前言 Preface Acknowledgements Chapter One INTRODUCTION 1.1 Background 1.2 Need for the study 1.3 Outline of the dissertation Chapter Two LITERATURE REVIEW 2.1 Definition of key concepts 2.1.1 Pronunciation 2.1.2 Criterial features 2.2 Second language pronunciation assessment 2.2.1 Competing paradigms 2.2.2 Factors influencing the pronunciation assessment 2.2.3 Automated scoring 2.3 Criterial features in L2 pronunciation assessment 2.3.1 The Mixed-method approach to identify criterial features 2.3.2 Phonological categories derived from theories 2.3.3 Phonological categories derived from rating scales 2.3.4 Phonological features derived from empirical studies 2.3.5 Summary of the phonological features identified in the literature 2.3.6 Relative contribution of the phonological features Chapter Three METHODOLOGY 3.1 Research questions 3.2 Speech corpus 3.2.1 RESCCL 3.2.2 The speech samples for this study 3.2.3 Read English Speech Corpus of Native American Speakers 3.3 Phase- I study 3.3.1 Speech samples in Phase I 3.3.2 Speech coding in Phase I 3.3.3 Data analysis in Phase I 3.4 Phase-II study 3.4.1 Speech coding in Phase II 3.4.2 Data analysis in Phase II 3.5 Summary Chapter Four RESULTS AND DISCUSSION I : The Validation of Criteriai Features 4.1 Segmental features 4.2 Prosodic features 4.2.1 Intonation features 4.2.2 Stress features 4.2.3 Rhythmic features 4.3 Fluency features 4.4 Discussion 4.4.1 Features that significantly contributed to the pronunciation ratings 4.4.2 Features that failed to significantly contribute to the pronunciation ratings Chapter Five RESULTS AND DISCUSSION II : Distinguishing the Holistic Rating Levels 5.1 Segmental features 5.2 Prosodic features 5.2.1 Intonation features
5.2.2 Stress features 5.2.3 Rhythmic feature 5.3 Fluency features 5.3.1 Speed 5.3.2 Breakdown features 5.4 Discussion 5.4.1 Features distinguishing all the levels 5.4.2 Features distinguishing between the lower levels 5.4.3 Features distinguishing between the high and lower levels 5.4.4 The feature failing to distinguish across holistic rating levels 5.5 Summary Chapter Six RESULTS AND DISCUSSION III : Predicting the Analytic Ratings 6.1 Segmental features 6.1.1 Checking the assumptions for regression 6.1.2 The regression model for the segmental ratings 6.2 Prosodic features 6.2.1 Checking the assumptions for regression 6.2.2 The regression model for the prosodic ratings 6.3 Fluenc3, features 6.3.1 Checking the assumptions for regression 6.3.2 The regression model for the fluency ratings 6.4 Discussion 6.4.1 Segmental predictors 6.4.2 Prosodic predictors 6.4.3 Fluency predictors 6.5 Summary Chapter Seven CONCLUSION 7.1 Major findings 7.1.1 The validation of the criterial features 7.1.2 The role of the criterial features in distinguishing the holistic rating levels 7.1.3 The differential contribution of the criterial features to the analytic ratings 7.2 Implications 7.2.1 Implications for the conceptualizations of pronunciation 7.2.2 Implications for pronunciation teaching 7.2.3 Implications for pronunciation testing 7.3 Limitations 7.4 Suggestions for future research References Appendix A: Text of the read dialogues Appendix B: Categorization of the within-sentence pauses Appendix C. Locations of boundary tones examined for tone deviation Appendix D. Syllables possible to be stressed in the read text Appendix E: Possible locations for accented syllables and sampled tonic/weak syllables
[