published_or_final_version / Chinese / Master / Master of Philosophy

published_or_final_version / abstract / toc / Linguistics / Doctoral / Doctor of Philosophy

The results indicate that such effects could be mainly attributed to activation of multiple semantic representations, in line with the primary goal of meaning access for spoken word comprehension. However, it seems that listeners sometimes do activate orthographic information for fine discrimination when the meaning was uncertain. / The study provides the first systematic set of data for the homophone family size effect, demonstrating its presence in spoken Chinese word recognition. / This study examined the homophone family size effect in spoken Chinese using monosyllabic and disyllabic homophones, and disyllabic words with a homophonic syllable. For each of these three types of words, the family size effect was measured in Exp. 1 to 3 using the auditory lexical decision task. In Exp. 1, participants showed slower response to monosyllabic homophones with a larger family size than those with a smaller family size. In Exp. 2, participants showed an opposite family size effect for disyllabic homophones (e.g. /mi4 feng1/[special characters omitted] and [special characters omitted]), that is, higher accuracy to disyllabic homophones than disyllabic non-homophones. In Exp. 3, the family size effect for disyllabic words with a homophonic syllable was found to be dependent on word frequency, an inhibitory effect was present for low frequency words, regardless of whether the first or the second syllable was the homophonic syllable. Using the same three types of words and the same task, Exp. 4 to 6 examined the neural basis of the family size effect with event-related potentials (ERPs). Results showed that the family size effect mainly resulted from access to the multiple semantic meanings associated with a spoken word. In addition, there was evidence that the multiple word-forms of a homophone were activated in a relatively early state (300--500 ms) during spoken word recognition when the first syllable of a disyllabic word was homophonic. / Wang, Wenna. / Advisers: Alan Chun-nang Wong; John Xuexin Zhang. / Source: Dissertation Abstracts International, Volume: 73-07(E), Section: B. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2011. / Includes bibliographical references (leaves 89-94). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. Ann Arbor, MI : ProQuest Information and Learning Company, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract and appendix also in Chinese.

published_or_final_version / Language Studies / Master / Master of Arts

本研究探究了香港某所國際學校中漢語作為第二語言學習者的中文交際意願。研究對象主要是香港某英基國際學校的170名漢語學習者；主要採用定量的研究方法，通過研究者的課堂觀課和學生的“自我報告”（Self-report）來搜集數據，通過五個量表（背景變量量表、心理變量量表和課堂內外漢語交際意願量表）來實現研究；研究的內容包括學生們課堂內、外漢語交際意願的總體水平以及可能存在的差異，並總結前人研究發現，在新的研究背景下對一些可能存在影響關係的背景變量（年齡、性別、漢語水平、學習漢語的社會支持、母語、性格、對待學習漢語的態度）和心理變量（學習漢語時的焦慮、動機、自我覺察的交際能力）進行數據分析并驗證假設，通過SPSS19.0來付諸實現；最終就如何提高學生的課堂內、外漢語交際意願提出教學建議。 本次研究的主要研究發現有：（1）該國際學校漢語學習者的總體漢語交際意願水平偏低，介於“較低交際意願”和“中度交際意願”之間，而且課內漢語交際意願略高於課外漢語交際意願。（2）總體來看，男生和女生之間不存在漢語交際意願上的顯著差異，但在課內，女生比男生更容易不懂就問、更關注語言知識的細節；男生比女生更願意作為發言代表來公開表達自己觀點；在課外，男生比女生更願意在校園裡與陌生人說漢語，更願意和朋友們或者陌生人上網用漢語交流。（3）預科項目（DP）和中學項目（MYP）的學生群體間不存在漢語交際意願的顯著差異。但在課堂上，DP學生比MYP的學生更願意進行公開性的漢語表達，更願意深入探討相關問題，更願意進行一些總結概括能力較強的發言。（4）DP年級學生的漢語水平和交際意願呈顯著正相關，即漢語水平高越高，課內、課外和總體的漢語交際意願越高,漢語水平越低，課內、課外和總體的漢語交際意願越低。（5）在用漢語進行“主動性發言”上的意願，漢語水平高的群體明顯高於漢語水平低的群體，說明高水平漢語學習者更願意掌握交際的主動權。（6）學習漢語社會支持較多的群體表現出明顯更高的漢語交際意願，呈顯著正相關。（7）以廣東話作為母語的學生比以英語作為母語的學生具有更高的漢語交際意願，這種差異性在課堂外比課堂內體現得更為明顯。（8）性格較外向的學生比性格較內向的學生具有更高的課內漢語交際意願。（9）喜歡學漢語的學生比不喜歡學漢語的學生具有更高的課內、課外和總體漢語交際意願。（10）總體來看，學生學習漢語的動機和自我覺察的漢語交際能力都偏高，且兩者與課內、課外和總體漢語交際意願均呈十分顯著正相關，其中動機與課外方面相關性略大，而自我覺察的交際能力與課內方面相關性略大。學生的總體焦慮水平偏低，與課內、課外和總體交際意願均呈十分顯著負相關，與課內方面相關性略大課外方面。無論是課內還是課外漢語交際意願，“自我覺察的交際能力”對漢語交際意願來說是預測性最強的心理變量。 This research investigated Chinese as second language (CSL) learner’s willingness to communicate (WTC) both inside and outside classroom in one of international schools, Hong Kong. The research objects are 170 Chinese Language B learners from an English School Foundation (ESF) international school. The main methodology adopted here was quantitative method. Classroom observations by researcher and self-reports by students were firstly done to collect background data, and then five scales were employed to do the main investigation, including a WTC Inside–the-classroom Scales (WTCIS), a WTC Outside-the-classroom Scale (WTCOS), a Language Anxiety Scale (LAS), a Motivation Scale (MS) and a Self-perceived Communicative Competence Scale (SCCS). The research objectives were to explore the general situation of CSL learners’ WTC, and the specific characteristics as well as possible differences of their WTC inside and outside the classroom; then to further examine the possible affecting variables from both background and psychological perspectives including gender, age, Chinese language proficiency, mother tongue, social support, personality and attitudes towards learning Chinese, which were regarded as background variables, and language anxiety, motivation and self-perceived communicative competence, which were regarded as psychological variables. After that, data was processed and analyzed by SPSS 19.0 for hypothesis verification. Finally, pedagogical implications on how to improve CSL learners’ WTC both inside and outside classroom were generalized from the study. 　　The major findings showed that: (1) the general WTC of CSL learners from this school were slightly on the low side, between the Lower and Middle level, and the WTC inside the classroom was comparatively higher than outside of classroom. (2) On overall WTC, there was no significant difference between male and female students. But inside the classroom, girls were more likely to ask questions and pay more attention on language details than boys, while boy were more active to be the presenters and voice out in public. Outside the classroom, boys were more willing to communicate with strangers or chat online with friends in Chinese. (3) No significant WTC difference was found between Diploma Program (DP) and Middle Years Program (MYP), but DP CSL learners showed more willingness to make public speech, to be more of an inquirer in problems, and to do more generalization in speaking. (4) Among DP group, significant positive correlation was found between language proficiency and WTC, suggesting that the more proficient CSL learners showed higher WTC, and vice versa. (5) The more proficient group were more willing to speak Chinese voluntarily than the less proficient ones, implying that the more competent language learners are more likely to the take the initiative in communication. (6) CSL learners with more social supports to learn Chinese showed higher WTC, and a significant positive correlation was found here. (7) Students with Cantonese as mother tongue had higher WTC than their English as mothertougue counterparts, and this difference was more obviously outside of classroom. (8) Extroverted students showed significantly higher WTC inside of the classroom than the introverted. (9) Students who liked learning Chinese demonstrated higher WTC than those who didn’t. (10) Students’ motivation to learn Chinese and their SPCC were both on the high side, and both correlated positively on WTC, with motivation correlated more closely on outside WTC while SPCC more on inside WTC. Students showed slight lower anxiety level in speaking Chinese. Anxiety had significant negative correlation with WTC, and it mattered more inside than outside of classroom. No matter for WTC inside or outside, SPCC was the most predictive variable among the three. / published_or_final_version / Education / Master / Master of Education

本博士論文主旨為在漢語作為第二語言（以下簡稱“漢語“）的口語表現評分中連接分數與考生表現。為此，本文設計了三個相互關聯的研究，即：研究一、研究二以及研究三。研究一採用傳統評分法獲取傳統分數，研究二設計自信評分法獲得自信分數，研究三則將傳統評分法與自信評分法進行比較。 / 研究一檢驗了傳統分數與考生表現之間的關係。基於代表漢語口語構念四大維度的七個特徵，研究一分析了66個考生在一次漢語口語測試中的表現（每個考生9分鐘），並採用了相關分析和標準多元回歸作進一步分析。結果表明：第一，七個特徵中的每一個特徵都與傳統分數顯著相關，呈現高等或中等效應量。第二，79%和77%的分數差異可以被兩個回歸分析中相應的特徵所解釋。因此，研究一提供了連接分數與考生表現的實證證據，進而驗證了漢語口語能力評估中的效度問題。 / 然而，研究一採用傳統評分法連接分數與考生表現時，還發現了兩個問題：相鄰等級界限模糊和維度之間的交迭。為解決這兩個問題，研究二提出了一種新的評分方法：自信評分法。然而，自信評分法卻在兩個相鄰等級和三個維度上產生了眾多原始自信分數。這些原始自信分數須轉換成一個準確的自信分數便於分數解釋和使用。為此，研究二採用了隸屬度函數和推理規則，並設計了自信評分演算法。隨後，研究二對自信評分進行了試驗，並對傳統評分法和自信評分法進行了初步比較。分數可靠性和關鍵資訊點相關係數結果表明，與傳統評分法相比，自信評分法具有優越性。 / 研究三採用混合研究法對傳統評分法和自信評分法進行比較，便於我們更為全面的理解這兩種方法的異同。量化分數資料包括五名評分員的傳統分數和自信分數。質化訪談資料則包括這五名評分員對傳統評分法和自信評分法評分過程的看法。量化分數資料分析表明，與傳統評分法相比，自信評分法能在分數和考生表現之間建立更加緊密的連接。質化訪談資料分析發現，自信評分法基於傳統評分法，並從傳統評分法發展而來。更為重要的是，自信評分法不但包含了傳統評分法，而且還承認並利用了評分員的評分自信對口語表現進行評分。 / 本博士論文旨在驗證漢語口語能力評估的效度問題，主要貢獻在於對連接分數與考生表現這一議題上，提供了概念解釋、實證證據以及方法論創新。 / The major objective of the thesis is to link scores to candidate performance in scoring speaking performance for Chinese as a second language (L2 Chinese). To this end, the thesis has been designed comprising three coordinated studies, namely, Study One, Study Two and Study Three. Study One employs Traditional Scoring for obtaining traditional scores, Study Two develops Confidence Scoring for producing confidence scores and Study Three compares Traditional Scoring and Confidence Scoring. / In Study One, the relationship between traditional scores and candidate performance was examined. Seven features--representing four major categories in the L2 Chinese speaking construct--were employed. Speech samples of 66 candidates on an L2 Chinese speaking test (i.e., 9 minutes’ speech length for each) were analyzed in terms of the seven features, with correlations and standard multiple regression being employed. Results indicated that, first, each of the seven features was significantly correlated to the traditional scores, producing large or medium effect sizes; second, 79% and 77% of the variance in the scores could be explained by the features involved in two regression analyses respectively. Study One therefore provides empirical evidence for linking scores to candidate performance to validate the assessment of L2 Chinese speaking proficiency. / Two problems were, however, identified in Study One when linking scores to candidate performance in using Traditional Scoring--indistinction between adjacent levels and overlap between scales. To address these two problems, Study Two therefore proposed a new approach, Confidence Scoring, leading to raw confidence scores between two adjacent levels applied to three rating scales. Since raw confidence scores had to be transformed to an exact confidence score for score interpretation and use, membership functions and rule bases were applied and a Confidence Scoring Algorithm was developed. A pilot study was subsequently conducted in Study Two to try out Confidence Scoring. An initial comparison was also made in the pilot study between Traditional Scoring and Confidence Scoring. Results of scoring dependabilities and correlations with Key Message Points (KMPs) indicated that Confidence Scoring outperformed Traditional Scoring. / In Study Three, a mixed methods study was conducted to provide a more comprehensive picture in comparing Traditional Scoring and Confidence Scoring. Quantitative score data comprised traditional scores and confidence scores from five raters. Qualitative interview data encompassed five raters’ perceptions on the scoring process using Traditional Scoring and Confidence Scoring. The analysis of quantitative score data indicated that, in relation to Traditional Scoring, a closer link between scores and candidate performance was established through Confidence Scoring. The investigation of qualitative interview data found that Confidence Scoring was based on and developed from Traditional Scoring. More importantly, while Confidence Scoring embraced Traditional Scoring, it provided a more flexible way of acknowledging and incorporating raters’ confidence in scoring speaking performance. / The contribution of the thesis therefore rests on conceptual understanding, empirical evidence as well as methodological innovationin linking scores to candidate performance--in the context of validating the assessment of L2 Chinese speaking proficiency. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Jin, Tan. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references (leaves 161-175). / Abstract also in Chinese; appendixes includes Chinese. / ABSTRACT --- p.I / 摘要 --- p.III / ACKNOWLEDGEMENTS --- p.IV / FIGURES --- p.IX / TABLES --- p.X / ABBREVIATIONS --- p.XI / PUBLICATIONS --- p.XII / Journal articles (podcast) --- p.XII / Conference proceedings --- p.XII / Chapter 1 --- INTRODUCTION --- p.1 / Chapter 1.1 --- Context of the research --- p.1 / Chapter 1.2 --- Research questions --- p.4 / Chapter 1.2.1 --- Research Question 1 (Study One) --- p.5 / Chapter 1.2.2 --- Research Question 2 (Study Two) --- p.5 / Chapter 1.2.3 --- Research Question 3 (Study Three) --- p.5 / Chapter 1.3 --- Research design overview --- p.5 / Chapter 1.3.1 --- Study One: Traditional Scoring --- p.6 / Chapter 1.3.2 --- Study Two: Confidence Scoring --- p.6 / Chapter 1.3.3 --- Study Three: Traditional Scoring and Confidence Scoring --- p.8 / Chapter 1.4 --- Potential Contribution --- p.10 / Chapter 2 --- REVIEW OF THE LITERATURE --- p.12 / Chapter 2.1 --- Early Development: Linking scores to expert experience --- p.16 / Chapter 2.1.1 --- Expert experience: The “native speaker“ benchmark --- p.16 / Chapter 2.1.2 --- Practice perspective: (I)ELTS (1986 & 1989) --- p.17 / Chapter 2.2 --- Major Contribution: Linking scores to rater perception --- p.19 / Chapter 2.2.1 --- Teacher/Rater interpretation: “scaling descriptors“ --- p.20 / Chapter 2.2.2 --- Rater judgment: “binary comparisons“ --- p.21 / Chapter 2.2.3 --- Practice perspective: IELTS revision (1998-2001) --- p.23 / Chapter 2.3 --- Work in Progress: Linking scores to candidate performance --- p.26 / Chapter 2.3.1 --- Identifying features from rater perception --- p.28 / Chapter 2.3.2 --- Identifying features from documents/rating scales --- p.30 / Chapter 2.3.3 --- Practice perspective: TOEFL iBT and IELTS (operational) --- p.31 / Chapter 2.4 --- The L2 Chinese context and identifying L2 Chinese features --- p.35 / Chapter 2.4.1 --- Pronunciation --- p.38 / Chapter 2.4.2 --- Fluency --- p.39 / Chapter 2.4.3 --- Vocabulary --- p.40 / Chapter 2.4.4 --- Grammar --- p.40 / Chapter 2.5 --- Traditional Scoring and problems of “indistinction“ and “overlap“ --- p.41 / Chapter 2.6 --- Summary --- p.46 / Chapter 3 --- STUDY ONE: TRADITIONAL SCORING --- p.48 / Chapter 3.1 --- Introduction --- p.48 / Chapter 3.1.1 --- Traditional Scoring --- p.48 / Chapter 3.1.2 --- Research Question 1 --- p.49 / Chapter 3.2 --- Method --- p.50 / Chapter 3.2.1 --- Instrument: An L2 Chinese speaking test --- p.50 / Chapter 3.2.2 --- Participants --- p.53 / Chapter 3.2.3 --- Coding --- p.54 / Chapter 3.2.4 --- Statistical analysis --- p.56 / Chapter 3.3 --- Results --- p.58 / Chapter 3.3.1 --- Correlations --- p.58 / Chapter 3.3.2 --- Standard multiple regression --- p.59 / Chapter 3.4 --- Discussion --- p.61 / Chapter 3.5 --- Summary --- p.74 / Chapter 4 --- STUDY TWO: CONFIDENCE SCORING --- p.76 / Chapter 4.1 --- Introduction --- p.76 / Chapter 4.1.1 --- Confidence Scoring --- p.77 / Chapter 4.1.2 --- Research Question 2 --- p.80 / Chapter 4.2 --- Confidence Scoring design --- p.82 / Chapter 4.2.1 --- Raw confidence scores of adjacent levels --- p.82 / Chapter 4.2.2 --- Raw confidence scores from different scales --- p.88 / Chapter 4.2.3 --- Raw confidence scores to a confidence score --- p.90 / Chapter 4.2.4 --- Score interpretation and use --- p.97 / Chapter 4.3 --- Pilot study --- p.98 / Chapter 4.3.1 --- Candidates and instruments --- p.98 / Chapter 4.3.2 --- Coding system --- p.99 / Chapter 4.3.3 --- Confidence scores and traditional scores --- p.100 / Chapter 4.4 --- Discussion --- p.104 / Chapter 4.5 --- Summary --- p.106 / Chapter 5 --- STUDY THREE: TRADITIONAL SCORING AND CONFIDENCE SCORING --- p.108 / Chapter 5.1 --- Introduction --- p.108 / Chapter 5.1.1 --- Mixed methods: The convergent parallel design --- p.109 / Chapter 5.1.2 --- Research Question 3 --- p.110 / Chapter 5.2 --- Method --- p.111 / Chapter 5.2.1 --- Quantitative score data --- p.111 / Chapter 5.2.2 --- Qualitative interview data --- p.112 / Chapter 5.3 --- Analysis --- p.114 / Chapter 5.3.1 --- Quantitative data analysis --- p.114 / Chapter 5.3.2 --- Qualitative data analysis --- p.121 / Chapter 5.4 --- Results and Findings --- p.123 / Chapter 5.4.1 --- Quantitative results --- p.123 / Chapter 5.4.2 --- Qualitative findings --- p.126 / Chapter 5.5 --- Discussion --- p.141 / Chapter 5.6 --- Summary --- p.145 / Chapter 6 --- GENERAL DISCUSSION AND CONCLUSION --- p.147 / Chapter 6.1 --- Study One: Traditional Scoring --- p.147 / Chapter 6.1.1 --- Constructing rating scales based on candidate performance --- p.148 / Chapter 6.1.2 --- Establishing a potential alignment of L2 speaking tests --- p.148 / Chapter 6.2 --- Study Two: Confidence Scoring --- p.151 / Chapter 6.2.1 --- Applying Confidence Scoring in other educational contexts --- p.151 / Chapter 6.2.2 --- Developing computation package for Confidence Scoring --- p.152 / Chapter 6.3 --- Study Three: Traditional Scoring and Confidence Scoring --- p.153 / Chapter 6.4 --- Limitations --- p.154 / Chapter 6.5 --- Conclusion --- p.155 / Chapter 6.6 --- Future agendas: Where are we heading? --- p.157 / Chapter 6.6.1 --- Investigating more features representing the construct --- p.158 / Chapter 6.6.2 --- Applying Confidence Scoring to different contexts --- p.159 / Chapter 6.6.3 --- Combining automated scoring and raters’ scoring --- p.160 / REFERENCES --- p.161 / APPENDICES --- p.176 / Appendix 1 --- p.176 / Appendix 2 --- p.177 / Appendix 3 --- p.184 / Appendix 4 --- p.188 / Appendix 5 --- p.192 / Appendix 6 --- p.193 / Appendix 7 --- p.195 / Appendix 8 --- p.199 / Appendix 9 --- p.203 / Appendix 10 --- p.207

published_or_final_version / Language Studies / Master / Master of Arts

In addition to the ED-EC framework, this thesis proposes a discriminative lattice rescoring (DLR) algorithm to facilitate the investigation of the extensibility of the framework. The DLR method recasts a discriminative n-gram model as a pseudo-conventional n-gram model and then uses this recast model to perform lattice rescoring. DLR improves the efficiency of discriminative n-gram modeling and facilitates combined processing of discriminative n-gram modeling with other post-processing techniques such as the ED-EC framework. / This thesis proposes an error detection and correction (ED-EC) framework to incorporate advanced linguistic knowledge sources into large vocabulary continuous speech recognition. Previous efforts that apply sophisticated language models (LMs) in speech recognition normally face a serious efficiency problem due to the intense computation required by these models. The ED-EC framework aims to achieve the full benefit of complex linguistic sources while at the same time maximize efficiency. The framework attempts to only apply computationally expensive LMs where needed in input speech. First, the framework detects recognition errors in the output of an efficient state-of-the-art decoding procedure. Then, it corrects the detected errors with the aid of sophisticated LMs by (1) creating alternatives for each detected error and (2) applying advanced models to distinguish among the alternatives. In this thesis, we implement a prototype of the ED-EC framework on the task of Mandarin dictation. This prototype detects recognition errors based on generalized word posterior probabilities, selects alternatives for errors from recognition lattices generated during decoding and adopts an advanced LM that combines mutual information, word trigrams and POS trigrams. The experimental results indicate the practical feasibility of the ED-EC framework, for which the optimal gain of the focused LM is theoretically achievable at low computational cost. On a general-domain test set, a 6.0% relative reduction in character error rate (CER) over the performance of a state-of-the-art baseline recognizer is obtained. In terms of efficiency, while both the detection of errors and the creation of alternatives are efficient, the application of the computationally expensive LM is concentrated on less than 50% of the utterances. We further demonstrate that the potential benefit of using the ED-EC framework in improving the recognition performance is tremendous. If error detection is perfect and alternatives for an error are guaranteed to include the correct one, the relative CER reduction over the baseline performance will increase to 36.0%. We also illustrate that the ED-EC framework is robust on unseen data and can be conveniently extended to other recognition systems. / Zhou, Zhengyu. / Adviser: Helen Mei-Ling Meng. / Source: Dissertation Abstracts International, Volume: 72-11, Section: B, page: . / Thesis (Ph.D.)--Chinese University of Hong Kong, 2009. / Includes bibliographical references (leaves 142-155). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [201-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese.

published_or_final_version / Education / Master / Master of Education

The purpose of this dissertation was to evaluate the Mandarin Chinese course taught as a foreign language for distance learners through a review of distance education and how it applied in foreign language teaching, a learning package based on Unisa's Mandarin Chinese 1, a beginners' course, was fully discussed and demonstrated. An analysis was made of what had been utilised in the learning package and how each element helped distance learners to develop their language skills. Based on the course's learning outcomes, learning material, assessments and learning support, empirical research was done through focus group interviews and questionnaires to test the effectiveness of the Mandarin course. Analysis of the results suggested that: • Students needed to build up a more realistic expectation about learning Chinese. • Students' workloads needed to be re-examined and re-measured. • Certain students lacked learner autonomy. • Mandarin courses should be produced jointly with distance teaching institutions abroad. / Educational Studies / M.Ed. (Didactics)