Feedback on HKDSE Practice Papers School Leaders Briefing
Feedback on HKDSE Practice Papers School Leaders’ Briefing Session 5 March 2012
HKDSE Practice Papers l l Papers for 24 Category A subjects produced to familiarise teachers and students with the format and requirements of the HKDSE exams Funded by EDB Delivered to schools under secure conditions in early January, with provisional marking schemes uploaded the next day for teachers' reference All papers and marking schemes available on HKEAA website from mid February (www. hkeaa. edu. hk/tc/HKDSE/Practice_Papers/index. html) 2
Research Study (Core Subjects) l l l 20 schools selected based on 2010 HKCEE results; stratified sampling but on voluntary basis About 2, 800 students' scripts collected CEML practice papers taken as school mock exams under standard exam conditions, with HKEAA observer Answers scripts scanned and marked onscreen Marking statistics and samples of performance analysed 3
Research Study (Elective Subjects) l l l About 100 to 400 scripts collected from a range of schools Marked and samples of performance analysed Valuable information about actual performance of students Good learning opportunity for HKEAA staff and exam personnel Trial run of marking and grading system 4
Feedback to Schools l l Over 50 briefing sessions conducted for subject teachers between mid January and 29 February Exam requirements, assessment criteria and standards, and annotated samples of performance provided Comments on strengths and weaknesses of candidates and advice given Information available on HKEAA website 5
Limitations of the Research Study l l l Sample size Student readiness Marking done by a small proportion of experienced CE/AL examiners/markers Marks for speaking papers collected from schools for reference, not standardised Only paper-level statistics available at this stage Statistical data to be viewed with caution 6
English Language
Assessment Framework Component Public exam School-based assessment Weighting Duration Paper 1 Reading 20% 1 hr 30 mins Paper 2 Writing 25% 2 hours Paper 3 Listening & Integrated Skills 30% 2 hours Paper 4 Speaking 10% 20 minutes 15% 22
Summary Statistics Paper Max mark Mean (%) S. D. (%) 1 86 38. 2 19. 2 2 210 44. 0 21. 0 3 104 42. 7 22. 5 23
Paper 2 Writing Part A (10%) l Short, guided task l About 200 words l Situation, purpose & relevant information provided Part B (15%) l Longer, more open-ended task l About 400 words l Choice of one out of 8 questions l Each based on one of the 8 modules in the Elective Part 24
Marking l l Double marking by question Three domains for assessment: Content, Language and Organisation, with equal weighting for each 25
Paper 2 Raw Mark Statistics Marks Paper awarded per marker 2 A 21 2 B 21 Mean (%) S. D. (%) 8. 8 (41. 9) 4. 5 (21. 4) 9. 8 (46. 7) 4. 4 (21. 0) 26
Paper 2 B Raw Mark Statistics Popularity No. of cands. % Mean (out of 21) Drama 210 7. 39 6. 96 5. 54 Short stories 209 7. 35 6. 33 4. 72 Poems & songs 101 3. 55 10. 77 5. 13 Popular culture 174 6. 12 7. 29 4. 8 Theme S. D. (out of 21) 27
Paper 2 B Raw Mark Statistics Theme Popularity No. of % cands. Mean (out of 21) S. D. (out of 21) Sports comm. 479 16. 85 11. 02 4. 08 Debating 334 11. 75 12. 05 4. 2 Workplace English 852 29. 97 9. 47 4. 08 Social issues 484 17. 05 10. 91 4. 55 28
Advice for Candidates (1) l l l Read the questions carefully and try to fulfil all the requirements of the task Develop ideas with examples and supporting details Pay attention to who the writer and the intended readers are Don’t use prepared sentences, paragraphs and stock phrases Memorised paragraphs / essays will be disregarded and not awarded any marks. Zero marks will be given to totally memorised essays 29
Advice for Candidates (2) l l l Candidates should not waste time counting the no. of words written. As a general guide, + or – 10% is acceptable (so between 360 to 440 for ‘about 400 words’). Quality is more important than quantity. 30
Paper 1 Reading l l Consists of two parts: Part A - Compulsory Part B - Section 1 (easier) Section 2 (more difficult) Candidates choose either Section 1 or 2 Candidates attempting Section 2 can attain up to Level 5** Candidates attempting Section 1 can attain up to Level 4 only 31
Paper 1 Reading Number of scripts Part A 2, 839 Part B 1 1, 431 (~50%) Part B 2 1, 419 (~50%) * 11 students answered both B 1 and B 2 32
Results in Part A Full Mean Mark (%) SD (%) Reliability MCQ 9 57. 18 23. 33 0. 59 Non-MCQ 31 44. 83 22. 34 0. 90 Total: 40 47. 61 33
Results in Part B 1 Full Mark Mean (%) SD (%) Reliability MCQ 7 35. 38 20. 14 0. 27 Non-MCQ 36 35. 37 20. 10 0. 89 Total: 43 35. 37 Part B 1 34
Results in Part B 2 Full Mark Mean (%) SD (%) Reliability MCQ 12 48. 83 20. 33 0. 56 Non-MCQ 34 39. 48 18. 04 0. 86 Total: 46 41. 92 Part B 1 35
Paper 3 Listening and Integrated Skills Consists of two parts: Part A (compulsory) Part B - Section 1 (easier) - Section 2 (more difficult) l Candidates choose to do either Section 1 or 2 l Candidates attempting Section 2 can attain up to Level 5** l Candidates attempting Section 1 can attain up to Level 4 only l 36
Paper 3 Listening and Integrated Skills Number of candidates: Part 3 A: 2, 845 Part 3 B 1: 1, 332 (46. 74%) Part 3 B 2: 1, 513 (53. 26%) 37
Paper 3 A Results Full Mark Mean (%) 50 49. 04 SD (%) Reliability 24. 53 0. 95 38
Part 3 A Score Distribution 39
Paper 3 Part B 1 Results Task Full Mark Mean (%) SD (%) Reliability 5 12 40. 25 24. 83 0. 83 6 18 34. 13 24. 38 0. 89 7 18 29. 22 24. 84 0. 87 40
Paper 3 Part B 2 Results Task Full Mark Mean (%) SD (%) Reliability 8 18 54. 71 18. 93 0. 81 9 18 53. 9 14. 83 0. 69 10 18 44. 87 20. 31 0. 81 41
Papers 1 and 3: Graded Approach l l l Need to convert B 1 scores to the same scale as B 2 scores After equating, scores can be compared across these two parts of the paper, even though they differ in number of items, difficulty, etc. Paper score = Part A score + Part B score 42
Part 3 A Score Distribution 43
How 3 B 1 Students Performed in 3 A 44
How 3 B 2 Students Performed in 3 A 45
Equating (e. g. Paper 3) 3 B 1 Scores 3 B 2 scores 54 48. . . 0 0 46
Equating 3 B 1 Scores 3 B 2 scores 54 48. . . 0 0 47
Equating 3 B 1 Scores 3 B 2 scores 54 48. . . 0 0 48
Equating 3 B 1 Scores 3 B 2 scores 54 48. . . 0 0 49
Part B 1 to Part B 2 Conversion Factor Paper 1 3 B 1 converted to B 2 score Multiply by 0. 652 Multiply by 0. 735 50
Part B 1 to Part B 2 Conversion Factor l Example: Paper 3 Candidate’s score in 3 B 1 = 26 Candidate’s score after conversion = 26 x 0. 735 = 19 51
Part B 1 to Part B 2 Conversion Factor In live exam l Carried out with 70, 000+ candidates l More sophisticated method used (see page 19 of ‘Grading Procedures and SRR’ booklet) l ‘Conversion factor’ will vary from paper to paper and from year to year, depending on the difficulty of the items in each part l Maximum level a candidate who attempted B 1 can get is Level 4 in the relevant component 52
Liberal Studies
Paper Structure l Paper 1 (50%): Data-response questions l l l 3 questions All compulsory Paper 2 (30%): Extended-response questions l l 3 extended-response questions Answer one question only 62
Marking l l l Marked by question Double marking Third marking and fourth marking whenever necessary l Criterion for 3 rd marking: >3 marks difference between the first two markers l Criterion for 4 th marking: after third marking, no closest pair occurs l 4 th marking by Chief/Assistant Examiner 63
Paper 1: Data-response Questions l l All questions compulsory Full marks of each question: Q 1 (18 marks), Q 2 (20 marks), Q 3 (20 marks) Mean: Q. 1 50. 0%, Q. 2 44. 4%, Q. 3 44. 2%; paper 46. 0% S. D. of paper: 12. 5% 39 students (1. 4%) got 0 marks in the last question which may be due to their problem with time allocation 64
Paper 2: Extended-response Questions l l l Choose ONE question out of three Each question has the same full marks (20 marks) Mean: Q. 1 42. 2%, Q. 2 39. 0%, Q. 3 52. 0%; paper 46. 6% S. D. of paper: 16. 0% Question popularity: l Q. 1 (gender stereotyping): 15% l Q. 2 (Chinese economic growth and energy consumption): 30% l Q. 3 (ethnic minorities in HK): 55% 65
Marking Guidelines l l l Suggested marking criteria for each question for markers' reference Not to be regarded as model answers Alternative answers are acceptable as long as they are reasonable Candidates are entitled to take any position they wish on an issue. They are not to be marked on the ‘correctness of their views' or on how close their views correspond to those of the markers. What counts is whether they can put forward a well-reasoned argument for their view. A script scoring the top mark range need not be perfect. It may contain minor flaws in content, approach or presentation. 66
Thank YOU! 67
- Slides: 67