Online language learning for addressing Hong Kong tertiary students - - PDF document

Online language learning for addressing Hong Kong tertiary students’ needs in academic writing eLearning Forum Asia 2011, NTU Singapore 1 Online language learning for addressing Hong Kong tertiary students’ needs in academic writing

Jonathan J WEBSTER, Angela CHAN, & John S.Y. LEE

The Halliday Centre for Intelligent Applications of Language Studies & Department of Chinese, Translation and Linguistics City University of Hong Kong

City University of Hong Kong City University of Hong Kong

 Established in 1984  3 Colleges, 4 Schools

• College of Business
• College of Liberal Arts & Social Sciences
• College of Science & Engineering
• School of Creative Media
• School of Energy and Environment
• School of Law
• School of Graduate Studies

 Programmes from Bachelor to PhD  > 12,000 FTE students in 2010/11  Ranked 124th in 2009 THES ranking (or 129th in 2010 QS

ranking)

3 3

Online language learning for addressing Hong Kong tertiary students’ needs in academic writing eLearning Forum Asia 2011, NTU Singapore 2

Outline

 The Language Companion Course Project  The CityU Learner corpus of academic writing  Preliminary studies based on the corpus  Further research  Conclusion

4

The Language Companion Course Project

Rationale (1/2)

 Language enhancement is seen as a companion to the

student’s normal course of study.

 Language enhancement is not seen as something

extraneous to the student’s normal learning, instead it must come along side of a designated course which is part of the student’s core curriculum.

 Students are made aware of the fact that their ability

in English language writing goes hand‐in‐hand with their learning in their chosen discipline.

6

Online language learning for addressing Hong Kong tertiary students’ needs in academic writing eLearning Forum Asia 2011, NTU Singapore 3

Rationale (2/2)

 Our aim is to provide a linguistically‐sophisticated

pedagogically‐motivated e‐learning platform for assisting CityU students to improve their English language writing ability within the context of courses in the student’s chosen discipline.

7 subject teacher

• nline

language specialist coach

Involved parties & generic workflow

8

Key:

• 1. Designing an assignment & submission schedule
• 2. Giving an assignment
• 3. Providing support materials for assignment writing
• 4. Submitting a draft for comments
• 5. Returning a commented draft
• 6. Submitting a finalized assignment for assessment

Subject teacher Online tutor coordinator Online tutor Student Subject teacher Online tutor

4 1 2 5 6

Student

3

Online tutor coordinator

Blog‐based interface

Online language learning for addressing Hong Kong tertiary students’ needs in academic writing eLearning Forum Asia 2011, NTU Singapore 4

Built‐in generic word processor

A codified comment is linked to a page showing explanations of the comment

Other types of comments (1/2)

 Additional comments inserted behind a comment

code



Online language learning for addressing Hong Kong tertiary students’ needs in academic writing eLearning Forum Asia 2011, NTU Singapore 5

Other types of comments (2/2)

 End comments provided to point out general

problems

The CityU Learner corpus of academic writing

Downloading student writing

 The blog‐based interface made the downloading of

student essays possible

 The download process took place at the end of each

semester

 Up to five versions of an essay are downloaded:

• The very first version
• The first version with a tutor’s comments
• The second version
• The second version with comments
• The final version submitted for grading

Online language learning for addressing Hong Kong tertiary students’ needs in academic writing eLearning Forum Asia 2011, NTU Singapore 6

Texts by discipline

Faculty Discipline (Abbreviation) No of texts Subtotal (%) Business Business (FB) 2406 4231 (27%) Information Systems (IS) 549 Management Sciences (MS) 1276 Humanities and Social Sciences Applied Social Studies (SS) 877 3470 (23%) Asian and International Studies (AIS) 383 Linguistics (LING) 2210 Science and Engineering Applied Physics (AP) 1001 7312 (48%) Biology (BIO) 3585 Building Science and Technology (BST) 703 Computer Science (CS) 460 Electronic Engineering (EE) 1563 Other faculties Creative Media (CM) 117 379 (2%) Law (LW) 262 Total 15392 (100%)

Genres

 Different subject disciplines gave assignments of

different genres

• Science and Engineering courses lab reports
• Business courses  case study reports
• Linguistics students  descriptions of data analysis
• Social sciences  argumentative essays

 Yet, determining the genre of an assignment is not

always clear‐cut. An assignment may consist of several components and each component constitutes one single genre.

Preliminary studies based on the corpus

Online language learning for addressing Hong Kong tertiary students’ needs in academic writing eLearning Forum Asia 2011, NTU Singapore 7

Study 1:Comment rates

 A computer tool was developed for extracting

comment bank (CB) codes and producing statistics on comment usage in texts.

 Comment rates were compared across drafts and

assignments to measure students’ progress.

 Main findings: the numbers of CB codes inserted

reduced remarkably across almost all subjects as the semester progressed

 The findings suggest that the language tutors’

feedback did help students reduce errors in their academic writing and the effect was carried over into their subsequent work.

Comment rates across versions

6.3 2.6 3.8 3.6 3.3 7.5 6.5 5.3 2.9 2.7 1.9 5.4 4.2 2.2 1.8 1.9 2.4 4.3 3.7 3.8 2.8 1.0 1.3 2.2 AIS_A1 AP_A1 BIO_A1 BST_A2 CS_A2 LING_A2

• No. of CB codes (per 100 words)

Version 1 Version 2

Comment rates across assignments

6.3 7.9 3.8 3.1 7.5 5.3 2.7 7.0 4.4 5.9 2.7 3.9 7.2 2.6 7.6 3.6 4.1 6.5 2.8 1.6 4.5 2.4 5.0 1.9 2.9 4.8 AIS_i AIS_ii AP BIO_i BST CS EE_i EE_ii FB_i FB_ii LING LING_i MS_i

• No. of CB codes (per 100 words)

Assignment 1 Version 1 Assignment 2 Version 1

Online language learning for addressing Hong Kong tertiary students’ needs in academic writing eLearning Forum Asia 2011, NTU Singapore 8

Study 2: Students’ responses to tutors’ comments

 Drew on 35 assignments by 30 students from 3

subject disciplines

 Each assignment was commented on by a language

tutor twice and revised by the author twice

 In total 1901 comments were identified  Each comment was categorized by

• Strategy used to insert the comment
• Type of issue it addresses
• Response taken in the revision
• Outcome of the revision

Language tutors’ commenting strategies (1/2)

 Example 1: insertion of a CB code

• Its aims are to identify who had been [53] most

adversely affected by the movement, and to compare them with had moved [53] into tertiary sectors or remained in manufacturing. (where [53] represents “Verb – past simple”)

 Example 2: OECs which provide explanation for an error (an

example of CB + OEC)

• the job duties they were doing in clause 6) response

[58 ] ("response" is not a verb) to each other. (where [58] represents “word choice”)

 Example 3: OECs which provide revision suggestion

• Conjunction is a mechanism for expressing the links of

semantic connection between spans in a text. (This is the topic sentence for the paragraph about conjunction. Join it to the examples that follow)

Language tutors’ commenting strategies (2/2)

 Example 4: OECs which raised questions for clarification

• I will explore the letter by analyzing how cohesive it is and

thus reach the decision to what extent the letter can be considered as a text. (Aren't you supposed to analyze the letter based on seven criteria? If so, you should state that in the introduction)

 Example 5: Corrections

• The quota samples are also inappropriate since this proposal

aim is to study the SMEs and the large companies (those companies who have fewer [25] (fewer than) 50 employees), so that the research method should include SMEs. (where [25] represents “Preposition – missing”)

 Example 6: Highlighted with no comments

• Also, the reasers had no idea about the research purpose,

whether the researchers were in favour of such policies or they were just curious about why these policies were not widely implemented.

Online language learning for addressing Hong Kong tertiary students’ needs in academic writing eLearning Forum Asia 2011, NTU Singapore 9

Breakdown of comments by commenting strategy

Commenting strategy Token % Insertion of CB code (CB) 1290 67.9% Open‐ended comments (OEC) 379 20.0% Combination of CB and OEC (CB+OEC) 180 9.5% Corrections (Corr.) 45 2.4% Highlighted with no comments (Highlighted) 7 0.4% Total 1901 100%

Breakdown of comments by comment type

Comment type Token % 1 Grammar 1154 60.7% 2 Vocabulary 162 8.5% 3 Clarification 144 7.6% 4 Delete this 124 6.5% 5 Sentence structure 100 5.3% 6 Paragraph structure 54 2.8% 7 Style 48 2.5% 8 General advice 45 2.4% 9 Comments (no response required) 32 1.7% 10 Referencing 11 0.6% 11 Coherence 10 0.5% 12 Essay organization 8 0.4% 13 Others 9 0.5% Total 1901 100%

Breakdown of comments by commenting strategy and comment type

CB OEC CB+OEC Correction Highlighted Total Grammar 942 116 51 40 5 1154 Vocabulary 114 19 27 2 162 Clarification 43 65 34 2 144 Delete this 81 12 30 1 124 Sentence structure 55 22 22 1 100 Paragraph structure 9 41 4 54 Style 31 9 8 48 General advice 45 45 Comments (no response required) 32 32 Referencing 4 3 3 1 11 Coherence 6 3 1 10 Essay organization 5 3 8 Others 9 9 Total 1290 379 180 45 7 1901

Online language learning for addressing Hong Kong tertiary students’ needs in academic writing eLearning Forum Asia 2011, NTU Singapore 10

Breakdown of comments by comment type and response outcome

Revised and correct Revised and incorrect Rewritten and correct Rewritten and incorrect Error source deleted Comment ignored Others Total Grammar 865 63 51 7 89 78 1 1154 Vocabulary 102 22 10 1 12 14 1 162 Clarification 69 12 10 5 12 35 1 144 Delete this 95 5 10 9 5 124 Sentence structure 55 11 8 11 15 100 Paragraph structure 32 2 6 1 12 1 54 Style 30 1 3 6 8 48 General advice 45 45 Referencing 7 1 1 1 1 11 Coherence 6 1 2 1 10 Essay

• rganization

5 3 8 Others 5 2 1 1 9 Total 1271 118 95 17 142 173 92 1869

Summary of the preliminary studies

 We found that most of the tutors’ comments are

related to grammatical problems and the majority of students were willing to revise their errors as advised.

 It appears that language tutors’ comments were able

to reduce students’ errors in their subsequent works.

 However, these findings were drawn on a limited

number of essays and therefore should be interpreted with caution and be seen as suggestive rather than

• conclusive. It is our plan to further investigate into

these issues with a larger sample size from the learner corpus under development.

Further research

Online language learning for addressing Hong Kong tertiary students’ needs in academic writing eLearning Forum Asia 2011, NTU Singapore 11

Nesi et al. (2004: 447)

One characteristic of writing corpora is focus on product, and this might be seen as a limitation for the use of corpora in academic writing research into composing

• processes. Whereas the assignments in most writing

databanks are assessed products, a significant line of enquiry for writing developers is writing processes [...]. Hence, from a process research point of view, it would be ideal if writing corpora could be designed to include, for instance, assignment guidelines, student plans and drafts, and teachers’ comments.

Our corpus

 Large quantity of texts from a range of subject

discipline

 A variety of data: students’ drafts, language tutor’s

comments, and assignment guidelines (for most of the assignments)

 Wide coverage of student writings from all levels from

poor to good

 The corpus is able to provide a more complete

picture of student writing and would be useful to facilitate understanding of specific writing needs of students at different levels.

Future plans

 We plan to expand our early study on language tutors’

comments and students’ responses to the comments.

 We also plan to investigate the progress of repeat

participants to see how essential academic writing skills are acquired over the semesters and to explore the short‐ /long‐term effects of tutors’ feedback on students’ writing.

 The LCC project has involved different tutor groups,

resulting in variation in practice across the different groups

• f language tutors. Such variation may have influenced

students’ responses towards the comments they received. We plan to explore this in detail so as to identify which approach better suits Hong Kong students.

Online language learning for addressing Hong Kong tertiary students’ needs in academic writing eLearning Forum Asia 2011, NTU Singapore 12

Conclusion (1/2)

 We illustrated an online platform designed to provide

feedback on students’ essays so as to help students improve their English language writing ability.

 We also introduced a learner corpus developed from the

data collected from the online platform. Currently, more than 15300 texts from thirteen disciplines have been collected, consisting of language tutors’ comments and students’ drafts as well as assignment guidelines.

 Initial studies based on the corpus data have shown that

the corpus can be used to conduct empirical research in many aspects in language teaching and e‐learning, including L2 students’ writing processes and the effect of language tutors’ feedback on students’ writing.

Conclusion (2/2)

 As the project involved language tutors from different

institutions, different approaches to student essays were observed in the data. The data would be useful to explore which approach is better suited to helping students develop their academic writing skills in the Hong Kong context.

 By using naturalistic (instead of experimental) data,

the corpus provides researchers with a more complete picture of Hong Kong tertiary students’ needs in academic writing for their subject disciplines. We believe that the corpus will make a significant contribution to the research and pedagogy of academic writing.