- Email: [email protected]
Using CALL to bridge the culture gap between discourse communities and ESP learners
Available online at www.sciencedirect.com
Procedia - Social and Behavioral Sciences 34 (2012) 122 – 127
Languages, Cultures and Virtual Communities Les Langues, les Cultures et les Communautés Virtuelles
Using CALL to bridge the culture gap between discourse communities and ESP learners Danielle Joulia* Département Informatique, IUT Paul Sabatier Toulouse 3, 133 B avenue de Rangueil, F-31077 Toulouse cedex 4, France
Abstract Learners of English for computer science have to master one essential skill: reading technical documentation. This documentation, available online and consulted whenever programmers need information to code programs, is barely understood by French first-year university students in computing. Faced with these professional texts, they experience linguistic as well as cultural problems due to lack of fluency in English and lack of knowledge in computer science. Our pedagogic aim is to lighten the cognitive overload induced when they read texts written by members of the discourse community. To this end, we have developed a software application to help students understand these authentic texts. Based on computer-assisted reading research and socio-constructivist learning theories, different types of aids were designed to alleviate the learner's difficulties and make the input comprehensible. Our software provides scaffolding on the macrostructural level (textual and rhetorical organisation) and the microstructural level (lexis and syntax). It was tested with 112 students who had to answer comprehension questions and carry out professional tasks similar to the activities programmers undertake in real life. We present the results of our experiment and reflect on a genre-centred, action-oriented, task-based learning approach that requires the learner to participate actively in these computer-mediated learning systems. © 2012 Published by Elsevier B.V. Selection and/or peer-review under responsibility of EUROCALL2010 Scientific © 2012 Published by Elsevier Ltd. under responsibility of Dominique Macaire and Alex Boulton Committee Keywords: computer-assisted reading; ESP; English for computing; scaffolding;TBL
* Danielle Joulia. Tel.: +00 335 622 588 09 [email protected]
1877-0428 © 2012 Published by Elsevier B.V. Selection and/or peer-review under responsibility of EUROCALL2010 Scientific Committee doi:10.1016/j.sbspro.2012.02.025
Danielle Joulia / Procedia - Social and Behavioral Sciences 34 (2012) 122 – 127
1. Purposes Our aim is to make learners into better processors of information, i.e. to improve reading skills in their own specialist field. To this end, we need to: x develop an interactive approach to reading; x combine lower-level decoding processes with higher-level comprehension and interpretation processes (Ariew & Ercetin, 2004). In other words, we need to help automatize bottom-up processing (word recognition) at the local level and facilitate top-down processing (anticipation, selection, inferencing, etc.) at the global level, the latter usually short-circuited by the linguistic threshold (i.e. cognitive overload induced by focusing on low-level process strategies). In short, our purpose is to reduce the cognitive overload to allow high-level processes, in order to achieve a deeper level of comprehension (Plass, 1998). The instructional design is based on the following features: x cognitive approach: process-oriented; x socio-constructivism: learner-centered, problem-solving, scaffolding; x action-oriented approach: academic tasks related to real-life professional activities, Common European Framework of Reference, macro-task; x task-based learning: focus on meaning, authenticity, using language as a means to an end; x interactionist view of SLA: modification of input, comprehensible input, negotiation of meaning (Chapelle, 1998); x genre-based: exemplars written by target discourse communities, top-down approach. 2. Materials A computer-assisted reading application was developed by the author (Joulia, 2011), involving: x three types of hypertext aids to facilitate the processes involved in reading and enhance comprehension: ¾ macro scaffolding: 1. content schemata: - 35 concept glosses, providing a definition or explanation in French (L1) of key computing concepts (Lenders, 2008) Æ to give learners domain-specific knowledge 2. formal schemata: - textual model of a Unix Manual page - rhetorical organisation of the Description section: outline, overview map, tree graph, with hyperlinks, highlighting the corresponding passage in text Æ to prompt learners to build internal connections - key words, topic sentences highlighted - simplified version of the text available (Yano et al., 1994) Æ to prompt learners to notice and select important information ¾ micro scaffolding: 3. linguistic schemata:
123
124
Danielle Joulia / Procedia - Social and Behavioral Sciences 34 (2012) 122 – 127
- 89 linguistic glosses (lexis and syntax), providing a translation in French (unsignaled glosses, displayed in pop-up windows) (De Ridder, 2002) Æ to automatize linguistic decoding x a teaching agent (Robby) giving advice at key stages x a tracker to record the number, type and order of aids consulted, reading times, and answers to comprehension questions into a log-file x development using Toolbook.
Figure 1. User interface showing text, concept and linguistic glosses, highlighted keywords, outline of the description section with enhanced passage selected
3. Method The study aimed to explore how hypertext aids influence the level of comprehension in the following context: x 112 first-year university students in computing (28 x 4 groups) participated in the study during their regular class period. x Students were grouped into two levels of L2 linguistic ability (low and intermediate) based on their IELTS reading scores and high school grades in the L2. Half of each group was given access to the
Danielle Joulia / Procedia - Social and Behavioral Sciences 34 (2012) 122 – 127
hypertext aids (experimental groups) while the other half was not (control groups). x The task was to read and understand the text about a Unix command: chmod. The students were instructed to read carefully and told they would be asked questions about the text later. x After task completion, participants were tested with a series of exercises which allowed us to assess the level of comprehension they had achieved. x The levels of comprehension were measured (Van Dijk & Kintsch, 1983): ¾ surface structure level (linguistic representation) Æwith a true/false recognition task ¾ textbase / macrostructure level (semantic representation) Æwith a true/false recognition task and a recall task in L1 (summary of command) ¾ situation model level (explicit and implicit text content representation, deeper level of comprehension) Æwith a true/false inference recognition task and practise exercises (transfer tasks similar to professional tasks) the text was again available for this series of tasks. 4. Results Research question 1: Did the experimental groups demonstrate better comprehension than the control groups? Yes. The mean score for correct answers was significantly higher for the experimental groups, but the differences were moderate and poor comprehension scores were observed. The effect of L2 ability was more important than that of aids. No significant correlation was noted between look-up behaviour and performance on test (Chun & Plass, 1996). Q1 : Comprehension
Scores 0-30
16 14 12 10
15.63
8 6 4 2 0
13.14 9.55 7.29
G1A
G1WA
G2A
G2WA
G1A: middle L2 ability group, with aids G1WA: middle L2 ability group, without aids G2A: low L2 ability group, with aids G2WA: low L2 ability group, without aids G1A / G1WA: F (1,54) = 4.54, p = .0376 G2A / G2WA: F (1,54) = 7.41, p = .0087
Groups
Figure 2. Summary of results
Research question 2: Did macrostructural input enhancement help to construct a textbase? No. No strong textbase was created. No significant correlation between consulting of macro aids and performance on the recall task was found (Ercetin, 2010). Macro aids (outline, overview map, tree graph, topic sentences, simplified version of text) were moderately used (Lomicka, 1998).
125
126
Danielle Joulia / Procedia - Social and Behavioral Sciences 34 (2012) 122 – 127
Q2: Textbase
Scores 0-10
5 4 4 3 3 2 2 1 1 0
4.18 3.61
2.12 1.34
G1A
G1WA
G2A
G2WA
Groups Figure 3. Results for macrostructural enhancement and textbase
Research question 3: Did low L2 ability students consult more aids than middle L2 ability students? Yes, for overall look-up (F (1, 54) = 8.62, p = .0049) and for glosses (F (1, 54) = 6.91, p = .0111). No, for macro consulting, no significant difference was found. Q3: Consulting of aids Macro aids
Glosses
60 50 40 Nb of 30 look-ups 20
58 38 13
15
G1A Groups
G2A
10 0 G1A
G2A
Figure 4. Results for proficiency and consultation
5. Conclusion The overall conclusions of the study were as follows: x students did not develop an interactive approach to reading, and did not activate high-level process strategies; x they did not achieve a deeper level of comprehension, and the situation model was not constructed; x they did not understand the relevance of macro aids, their use of aids was oriented toward the goal of translating, and they were satisfied with a minimal level of comprehension: ¾ the teacher’s objective needs contrast with the students’ subjective needs;
Danielle Joulia / Procedia - Social and Behavioral Sciences 34 (2012) 122 – 127
x the software program required the learners to participate actively, to be autonomous and take responsibility for their own learning: ¾ visualising the macrostructure is not sufficient to build a textbase, learners need more guidance and macro exercises to prompt them to notice the macrostructure of the text; x the tasks represented considerable conceptual and procedural difficulty requiring metacognitive strategies, and need to be simplified (i.e. use only the simplified version of the text); x top-down strategies and an interactive approach are long term processes, developed over longer periods of time; they can be instructed but the learners’ support is needed; x a discrepancy was noted between theoretical models and the students’ behaviour (internal and external constraints).
References Ariew, R., & Ercetin, G. (2004). Exploring the potential of hypermedia annotations for second language reading. Computer Assisted Language Learning, 17(2), 237-259. Chapelle, C. (1998). Multimedia CALL: Lessons to be learned from research on instructed SLA. Language Learning & Technology, 2(1), 22-34. Chun, D., & Plass, J. (1996). Facilitating reading comprehension with multimedia. System, 24(4), 503-519. De Ridder, I. (2002). Visible or invisible links: Does the highlighting of hyperlinks affect incidental vocabulary learning, text comprehension and the reading process? Language Learning & Technology, 6(1), 123-146. Ercetin, G. (2010). Effects of topic interest and prior knowledge on text recall and annotation use in reading a hypermedia text in the L2. ReCALL, 22(2), 228-246. Joulia, D. (2011). Faciliter la lecture-compréhension de documentations informatiques en anglais : de l'analyse à l'expérimentation didactique assistée par ordinateur (Facilitating reading comprehension in English of computer science documentation:from theory to computer-assisted teaching experiment). Unpublished doctoral dissertation. University of Toulouse. Lenders, O. (2008). Electronic glossing: Is it worth the effort? Computer Assisted Language Learning, 21(5), 457-483. Lomicka, L. (1998). To gloss or not to gloss: An investigation of reading comprehension online. Language Learning & Technology, 1(2), 41-50. Plass, J. (1998). Design and evaluation of the user interface of foreign language multimedia software: A cognitive approach. Language Learning & Technology, 2(1), 35-45. Van Dijk, T., & Kintsch, W. (1983). Strategies of discourse comprehension. New York: Academic Press. Yano, Y., Long, M., & Ross, S. (1994). The effects of simplified and elaborated texts on foreign language reading comprehension. Language Learning, 44, 189-219.
127
Procedia - Social and Behavioral Sciences 34 (2012) 122 – 127
Languages, Cultures and Virtual Communities Les Langues, les Cultures et les Communautés Virtuelles
Using CALL to bridge the culture gap between discourse communities and ESP learners Danielle Joulia* Département Informatique, IUT Paul Sabatier Toulouse 3, 133 B avenue de Rangueil, F-31077 Toulouse cedex 4, France
Abstract Learners of English for computer science have to master one essential skill: reading technical documentation. This documentation, available online and consulted whenever programmers need information to code programs, is barely understood by French first-year university students in computing. Faced with these professional texts, they experience linguistic as well as cultural problems due to lack of fluency in English and lack of knowledge in computer science. Our pedagogic aim is to lighten the cognitive overload induced when they read texts written by members of the discourse community. To this end, we have developed a software application to help students understand these authentic texts. Based on computer-assisted reading research and socio-constructivist learning theories, different types of aids were designed to alleviate the learner's difficulties and make the input comprehensible. Our software provides scaffolding on the macrostructural level (textual and rhetorical organisation) and the microstructural level (lexis and syntax). It was tested with 112 students who had to answer comprehension questions and carry out professional tasks similar to the activities programmers undertake in real life. We present the results of our experiment and reflect on a genre-centred, action-oriented, task-based learning approach that requires the learner to participate actively in these computer-mediated learning systems. © 2012 Published by Elsevier B.V. Selection and/or peer-review under responsibility of EUROCALL2010 Scientific © 2012 Published by Elsevier Ltd. under responsibility of Dominique Macaire and Alex Boulton Committee Keywords: computer-assisted reading; ESP; English for computing; scaffolding;TBL
* Danielle Joulia. Tel.: +00 335 622 588 09 [email protected]
1877-0428 © 2012 Published by Elsevier B.V. Selection and/or peer-review under responsibility of EUROCALL2010 Scientific Committee doi:10.1016/j.sbspro.2012.02.025
Danielle Joulia / Procedia - Social and Behavioral Sciences 34 (2012) 122 – 127
1. Purposes Our aim is to make learners into better processors of information, i.e. to improve reading skills in their own specialist field. To this end, we need to: x develop an interactive approach to reading; x combine lower-level decoding processes with higher-level comprehension and interpretation processes (Ariew & Ercetin, 2004). In other words, we need to help automatize bottom-up processing (word recognition) at the local level and facilitate top-down processing (anticipation, selection, inferencing, etc.) at the global level, the latter usually short-circuited by the linguistic threshold (i.e. cognitive overload induced by focusing on low-level process strategies). In short, our purpose is to reduce the cognitive overload to allow high-level processes, in order to achieve a deeper level of comprehension (Plass, 1998). The instructional design is based on the following features: x cognitive approach: process-oriented; x socio-constructivism: learner-centered, problem-solving, scaffolding; x action-oriented approach: academic tasks related to real-life professional activities, Common European Framework of Reference, macro-task; x task-based learning: focus on meaning, authenticity, using language as a means to an end; x interactionist view of SLA: modification of input, comprehensible input, negotiation of meaning (Chapelle, 1998); x genre-based: exemplars written by target discourse communities, top-down approach. 2. Materials A computer-assisted reading application was developed by the author (Joulia, 2011), involving: x three types of hypertext aids to facilitate the processes involved in reading and enhance comprehension: ¾ macro scaffolding: 1. content schemata: - 35 concept glosses, providing a definition or explanation in French (L1) of key computing concepts (Lenders, 2008) Æ to give learners domain-specific knowledge 2. formal schemata: - textual model of a Unix Manual page - rhetorical organisation of the Description section: outline, overview map, tree graph, with hyperlinks, highlighting the corresponding passage in text Æ to prompt learners to build internal connections - key words, topic sentences highlighted - simplified version of the text available (Yano et al., 1994) Æ to prompt learners to notice and select important information ¾ micro scaffolding: 3. linguistic schemata:
123
124
Danielle Joulia / Procedia - Social and Behavioral Sciences 34 (2012) 122 – 127
- 89 linguistic glosses (lexis and syntax), providing a translation in French (unsignaled glosses, displayed in pop-up windows) (De Ridder, 2002) Æ to automatize linguistic decoding x a teaching agent (Robby) giving advice at key stages x a tracker to record the number, type and order of aids consulted, reading times, and answers to comprehension questions into a log-file x development using Toolbook.
Figure 1. User interface showing text, concept and linguistic glosses, highlighted keywords, outline of the description section with enhanced passage selected
3. Method The study aimed to explore how hypertext aids influence the level of comprehension in the following context: x 112 first-year university students in computing (28 x 4 groups) participated in the study during their regular class period. x Students were grouped into two levels of L2 linguistic ability (low and intermediate) based on their IELTS reading scores and high school grades in the L2. Half of each group was given access to the
Danielle Joulia / Procedia - Social and Behavioral Sciences 34 (2012) 122 – 127
hypertext aids (experimental groups) while the other half was not (control groups). x The task was to read and understand the text about a Unix command: chmod. The students were instructed to read carefully and told they would be asked questions about the text later. x After task completion, participants were tested with a series of exercises which allowed us to assess the level of comprehension they had achieved. x The levels of comprehension were measured (Van Dijk & Kintsch, 1983): ¾ surface structure level (linguistic representation) Æwith a true/false recognition task ¾ textbase / macrostructure level (semantic representation) Æwith a true/false recognition task and a recall task in L1 (summary of command) ¾ situation model level (explicit and implicit text content representation, deeper level of comprehension) Æwith a true/false inference recognition task and practise exercises (transfer tasks similar to professional tasks) the text was again available for this series of tasks. 4. Results Research question 1: Did the experimental groups demonstrate better comprehension than the control groups? Yes. The mean score for correct answers was significantly higher for the experimental groups, but the differences were moderate and poor comprehension scores were observed. The effect of L2 ability was more important than that of aids. No significant correlation was noted between look-up behaviour and performance on test (Chun & Plass, 1996). Q1 : Comprehension
Scores 0-30
16 14 12 10
15.63
8 6 4 2 0
13.14 9.55 7.29
G1A
G1WA
G2A
G2WA
G1A: middle L2 ability group, with aids G1WA: middle L2 ability group, without aids G2A: low L2 ability group, with aids G2WA: low L2 ability group, without aids G1A / G1WA: F (1,54) = 4.54, p = .0376 G2A / G2WA: F (1,54) = 7.41, p = .0087
Groups
Figure 2. Summary of results
Research question 2: Did macrostructural input enhancement help to construct a textbase? No. No strong textbase was created. No significant correlation between consulting of macro aids and performance on the recall task was found (Ercetin, 2010). Macro aids (outline, overview map, tree graph, topic sentences, simplified version of text) were moderately used (Lomicka, 1998).
125
126
Danielle Joulia / Procedia - Social and Behavioral Sciences 34 (2012) 122 – 127
Q2: Textbase
Scores 0-10
5 4 4 3 3 2 2 1 1 0
4.18 3.61
2.12 1.34
G1A
G1WA
G2A
G2WA
Groups Figure 3. Results for macrostructural enhancement and textbase
Research question 3: Did low L2 ability students consult more aids than middle L2 ability students? Yes, for overall look-up (F (1, 54) = 8.62, p = .0049) and for glosses (F (1, 54) = 6.91, p = .0111). No, for macro consulting, no significant difference was found. Q3: Consulting of aids Macro aids
Glosses
60 50 40 Nb of 30 look-ups 20
58 38 13
15
G1A Groups
G2A
10 0 G1A
G2A
Figure 4. Results for proficiency and consultation
5. Conclusion The overall conclusions of the study were as follows: x students did not develop an interactive approach to reading, and did not activate high-level process strategies; x they did not achieve a deeper level of comprehension, and the situation model was not constructed; x they did not understand the relevance of macro aids, their use of aids was oriented toward the goal of translating, and they were satisfied with a minimal level of comprehension: ¾ the teacher’s objective needs contrast with the students’ subjective needs;
Danielle Joulia / Procedia - Social and Behavioral Sciences 34 (2012) 122 – 127
x the software program required the learners to participate actively, to be autonomous and take responsibility for their own learning: ¾ visualising the macrostructure is not sufficient to build a textbase, learners need more guidance and macro exercises to prompt them to notice the macrostructure of the text; x the tasks represented considerable conceptual and procedural difficulty requiring metacognitive strategies, and need to be simplified (i.e. use only the simplified version of the text); x top-down strategies and an interactive approach are long term processes, developed over longer periods of time; they can be instructed but the learners’ support is needed; x a discrepancy was noted between theoretical models and the students’ behaviour (internal and external constraints).
References Ariew, R., & Ercetin, G. (2004). Exploring the potential of hypermedia annotations for second language reading. Computer Assisted Language Learning, 17(2), 237-259. Chapelle, C. (1998). Multimedia CALL: Lessons to be learned from research on instructed SLA. Language Learning & Technology, 2(1), 22-34. Chun, D., & Plass, J. (1996). Facilitating reading comprehension with multimedia. System, 24(4), 503-519. De Ridder, I. (2002). Visible or invisible links: Does the highlighting of hyperlinks affect incidental vocabulary learning, text comprehension and the reading process? Language Learning & Technology, 6(1), 123-146. Ercetin, G. (2010). Effects of topic interest and prior knowledge on text recall and annotation use in reading a hypermedia text in the L2. ReCALL, 22(2), 228-246. Joulia, D. (2011). Faciliter la lecture-compréhension de documentations informatiques en anglais : de l'analyse à l'expérimentation didactique assistée par ordinateur (Facilitating reading comprehension in English of computer science documentation:from theory to computer-assisted teaching experiment). Unpublished doctoral dissertation. University of Toulouse. Lenders, O. (2008). Electronic glossing: Is it worth the effort? Computer Assisted Language Learning, 21(5), 457-483. Lomicka, L. (1998). To gloss or not to gloss: An investigation of reading comprehension online. Language Learning & Technology, 1(2), 41-50. Plass, J. (1998). Design and evaluation of the user interface of foreign language multimedia software: A cognitive approach. Language Learning & Technology, 2(1), 35-45. Van Dijk, T., & Kintsch, W. (1983). Strategies of discourse comprehension. New York: Academic Press. Yano, Y., Long, M., & Ross, S. (1994). The effects of simplified and elaborated texts on foreign language reading comprehension. Language Learning, 44, 189-219.
127