| 摘要 | 第6-15页 |
| Abstract | 第15页 |
| Acknowledgments | 第16-22页 |
| Figures | 第22-23页 |
| Tables | 第23-26页 |
| Abbreviations | 第26-27页 |
| Chapter 1 Introduction | 第27-39页 |
| 1.1 The interactive nature of the scoring of direct writing assessment | 第27-34页 |
| 1.2 Research questions | 第34页 |
| 1.3 Significance of the study | 第34-36页 |
| 1.4 Definition of the key terms | 第36-37页 |
| 1.5 Overview of the dissertation | 第37-39页 |
| Chapter 2 Literature Review:Research on Raters and Rating Scales | 第39-91页 |
| 2.1 Studies concerning rater factor | 第39-61页 |
| 2.1.1 Effects of rater characteristics on essay rating outcomes and processes | 第40-46页 |
| 2.1.2 Text features raters focus on | 第46-53页 |
| 2.1.3 Raters' mental processes | 第53-61页 |
| 2.2 Studies concerning rating scales | 第61-75页 |
| 2.2.1 Theoretical arguments concerning holistic and analytic scales | 第61-65页 |
| 2.2.2 Effects of interactions between raters and holistic and analytic scales on essay rating outcomes | 第65-68页 |
| 2.2.3 Interactions between raters and holistic and analytic scales and the effect of such interactions on essay rating processes | 第68-75页 |
| 2.2.3.1 Interactions between raters and holistic and analytic scales | 第68-70页 |
| 2.2.3.2 Effects of interactions between raters and holistic and analytic scales on essay rating processes | 第70-75页 |
| 2.3 Approaches to investigating essay rating outcomes and processes | 第75-90页 |
| 2.3.1 Essay rating outcomes:G-theory and MFRM | 第75-84页 |
| 2.3.2 Essay rating processes:think-aloud protocols | 第84-90页 |
| 2.4 Summary | 第90-91页 |
| Chapter 3 Research Design | 第91-134页 |
| 3.1 Overview of the research design | 第91-92页 |
| 3.2 Context of the study | 第92-96页 |
| 3.2.1 General description of CET | 第92-93页 |
| 3.2.2 Writing task in CET | 第93-94页 |
| 3.2.3 Rating scale and rating procedure used in CET 6 | 第94-96页 |
| 3.3 Data collection | 第96-111页 |
| 3.3.1 Materials | 第96-102页 |
| 3.3.1.1 Essays | 第96-98页 |
| 3.3.1.2 Rating scales | 第98-102页 |
| 3.3.2 Participants | 第102-106页 |
| 3.3.3 Data-collection tools | 第106-108页 |
| 3.3.3.1 Think-aloud protocols:training and instructions | 第106-107页 |
| 3.3.3.2 Questionnaires and semi-structured interviews | 第107-108页 |
| 3.3.4 Data-collection procedures | 第108-111页 |
| 3.4 Data analyses | 第111-132页 |
| 3.4.1 Quantitative analysis:essay scores | 第111-115页 |
| 3.4.1.1 G-theory analysis of the scores | 第111-113页 |
| 3.4.1.2 MFRM analysis of the scores | 第113-115页 |
| 3.4.2 Qualitative analysis | 第115-132页 |
| 3.4.2.1 Data transcription | 第115-117页 |
| 3.4.2.2 Segmentation of transcript | 第117-118页 |
| 3.4.2.3 Development of a coding scheme | 第118-125页 |
| 3.4.2.4 Reliability and validity of the coding scheme | 第125-126页 |
| 3.4.2.5 Validity issue concerning think-aloud protocols | 第126-129页 |
| 3.4.2.6 Quantitative and qualitative analysis of the coded data | 第129-132页 |
| 3.5 Summary | 第132-134页 |
| Chapter 4 Effects of Rater-scale Interactions on Essay Rating Outcomes | 第134-185页 |
| 4.1 G-theory analysis of the scores | 第134-141页 |
| 4.1.1 Estimated variance components for the two rating scales (G-study) | 第134-138页 |
| 4.1.2 Score generalizability (D-study) | 第138-141页 |
| 4.2 MFRM analysis of the scores | 第141-180页 |
| 4.2.1 FACETS screening | 第141-144页 |
| 4.2.1.1 Model-data fit | 第141-143页 |
| 4.2.1.2 Psychometric dimensionality of ratings | 第143-144页 |
| 4.2.2 MFRM results | 第144-180页 |
| 4.2.2.1 Examinee ability estimates | 第145-153页 |
| 4.2.2.2 Rater severity and self-consistency | 第153-157页 |
| 4.2.2.3 Scale functioning | 第157-169页 |
| 4.2.2.4 Bias interactions | 第169-180页 |
| 4.3 Summary | 第180-185页 |
| Chapter 5 Effects of Rater-scale Interactions on Essay Rating Processes | 第185-264页 |
| 5.1 Differences in the use of rating strategies across the two scales | 第185-211页 |
| 5.1.1 Differences in the use of general categories of rating strategies across the scales | 第185-188页 |
| 5.1.2 Differences in the use of specific rating strategies across the scales | 第188-209页 |
| 5.1.2.1 Self-monitoring-interpretation | 第188-193页 |
| 5.1.2.2 Interpretation of ideas, logic of argument, or problematic use of vocabulary | 第193-194页 |
| 5.1.2.3 Classifying errors into types | 第194-196页 |
| 5.1.2.4 Self-monitoring-judgment | 第196-203页 |
| 5.1.2.5 Assessing quality (overall and specific) | 第203-205页 |
| 5.1.2.6 Considering local language features | 第205-206页 |
| 5.1.2.7 Editing errors | 第206-208页 |
| 5.1.2.8 Summarizing judgments | 第208-209页 |
| 5.1.3 Summary of the differences in the use of rating strategies across the two scales | 第209-211页 |
| 5.2 Differences in raters' text focus across the two scales | 第211-239页 |
| 5.2.1 Differences in the main categories of text focus across the scales | 第211-213页 |
| 5.2.2 Differences in specific text focuses across the scales | 第213-236页 |
| 5.2.2.1 Content focus | 第213-217页 |
| 5.2.2.2 Coherence focus | 第217-221页 |
| 5.2.2.3 Quality of language (unspecified) focus | 第221-224页 |
| 5.2.2.4 Syntax focus | 第224-227页 |
| 5.2.2.5 Grammar focus | 第227-229页 |
| 5.2.2.6 Vocabulary focus | 第229-234页 |
| 5.2.2.7 Non-scale-related language feature focus | 第234-236页 |
| 5.2.3 Summary of the differences in raters' text focus across the two scales | 第236-239页 |
| 5.3 Difficulties raters encountered when applying the two scales | 第239-256页 |
| 5.3.1 Major difficulties in essay rating in the holistic rating condition | 第239-245页 |
| 5.3.2 Major difficulties in essay rating in the analytic rating condition | 第245-255页 |
| 5.3.3 Summary of the difficulties raters encountered when applying the two scales | 第255-256页 |
| 5.4 Main features of raters' interaction with the two scales | 第256-259页 |
| 5.4.1 Main features of raters' interaction with the holistic scale | 第256-258页 |
| 5.4.2 Main features of raters' interaction with the analytic scale | 第258-259页 |
| 5.5 Summary | 第259-264页 |
| Chapter 6 Conclusion and Discussion | 第264-288页 |
| 6.1 Summary and discussion of findings | 第264-272页 |
| 6.2 Implications | 第272-282页 |
| 6.2.1 Theoretical implications | 第273-276页 |
| 6.2.2 Practical implications | 第276-279页 |
| 6.2.3 Methodological implications | 第279-282页 |
| 6.3 Limitations | 第282-285页 |
| 6.4 Suggestions for further research | 第285-288页 |
| References | 第288-306页 |
| Appendices | 第306-365页 |
