeGrammarStructure9．Grammarteachingdogoodhelpto9213718910379600reading，writingandspeakingAccordingtothetable，clearinformationcanbeachievedthatmostanswersgoestothechoiceof“sometimes”．Butnotallarethesame，therearestillsomeexceptionsForquestion3，257outof600studentschoose“seldom”astheirmainchoice，and99outof600choose“never"’．Inasummaryof356(nearly60％)studentsdon’t1earnEnglishfromoriginalsoundmoviesorworks．Forquestion4，201studentsneveropenmouthinEnglishclass．163studentsseldomopenmouthprovidedtheyareforcedtoopen．Theyjustattendclassandlistentowhatteachersaid．Combinequestion6and8，wecarlknowteachersinclassroomspendquantitiesoftimedealingwithgrammarstructures，butseldomcreatingopportunityforstudentsopeningmouthtopracticeandstrengthentheylearntBasedontheanalysis，oneconsensushasbeenreached．Teachersconcentratedonexplicitknowledgeinstructionbutignoreimplicitknowledgeinternalization．Ontheotherhand，studentsputtheirfocusonclassroomteachingandlearningforEnglishlearning．Benotedthat，classroomteachingcanonlyaddstudents’explicitgrammaticalcompetence．Students’implicitgrammaticalcompetencecannotbeincreasedspontaneously．Students‘attitudeshouldbechangedandmorepracticesareneededoutofclass．Thismayexplainthegapbetweenthesetwotypesofgrammaticalcompetence55 万方数据华北电力大学硕士学位论文5．4．2GrammaticalCompetenceandEnglishProficiencyInthisstudy,learners’proficiencywasmeasuredbythefinalexaminationheldinJuly1“．2014foreverygrade．ThepartstheauthorconcernedaboutareVocabulary&Grammar,Cloze，ReadingandWritingfourparts．Afterdatawascollected，SpearmanCorrelationanalysiswascarriedouttotestthecontributionofimplicitgrammaticalcompetenceandexplicitgrammaticalcompetencetostudents’EnglishproficiencyDatingbacktotheresearchespreviousresearchers(Ellis，2006)havedone，strongcorrelationbetweengrammaticalcompetenceandEnglishoverallperformancewasdiscovered．Amongtheirexperiments．themajorityofthecorrelationcoefficientsarriveatthe0．05level；moreover,someevenreachthe0．01significancelevel．TheseresultssupportthefindingsthatstrongrelationshipexistsbetweenexplicitknowledgeandEnglishproficiency,whichwehavesummarizedinChapter2Throughtheanalysisofthestatistic，theauthorfoundtwotypesofknowledgebothcontributealottotheVocabularyandGrammar,Cloze，ReadingandWriting．Andasgradegrowth，thiskindofcontributionismoreobviously．Indetailedspeaking，implicitknowledgeshowshighcorrelationtoVocabularyandGrammar,however,implicitknowledgedoesn’tshowanystrongcorrelationtothefourparts．InGrade2，explicitandimplicitknowledgebothbegintofunctionalottotheperformanceExplicitknowledgeplaysamuchgreaterroleinwritingandgrammar,whereasimplicitknowledgeaccountsinalesspositioninwritingincontrastwithexplicitknowledge．InGrade3，implicitknowledgespeakshimselflouderthanexplicitknowledgeinreadingInpreviousstudies，researchershavesummarizedthreevariablestoillustratethestrengthofthecorrelation．ThatisL1·L2languagedistance，individualdifference，andsecondlanguagelevel．Inthisexperiment，alltheparticipantsarenatives．TheirmothertongueisChinese?andEnglishistheirsecondlanguage．TheybothlearnedEnglishfromGrade3inElementarySch001．Therefore，LI—L2languagedistancecanberBgardedasthesame．Moreover,accordingtothequestionnaire，weknowthestrategyandhabitforstudentsaredifferent．Sothereasonforthisdifferencemaybe56 万方数据华北电力大学坝”I：学位论文students’secondlanguagelevel，andtheirattitudeandawarenesstowardsEnglishlearning． 万方数据华北电力大学倾}1学位论文Chapter6ConclusionandLimitationThispartmainlydescribesthesummarizationofthefindings，fromwhichimplicationsandsuggestionsareputforward．Andatlast，pointsoutthelimitationsofthestudyandmakerecommendationsforfurtherstudy6．1FindingsoftheStudyToanswerthefirstquestionaboutMiddleSchoolstudents’explicitandimplicitgrammaticalcompetence，twotestswereconductedto600studentsfromGongqiingCityacrossthreegrades．Andafterwards，aquestionnairewascarriedoutatthesametimetosupporttheexplanationoftheresults．Theresultsshowedthatthemeanscoreofstudents’explicitknowledgeisalialebithigherthanthatofimplicitknowledgeParticipantsdonotexcelimplicitknowledgeaswellasexplicitknowledge．Forcertainstructures，implicitgrammaticalcompetencesurpassesexplicitknowledgecompetence．Theyhavebeeninternalizedwell．However,mostofthestructureshavenotbeenintemalizedaccordingtothetest．Asforstudents’Englishproficiency,thefinalexaminationwascaredonJuly2014．Themeanscoreforall9classesacrossthreeschoolshavereachedthequalifiedline，whichhavearrivedat70％．Anditisnotedthat，judgingfromthescoreeachgradehasgotten，asgradegrowth，theirEnglishproficiencyincreased．IntermsofthecorrelationbetweenEnglishgrammaticalcompetenceandEnglishproficiencyacrossgrade，aseriesofSpearmananalysisweredone．Basedontheanalysis，severalfindingshavebeendrawn．ComparedwithGradeThree，studentsinGradeOneandGradeTwobelongtoonegroup．neydon’tbearexplicitandimplicitgrammaticalcompetenceasmuchasGradeThree，whoseareinadvancedlevelStudentsinlowerlevelwouldliketorumexplicitknowledgetbrhelp，whilestudentsinhighproficiencylevel，theyprefertogethelpfromtheirimplicitknowledge．Notonlyexplicitknowledgebutalsoimplicitknowledgefunctionsdramaticallyt9students’Englishproficiency．ExplicitknowledgeplaysagreatroleinVocabulary58 万方数据毕北I乜力人学预。I：学位论：殳andGrammarandwriting,whereasimplicitknowledgeworkspowerfullyinReadingandCloze．Asgradegrowth，thecorrectionappearstobestronger．Specifically,thehigherthestudents’Englishproficiencyis，thestrongercorrelationbetweengrammaticalcompetenceandtheirproficiency6．2ImplicationsTheconclusionofthisstudyhasboththeoreticalandpedagogicalimplicationsforsecondlanguageteachingandsecondlanguagelearning．Explicitknowledgeandimplicitknowledgeworktogethertopromotestudents’Englishproficiency．Thiscanbeusedtoconstructsecondlanguageproficiency．SincestudentsinGradeOneandGradeTwoprefbrtoturntheirexplicitknowledgeforhelp．StudentsinGradeThreeturntoimplicitknowledgewhendoingpapers．SoinClass，indifferentgrade，teachingmethodsandstylesshouldbechangedaccordingtothenecessityofthesubject．Wecannotonlypayattentiontoimprovestudents’explicitgrammaticalknowledge，implicitknowledgealsoneedtobeconsidered．Teacherscouldnotjustdealwithgrammarstructures，however,morechancesandopportunitiesshouldbeprovidedforstudentstoopentheirmouth，whichhelpexplicitknowledgeinternalizesintoimplicitknowledge．AlsoasanEnglishlearner,oneshouldexposehimselffrequentlytotheEnglishoriginalworksSOastospeeduptheinternalization．Somestructuresthatseemtobeeasyinexplicitknowledgemightbedifficultinimplicitknowledge．Whatitwantstoconveyisthatcorrectunderstandingandgraspingofcertainstructurescannotguaranteethecorrectuseinrealcontext．Morepracticesandinputisamust．StudentscouldnotjustrelyOnclassroominstruction．Takingmoreactivitiesandreadingmoreoriginalflavorworksisagoodwayforintemalizingexplicitknowledge6．3LimitationandRecommendationsoftheStudyThefindingsofthisthesisarefarfromsatisfactionobviouslybecauseofthe，limitedabilityoftheauthorandoperationdifficultyinreality，Moredepthstudieson59 万方数据华北电力大学硕十学位论文thisissueareofgreatnecessity．First，thelimitationcomesfromthemeasurementsofexplicit＆implicitknowledgeandstudents’generallanguageproficiency．Inthisstudy,onlytwotestsareappliedandcarriedout．Duringtheoperationalprocess，manyflawsexistfornothavingfullconsideration．Aboutthemeasurementsofstudents’languageproficiency,onlyfourpartsofthecourseachievementsareconcerned．Thesecondlimitationcomesfromthenumberofgrammaticalstructuresandtest—takers．Inthisthesis，only179rammaticalstructuresareselectedfortestingOnlyasmallgroupofstudentsineachgradetakepartinthetest．Whethertheconclusionscanbeappliedtoallofthestudentsremainstobetestified．AlargeAresearchwithalargescaleofparticipantsandagreatnumberofgrammaticalstructuresarefavorable60 万方数据华北【U力大学硕二j二学位论文References[1】1Bachman，L．1990．FundamentalConsiderationsinLanguageTesting[M]．Oxford：OxfordUniversityPress．[2】2Bialystok，E．1979。ExplicitandImplicitJudgmentsofL2Grammaticality[J]．LanguageLearning，29(1)：81—103．[3]Bialystok，E．1990．TileCompetenceofProcessing：ClassifyingTheoriesofSecondLanguageAcquisition[J]．TeachingEnglishtoSpeakersofOtherLangTzagesQuarter，24(4)：635—648．[4】Bialystok，E．1994．RepresentationandWaysofKnowing：ThreeIssuesinSecondLanguageAcquisition[A]．InN．C．Ellis(ed．)．ImplicitandExplicitLearningofLanguages，549-569．SanDiego，CA：AcademicPress．[5]Butt，M．&Kiparsky,G．1972．TheGooficon：ARepairManual．forEnglish[M]．Rowley,MA：NewburyHouse．[6】Canale，M．&Swain，M．1980．TheoreticalBasesofCommunicativeApproachestoSecondLanguageTeachingandTesting[J】．AppliedLinguistics，(1)：1—47．[7】Canale，M．1983．FromCommunicativeCompetencetoLanguagePedagogy[A]．InR．Schmidt(ed．，．LanguageandCommunication，2—27．London：Longman．【8]Celce—Murcia，M．2002．WhyItMakesSensetoTeachGrammarinContextandThroughDiscourse【A】．InE．Hinkel&S．Fotos(eds．)．NewPerspectivesonGrammar—TeachinginSecondLanguageClassrooms，119—134．Mahwah，NewJersey：Drihaum．[9]Chomsky,1965．AspectsoftheTheoryof&ntax[MI．Cambridge，Mass：MassachusettsInstituteofTechnologyPress．[10]DeKeyser,R．1998．BeyondFocusonForm：CognitivePerspectivesODLearningandPracticingSecondLanguageGrammar[A]．InC．Doughty&J．Williams(eds．)．FocusonForminClassroomSecondLanguageAcquisition，422—463．Cambridge：CambridgeUniversityPress．[11]DeKeyser,R．2003．ImplicitandExplicitLearning[A】．InC．Doughty&M．H．Long(eds．)．ttandbookofSecondLanguageLearning,313-348．Maiden，MA：BlackwellPublishingLtd．[12]Oienes，Z．＆Pemer,J．1999．ATheoryofImplicitandExplicitKnowledge[J]．BehavioralandBrainSciences，(22)：735—808．[13]Doughty,C．1991．SecondLanguageInstructionDoesMakeaDifference：EvidencefromallEmpiricalStudyonSLRelativization【J]．StudiesinSecondLanguageAcquisition，(13)：431—469．[14]Doughty,C．2003．InstructedSLA：Constraints，CompensationandEnhancement[J]．InC．Doughty&H．Long(eds．)．TheHandbookofSecondLanguageAcquisition，256—310．Maiden，]VIA"BlackwellPublishingLtd．[15]Ellis，R．1993．TheStructuralSyllabusandSecondLanguageAcquisition[J]．TeachingEnglishtoSpeakersofOtherLanguagesQuarterly，27(1)：91-113．[16]Ellis，R．2001．Introduction：InvestigatingForm-focusedInstruction[J]．LanguageLearningsupplement，5l(1)：1-46，61 万方数据华北电力大学硕士学位论文【17]Ellis，R．2002．DoesForm—focusedInstructionAffecttheAcquisitionofImplicitKnowledge?AReviewoftheResearch[J]．StudiesinSecondLanguageAcquisition，24(2)：223-236．【l8]Ellis，R．2004．TheDefinitionandMeasurementofL2ExplicitKnowledge[J]．LanguageLearning，54(2)：227—275．【19]Ellis，R．2005．AttheInterface：DynamicInteractionsofExplicitandImplicitLanguageKnowledge[J]．StudiesinSecondLanguageAcquisition，27(5)：305．352．[20]Ellis，R．2005．MeasuringImplicitandExplicitKnowledgeofaSecondLanguage：APsychometricStudy[J]．StudiesinSecondLanguageAcquisition，27(2)：141．172．[21]Ellis，R．2006．ModelingLearningDifficultyandSecondLanguageProficiency；TheDifferentialContributionsofImplicitandExplicitKnowledge[J]．AppliedLinguistics，27(3)：431-463．[22]Ellis，R．2008．TheStudyofSecondLanguageAcquisition(2ndedition)[M]．Oxford：OxfordUniversityPress．[23]Ellis，R．2009．ImplicitandExplicitKnowledgeinSecondLanguageLearning,TestingandTeaching【M]．Buffalo：MultilingualMatters．【24]Erlam，R．2006．ElicitedImitationasaMeasureofL2ImplicitKnowledge：AnEmpiricalValidationStudy[J]．AppliedLinguistics，27(3)：464—491．[25]Fotos，S．1991．TheClozeTestasanIntegrativeMeasureofEFLProficiency：ASubstituteforEssaysonCollegeEntranceExaminations[J]．LanguageLearning，41(3)：313-336．[26]Frensch，A．P-&RlingeqD．2003．ImplicitLearning[J]．CurrentDirectionsinPsychologicalScience，(12)：13-18．[27]Fromkin，V．1988．AnIntroductiontoLanguage[M]．NewYork：Holt，RinchartandWinston．【28]Gombert，J．1992．MetalinguisticDevelopment[M]．Chicago：UniversityofChicagoPress．[29]Green，P_S．＆Hecht，K．1992．ImpliotandExpliotGrammar：．AnEmpiricalStudy[J]．AppliedLinguistics，(13)：168-184．[30]Han，Y．andEllis，R．1998．ImplicitKnowledge，ExplicitKnowledgeandGeneralLanguageProficiency[J]．LanguageTeachingResearch，(2)：l一23．【3l】Hanania，E．&Shikhani，M．1986．Interrelationshipsamongthreetestsoflanguageproficiency：standardizedESL，clozeandwriting[J】．TeachingEnglishtoSpeakersofOtherLanguagesQuarterly，(20)：97—109．[32]Hu，G．2002．PsychologicalConstraintsontheUtilityofMetalinguisticKnowledgeinSecondLanguageProduction[J】．StudiesinSecondLanguageAcquisition，24(3)：347—386．【33]Hulstijn，J．andHulstijn，w．1984．GranlmaticalErrorsasaFunctionofProcessingConstraintsandExplicitKnowledge[J]．LanguageLearning，34(1)：23—43．【34]Hymes，H．1971．OnCommunicativeCompetence【MI．Philadelphia：UniversityofPennsylvaniaPress． 万方数据华北}乜力火学颁I：学位论文[35]Krashen，S．1981．SecondLanguageAcquisitionandSecondLanguageLearning[M]．Oxford，England：Pergamon．f36]Krashen，S．1983．PrinciplesandPracticeinSecondLanguageAcquisition[M]．Oxford：Pergamon．[37]Larsen—Freeman，D．1991．TeachingGrammar[A]．InM．Celce-Murcia(ed．)．TeachingEnglishasaSecondLanguage．279—283．Boston：Heinle&Heinle．[38]Macrory,G．andStone，V2000．PupilProgressintheAcquisitionofthePerfectTenseinFrench：TheRelationshipbetweenKnowledgeandUse[J]，LanguageTeachingResearch，(4)：55—82．[39]Paradis，M．1994．NeurolinguisticAspectsofImplicitandExplicitMemory：ImplicationsforBilingualismandSLA[A]．InN．C．Ellis(edj．ImplicitandExplicitLearningofLanguages，393—419．SanDiego，CA：AcademicPress．[40]Roehr，K．2008b．MetalinguisticKnowledgeandLanguageAbilityinUniversity—LevelL2Learners[J]．AppliedLinguistics，29(2)：173—199．[41]Schmidt，R．&Frota，S．1986．DevelopingBasicConversationalAbilityinaSecondLanguage：ACase—StudyofanAdultLearner[A]．InR．Day(ed．)．翰l醚ngtolearn：ConversationinSecondLanguageAcquisition，237—326．Rowley,MA：NewburyHouse．[42]Schrnidt，R．2001．Attention[A】．InP．Robinson(ed．)．CognitionandSecondLanguageInstruction．3·32．Cambridge：CambridgeUniversityPress．[43]Scott，VM．1989．AnEmpiricalStudyofExplicitandImplicitTeachingStrategiesinFrench[J]．TheModernLanguageJournal，7(3)：14—22．[44]Sharwood—Smith，M．1981．Consciousness-raisingandtheSecondLanguageLearner[J]．AppliedLinguistics，(2)：159-169+[45]Stem，M．1983．StudiesinSecondLanguageAcquisition[J】．CambridgeUniversityPress，7(2)：251—253．[46]Widdowson，H．G．1989．AspectsofLanguageTeaching[M]．London：OxfordUniversityPress．[47]Yule，G．2002．ExplainingEnglishGrammar[M]，Shanghai：ShanghaiForeignLanguageEducationPress．【48]Dai，M．2005．SecondLanguageAcquisition“Explicit”and“Implicit”Discussion[J]．ForeignLanguageLiterature，(2)：101．111．(戴曼纯，2005．二语习得的“显性”与“隐性”问题探讨[J】．外国语言文学，(2)：101．1l1．)【49]Du，X．2009．AStudyonGrammarTeachingPatternsofCollegeStudentsofEnglishMajors[J】．ShangdongForeignLanguageTeaching，(5)：42—45．(杜小红，2009．高校英语教育专业语法教学模式研究[J]．山东外语教学，(5)：42．45．)【50]Du，X．2010．AResearchStudyonConstructionoftheGrammaticalCompetency[J]．TianjinForeignLanguageP圮sJ，(6)：62—66．(杜小红，2010．基于研究性学习的隐性语法能力构建探讨[J】．天津外国语学报，(6)：62．66．)【51]Gu，Q．2005．DisputeonImplicitKnowledge，ExplicitKnowledgeandtheirInterface【J]-ForeignLanguageEducation，(6)：45-50．(顾琦～，2005．隐性知识、显性Jm_i,J4及其接LI之争【J】．外语教学，(6)：45—50．)【52]He，Z．&Chen，X．2004．EnglishPragmaticGrammar[M]．Beijing：ForeignLanguageTeachingandResearchingPress．(何自然，陈新仁，2004．英语语用63 万方数据华北Ln力人学顾_Jj学位论文语法[M]．北京：外语教学与研究出版社．)【53]Sun，M．2007a．DeclarativeKnowledgeandEnglishCognition[J]．ForeignLanguageResearch，(1)：134-137．(孙呜，2007a．陈述性知识与英语认知【J]．外语学刊，(1)：134．137．)[54]Sun，M．2007b．ProblemsrethinkingonEnglishCommunicativeTeachingApproach[J]．ForeignLanguageandtheirTeaching，(7)：6-28．(孙鸣，2007b．我国英语交际法教学之若干问题再思考[J]．外语与外语教学，(7)：6．28．)[55]Yang，L．&Yang，Z。2007．CognitiveFunctionsofExplicitKnowledgeonGrammarAcquisition[J】．ForeignLanguageResearch，(5)：129—132．(杨烈祥，阳志清，2007．外显知识在语法习得中的认知作用[J]．外语学刊，(5)：129-132．1[56]Zeng，Y2009．AnEmpiricalStudyontheApplicationofImplicitKnowledgeofEnglishMajorStudents[J】．Fore／gnLanguageTeachingandResearch，(4)：298—302．(曾永红，2009．英语专业学生隐性语法知识应用实证研究[J]．外语教学与研究，(4)：298．302．)【57]Zeng，Y．&Bai，J．2010．ACorrelationStudybetweenEnglishImplicitKnowledgeandtheirEnglishProficiency[J]．JournalofHunanFf，苫fNormalUniversity，(3)：52．55．(曾永红，白解红，2010．英语隐性语法知识与语言水平相关性研究【J]．湖南第-t)0范学院学报，(3)：52．55．)[58]Zhang，D．&Zhao，S．2011．ImplicitandExplicitGrammaticalKnowledgeandSecondLanguageReading[J]．ForeignLanguageTeachingandResearch，(3)：387—399．(张东波，赵守辉，2011．隐性及显性语法知识与第二语言‘阅读[J]．外语教学与研究，(3)：387—399．)[59]Zhang，R．2004．ImplicitCognitionandItsImplicationsforEnglishGrammarTeaching[J]．ForeignLanguageWork(4)：43·47．(张人，2004．内隐认知及其对英语语法教学的启示[J1．外语界，(4)：43—47．)64 万方数据华北电力大学硕士学位论文AppendixIRevisedOralImitationTest问卷编号高中生隐性语法知识状况的调查(由问卷统计者编写)同学你好!我是外围语学院英语语言文学研究生，止在做一凋查研究。我们保证：所获得的数据仅会用丁．牛业论文写作绝对不会川作他州。谢谢你的支持与合作!本试题旨在凋查学生的隐性语法知识状况。采取改良版的OralImitation方式，试卷是基于要调查的17个结构展开。具体的17个结构如。卜：AcquistionStructureExamplesoflearnerserrorTypesequenceVerbLilysaysshewantsbuyingaEarlYSComplementsnewpencil．RegularpastJuliabuyacarthedaybeforeIntermediateMtenseyesterday．TheywillgoHongkong，aren7QuestionTagsLateSthey?Yes／NODoesJimlikesoranges?IntermediateMquestionsModalV色rbIshouldtogohomenow．EarlYMUnrealIfhehadknownhisphoneLateScnoditionalsnumber,hewil，callhim．ShehasbeeninBeijingsince2SinceandForIntermediateSyearsIndefiniteAritcalHehadthegreattimeyesterdayLateMBetween1980and1990，theErgativeVerbspopulationofBeijingwasLateSdecreased．LilyisstilllivinginhisgrandpaPossessive—SLateMhomePlural。STomatethreeapptejustnow．EarlyMKevingotoschoolbybikeThirdperson—SLateMeveryday．ThecakemyfatherbuysitisRelativeclausesLateSbad．65 万方数据华北电力大学硕士学位论文Embedded1wanttoknowwhatcan1doforLateSquestionyou．DativeFatherexplainedSamtheLateSalternationCtnsⅥ，er．MyapplesaremoyesweeterthanComparativesLateSyours．AdverbplacementShespeaksverywellJapanese．LateSNote：SstandsforSyntactic；MstandsforMorphologicalRevised0ralImitationTestDirections：请听录音，接下米会听到17个句子，每个句子结尾会有8秒的停顿时间，在停顿期间，学生需要完成两件事：首先，对听剑的句子做出正误判断，如果觉得句子符合语法表达又表意明确，就在True一栏画√；如果不符合语法规范，请在False一栏画√；如果对所听到的句子不确定，请在NotSure‘栏画√。其次，对听到的句子做出判断后，请重复你所听到的句子，如果听到的句子是正确的，可以原话重复；如果听到的句子是错误的，自己的话复述出表意正确的句子。请参照F面的两个例子，完成听力，并在答案sheet上做出相应的选择。Forexample：1．Mr．Smithhadalwaysspendhissummerholidaysinasmallhotelattheseaside．TrueFalseNotsureRepeatedcontent2．Heputssomehaysbesidesthedog．TmeFalseNotsureRepeatedContentSch001GradeStudentNumberListenContents：1．Theotherday,LilyfoundStephenlayinthefloorwithoutbreath．2．Mr．Jacksonoftensharehouseworkwithhiswifeinthepasttwentyyears3．Idon’tthinkyoucanswimverywell．can’tyou?4．⋯MustIfinishmyhomeworkbeforegoingtoplayfootballoutside?⋯NO．youcan‘t．5．ItmusthavebeentoWuzhen．foryouknowitSOwell． 万方数据华北电力大学硕I：学位论文6．Shouldyouhaveworkedabitharder,youcouldachievemore·7．NorthChinaE1ectricPowerUniversityhaschangedvigorouslyforIgraduatedinthevear2012．8．Achildassheis，shecancookaswellashermother．9．WhenⅣkMiaoenteredtheclassroom，hewelcomedbyawarmapplauseandwasmoved，10．ThisisJim，saJldJack’Sbed．Theyallpaintedyellowforyellowisfavoritecolor．11．Helpyourselvestosomejuicesandchickens．12．ThereareacupandtwobottlesoforangejuiceOilthetable·13．ThisisalltheavailablepaperwhichIhaveatthismoment．14．Idon’tl(110wwhethercanIbelieveyounowsinceyoucheatedmebetore·15．Canyouexplainwhattheteachersaidonclassyesterdayforme?16．ForvesterdaY’SEnglishcontests，CathyWOllthefirstprize．SheistheexcellentplayerOilthestageofalltheparticipants．17．Heavilyhethrewhimselfontothebed．School1．TIrue2．True3．7I’me4．True5．Tnle6．True7．True8．‘rme9．Tme10．TmeZJ，7向e12．True13．Tme14．True15．True16．True17．TrueGradeFalseAnswerSheetStudentNumberNotsureNotsureNotsureNotsureNotsureNotsureNotsureNotsureNotsureNotsureNotsureNotsureNOtsureNotsureNotsureNotsureNotsure 万方数据华北电力人学硕!iJ学位论文AppendixIIWrittenMetalinguisticTest司卷编号(由问卷统计者填写)高中生显性知识状况的调查同学你好!我是外国语学院英语语言文学研究生，正在做一调查研究。我们保证：所获得的数据仅会用丁^毕业论文写作绝对不会用作他州。谢谢你的支持与合作!此试题用米考察学生的显性知识状况；采取元语言测试(WrittenMetalinguisticTask)的方式试题涵盖的34个题目是根据要考察的17个语法结构而定的(这17个语法结构足按照四个原则挑选出来的)，具体考察结构如下表。为了增加问卷的有效性，特针对每一个考察对象，出』，两个题目。AcquisitionStructureExamplesoflearnerserrorTypesequenceVerbLilysaysshewantsbuyingaEarnSComplementsnewpencil．RegularpastJuliabuyacarthedaybeforeIntermediateMtenseyesterday．TheywillgoHongkong，arenttQuestionTagsLateSthey?Yes／NODoesJimlikesoranges?IntermediateMquestionsModalVerbIshouldtogohomenow．EarnMUnrealIfhehadknownhisphoneLateSconditionalsnumber,hewil，callhim．ShehasbeeninBeijingsince2SinceandForIntermediateSyearslndefiniteAritcalHehadthegreattimeyesterdayLateMBetween1980and1990，theErgativeVerbspopulationofBeijingwasLateSdecreased．LilyisstilllivinginhisgrandpaPossessive．SLateMhomePlural—STomatethreeapplejustnow．EarnMKevingotoschoolbybikeThirdperson—SLateMeveryday．，68 万方数据华北电力大学硕士学位沦文ThecakemyfatherbuysitisRelativeclausesLateSbad．Embedded1wanttoknowwhatcanIdoLateSquestionforyou．DativeFatherexplainedSamtheLateSaltemationansl巾Pr．MyapplesaremoresweeterComparativesLateSthanyours．AdverbplacementShespeaksverywellJapanese．LateSNote：SstandsforSyntactic；MstandsforMorphologicalWrittenMetalinguisticTestInstructions：改错问卷由34个题目组成，每个题目有且仅有一个错误，错误涵盖句子的时态、语态、性数等。指出每个句子的错误所在并改止，同时写山这么改止的理由，中英文都可以。请参照下面砥个例f，然后完成后面的34个小题。Forexample，1，Everyoneofusknowthereasonwhvsheisalwayslateforclass．Theungrammaticalpartis．k．．．n．．．．．o．．．—w—Thecorrectionis．k．．．．n．．．．．．o．．．．．w．．．—s—Itisnotgrammaticalbecause￡盘望点!!垒』堡￡!Q￡丛!苎』竺丝壶：!!￡垒i点竺!堡!卫点!呈芝zi』旦点i丝g丝』旦￡62￡：坐盟丝型!￡墨盘Q些!盘垒生＆』』Q丝堡盈垒z堡点i磐丛』堡￡!呈!=垒6亟!=塑：2，HeaskmetohelphimwithhisEnglishyesterday．Theungrammaticalpartisas__2kYourcorrectionisaskedItisnotgrammaticalbecauseyesterdavintheendofthesentenceindicatestheverbactionhappensinthepast,sothesentenceshouldusepasttense．．SchoolStudentNumberJudgmentContents：Grade：Date：1．LisasuggestbuyacakeforJim’SbirthdayTheungrammaticalpartisYourcorrectioniSUngrammaticalbecause2．KategotoDalian7TheungrammaticalforhersummervacationlastyeanpartlS69