Printer Friendly

Effects of lexical cohesion and macrorules on EFL students' main idea comprehension.

This study investigated the effects of an instructional procedure which incorporated lexical cohesion and macrorules to promote main idea comprehension of Thai EFL students. One hundred and six undergraduate students taking a reading module were randomly assigned to one of three teaching conditions: the combined use of lexical cohesion and macrorules, lexical cohesion alone, and traditional skill-based instruction. The main idea comprehension test was employed to assess students' performance on main idea comprehension. Interviews were used to investigate the strategies use during main idea processing. Post-intervention questionnaires were also applied to examine the students' attitudes towards the teaching models. Statistical analyses revealed a significant improvement in the main idea comprehension post-test of the experimental and control groups. Significance of group differences, however, was not found. The findings suggest that main idea instruction requires a simultaneous intergration of various reading techniques.

Keywords: Lexical Cohesion, Macrorules, Macrostructure, Main Idea Comprehension


Being able to determine the most important content, or macrostructure, of any written discourse is viewed as important in text processing (van Dijk & Kintsch, 1983; Kintsch, 2004; van den Brock et al., 2003; Wang, 2009). This high-order comprehension skill is often a complex task which necessarily involves multiple cognitive activities such as understanding the text structure and constructing a mental representation of the text information (Afflerbach, 1990; Pressley, 1998; van Dijk & Kintsch, 1983; Grabe, 2009; Wang, 2009).

As advocated by van Dijk and Kintsch (1983), discourse processing concerns local and global processes. The local discourse processes, or the microstructures, involve making connections between text segments. In other words, to understand the text structure, the readers must be able to capture semantic relations among elements within a sentence and across sentences in the text. The formation of meaning at this local level is a necessary stage on the way towards the higher-order comprehension of discourse (Vidal-Abarca and Sanjose, 1998). Kintsch and van Dijk (1978) proposed the use of argument repetition, reference to the same events or individuals. In this procedure shared arguments among sentences or propositions are traced in order to infer coherence of a discourse. This approach is claimed to help establish relations between facts denoted by propositions, and to construct a coherent text.

Similar to Kintsch and van Dijk's notion of argument overlap, Hoey (1991) claims that lexical cohesion is a major characteristic of coherent discourse and contributes significantly to the creation and organization of text. In his model, lexical cohesive relations are classified into these categories: simple lexical repetition (e.g. a bear--bears); complex lexical repetition (e.g. a drug--drugging); simple paraphrase (e.g. to sedate--to drug); complex paraphrase (e.g. heat--cold); superordinate, hyponymic (bears--animals), and co-reference repetition (e.g. Mrs. Thatcher--The Prime Minister); membership of a closed (lexical) set (e.g. March--April); personal pronouns (e.g. canal--it); deixis, i.e. demonstrative pronouns (the works of Plato and Aristotle--these writers); ellipsis (a work of art--the work); substitution (tennis balls--ones).

Hoey (1991) proposes the concepts of 'links' and 'bonds.' A link normally occurs when a lexical item is connected with every one of its previous sentences. A bond is a connection between sentences that share three or more links. Sentences, according to Hoey's framework, are categorized into two types: marginal and central. Marginal sentences show no signs of connection with other sentences in a text. Nor do they contribute to the development of the text's theme. Central sentences have a great number of bonds with the rest of the text, and are relevant to the development of the theme. Central sentences can then be used to create summaries of texts. Hoey (1991) asserts that his analysis of lexical cohesion can be utilized in EFL reading instruction. He suggests students be trained to explore shared arguments and to create semantic connections that operate between sentences in a text. The recognition of bonds can be useful because it provides clues for finding the relevant sentences, and gives rapid access to the text content (1991, p. 241).

The global processes, or the macrostructures, determine the overall meanings of the elements, or macrostructures, of the discourse. Coherent meaning representations occurring at this level are constructed by the interpretation of interrelated propositions, depending greatly upon the reader's integration of the message encountered in the text with their prior knowledge (van Dijk and Kintsch, 1983; Kintsch, 2004). Van Dijk and Kintsch (1983) maintain that the process of forming the macrostructure of a discourse generally occurs automatically while reading through macrorules which include:

Deletion: This rule eliminates the information which is not relevant, nor necessary to the interpretation of a global fact, or a central theme of the discourse.

Generalization: In this rule, a list of words or items is replaced by a more general word in the same class. It involves producing a superordinate term (e.g. the term pets for cats, dogs, goldfish and parrots).

Construction: In this rule, several propositions are taken together as one overall action or concept and

replaced by an appropriate proposition that signifies a global fact.

These rules, however, as many researchers argue, merely explain what can be done, but practically cannot be utilized as a precise means of producing macro-propositions (Afflerbach, 1990; Friend, 2001; Kintsch, 2002). In the deletion process, some readers are unable to distinguish between relevant and irrelevant information because they lack the skills to judge what information is important for comprehension. The rules of generalization and construction also require a higher level of comprehension using a top-down process to understand the text content.

Up to the present time, there has been a limited amount of research work concerning discourse macrostructures of first language readers, and even fewer studies in the area have been conducted with second or foreign language students. Gallini and Spires (1995) compared the impact of micro-level, macro-level, and combined micro and macro-level strategies on new text learning. The results indicated that students in the macro-level strategy condition outperformed participants in the other two groups and that the combined micro and macro-level strategy instruction did not reveal significant effects. Friend (2001) applied the argument-referent repetition (using cohesive elements of local coherence strategies) and generalization (using elements of the global relation strategies) models described by van Dijk and Kintsch (1983) to summarization instruction. Participants of her study were randomly assigned to one of three groups: a) argument repetition, b) generalization, and c) control. Analysis of test summaries indicated that instruction in generalization was significantly more effective than argument repetition for generating thesis statements, and that students in the argument repetition and the generalization conditions performed significantly better than those in the self-reflection instruction in determining the most important content of text.

The findings of previous research (e.g. Gallini & Spires, 1995; Friend, 2001), however, do not seem consistent with recent reading theories and research which assert that macro-strategic processes alone are inadequate characterizations of the comprehension of texts (Pressley, 1998; Celce-Murcia and Olshtain, 2000; Kintsch, 2005). Other studies (e.g. Jitendra et al., 2000; Stevens et al., 1991; Spiak, 1999), mostly conducted with native speakers of English, have incorporated microprocessing skills in the investigations, but they have tended to place emphasis on macroprocessing in which a sufficient level of language competence is a basic prerequisite, rather than microprocessing (Dhieb-Henia, 2003). The approaches rely heavily upon the readers' intuitions, world knowledge and knowledge about the text and are more difficult for the low ability students to acquire. These particular types of instruction, therefore, may produce a different result if performed with EFL learners who seem to experience greater difficulties with the linguistic knowledge (Grabe, 2002; Grabe and Stoller, 2002; Liu and Braine, 2005).

Recent studies have placed emphasis on the knowledge of text structure which, as they claimed, enables the reader to recognize connections among text ideas and thus build the global coherence more easily (Wang, 2009). The idea is necessary, especially in an EFL context where students are less efficient in processing English words and knowing how to synthesize the text content (Cohen et al, 1988, cited in Lin, 2008). Perfetti et al (1996) and Yang (2002) propose that, for the instruction of EFL reading, it is necessary to increase knowledge and skills in the foundational elements of text understanding before a reader conducts higher level reading processes. In order to assist EFL readers in accomplishing the overall meaning of a text, more attention to microprocessing should be assigned in the teaching of main idea. In this regard, it would be more appropriate to develop an instructional method that integrates the micro and macro-level strategies, rather than considering them as separate discourse processes.

Purpose and research questions

The current study was intended to examine the combined effect of an instructional treatment which incorporated the micro and macroprocesses on the improvement of Thai EFL students' main idea performance. The microprocess was based on Hoey's (1991) framework and taxonomy of lexical cohesion; the macroprocess applied van Dijk and Kintsch's (1983) macrorules.

A previous study (Wilawan, 2006) investigated the effect of Hoey's lexical patterns and metacognitive strategies on Thai EFL readers' main idea comprehension. The results revealed the positive role of lexical cohesion on the improvement of main idea comprehension. The process of recognizing bonds in the previous study, however, was found to be too complicated, difficult and time-consuming for many Thai EFL learners. The participants were then encouraged to apply only the 'link' model--searching for semantically related words in order to create mental representations. Although these lexical links are considered less reliable than bonded sentences, they can help readers to form text meaning, as Hoey explains 'if a bonding is not recognized, this may well in turn affect the kind of summary a reader is left with--his or her sense of what the text is 'about' (1991, p. 222). In this regard, Hoey's idea of finding lexically related items on the basis of multiple links would assist EFL students in constructing text meaning, thereby generating the central theme more easily. The model would also fill the gap in the macrorules' deletion process by increasing students' sensitivity of importance; students learn to ignore irrelevant, unimportant information, meanwhile focusing on key vocabulary necessary for the interpretation of a global fact. It was expected that lexical cohesive relations would also help students retrieve background knowledge to facilitate higher-order comprehension as required in the generalization and construction rules. Adopting these two approaches as the framework, this study addresses the following questions:

1. Do students who receive the Lexical Cohesion (LC) and the Lexical Cohesion and Macrorules (LCM) instruction achieve significantly higher scores from the pre-test to the post-test as compared to students in the skill-based group (ST)?

2. Are there any significant differences between the mean values of the experimental and control groups? Which of the three instructional methods is superior in producing higher levels of main idea comprehension improvement?

3. Do students of the experimental groups have positive attitudes towards their instructional methods as compared to the control group?

4. Are the main idea processes of students of the LCM class different from those of the LC and ST groups after instruction?

Research Method


The 106 participants were Year 2-4 undergraduate students (aged 18-21) who enrolled in Fundamental English Reading at Kasetsart University in Bangkok. These students had already completed Foundation English III, a prerequisite for Fundamental English Reading, and could be grouped into readers of B 1 intermediate level as classified by the Common European Framework of Reference. They were randomly assigned to one control group with 36 students, and two experimental groups with 36 and 34 subjects. All of them were native speakers of Thai.

After receiving the training, students in each group were requested to participate in the interview. Twenty-two students from three conditions volunteered: 7 from the LC; 9 from the LCM; and 6 from the ST.


Instructional Materials

Twenty-five informational expository passages used to train the students were selected from three ESL/EFL reading workbooks: Six-way Paragraph: Middle Level (Walter, 2000); Reading by Doing: An Introduction to Effective Reading (Simmons & Palmer, 1994); and More Reading Power (Mikulecky & Jeffries, 2003). These workbooks were chosen because main idea comprehension skill was one of their major focuses and they were considered to be the appropriate level for the participants. All texts used for instruction and practice contained either implicit or explicit main idea statements and were the same for all classes. They were also general enough to be understood by students of all disciplines. A practice exercise set of 5 training passages, was prepared for each instructional session. The passages chosen were between 120 and 200 words long.

For the experimental groups (LC and LCM), the reading passages for instructional session 1 were scanned sentence by sentence by the researcher for lexical relations. Lexically linked chains consisting of three or more items in each text were bold-faced and placed in a list at the top of the page to give the students an idea of how related words could be traced, as well as an overview of the theme of the reading selection.

Testing Materials

Two equivalent main idea comprehension tests were created to measure students' comprehension of main ideas. Five months before the intervention began, five test forms were piloted with other 62 students who had already completed Foundation English III. Scores from the 5 piloted tests were afterwards examined for item difficulty; items which had the degree of difficulty of 0.20-0.80 were maintained. Out of the five test forms, 30 test items were carefully selected for the pretest and the posttest; each consisted of 15 items: 10 selection (multiple-choice) responses in Part A and 5 production (constructed main idea) responses in Part B. The 15 test items included both implicit and explicit main ideas. Every single test item contained one English reading passage similar to those employed for training.


Instructional Procedures

The participants attended class from Monday to Thursday (2 hours per session). They were utterly oblivious of the instructional treatment they were receiving. The instruction took place in a naturally occurring classroom setting. The amount of time required for training was 10 hours. Explicit explanations and guided practice were provided at the very beginning of the intervention to reinforce comprehension and application. To avoid confusion the term main idea was defined as a one-sentence summary of the passage (Jitendra et al., 2000). To motivate the students for the best results, they were informed that their scores on the pre and posttest would account for 10% of their course grades.

The three teaching methods were: skill-based teaching (ST); lexical cohesion (LC); and lexical cohesion plus macrorules (LCM). Instruction in all three settings was similar in content, reading load and practice exercises. Participants were encouraged to use their first language (Thai) among their peers during the training in order to ease the linguistic burden, and also to increase understanding of important concepts (Fung et al., 2003). They were directed to apply the main idea processes learned in the context of their classes. The main objective of the instruction was to train students to recognize the main idea, or thematic elements, of a passage more effectively.

Skill-Based Teaching (ST): Students were provided with specific skills necessary for main idea comprehension: finding key words, identifying details and grasping main ideas. The instructor in this traditional method explained the syntactic structure and new vocabulary, and gave a brief introduction to main idea skills (e.g. locating particular portions of paragraphs). Afterwards, students, in small groups of 3-4, were given exercises to practice the skill. The instructor provided assistance when necessary.

Lexical Cohesion (LC): Following Hoey's (1991) framework, the instructor raised students' awareness of lexical relationships and explained how they could help readers in understanding what the text is about. Working collaboratively in groups of 3-4, on practice exercises, they were constantly encouraged to search for word links while reading, as well as to use them to establish connections among concepts across elements in the discourse. The students, then, discussed the topic, the text's content and the possible lexical connections. The instructor periodically checked the students' lists of appropriate lexical links for each text and provided assistance and corrective feedback as needed. When each group arrived at a main idea sentence, the instructors checked the answer. If the answer was inappropriate and therefore could not be a possible main idea, the instructor reminded the students to review their word relations which could help them recognize the thematic elements of the text.

Lexical Cohesion and Macrorules (LCM): The LCM treatment condition incorporated the use of lexical relations and macrorules. Similar to the LC group, the LCM participants were directed to observe elements or ideas in the text which referred to one another so that they could determine what content was most important. The LCM students were also engaged in the application of macrorules (van Dijk and Kintsch, 1983). As proposed by the theory, the rules are claimed to organize text information, condense it into its essential points, thus assisting a reader in indicating or forming the macropropositions.

The illustrations of the LC and LCM models are provided in Appendix B.

Main Idea Comprehension Test: product data

The pre-test was administered to all participants before the experiment. The post-test was carried out on the day after the last session, following the same procedures as the pre-test. Both administrations of the test were organized at exactly the same time for all classes. Students had 1 hour to complete the test.

Each selection response was scored as correct (1 point) or incorrect (0 point); each production response received score of 1 (correct and complete), 0.5 (correct but incomplete) or 0 (incorrect). Written responses were scored on the basis of appropriate, comprehensible main idea sentences rather than spelling and grammatical structure. All tests were independently marked by two raters. Interscorer agreement was calculated using this formula: Agreements/ (Agreements + Disagreements ) x 100. Agreement was found to be 90.92%. Any disagreements were resolved by discussion.

Questionnaire: perception data

A questionnaire was administered after the completion of the instruction (it had been piloted with 44 university students; the reliability coefficient was 0.82). It was developed to elicit the students' feedback concerning the effectiveness of the main idea processes and the acceptability of the training. Each item in Part A offered a closed response format with four options (Likert 4-point rating scale). Students were instructed to indicate whether they agree or disagree with it. For the open-ended questions in Part B, the students were reminded to provide answers for all items and give detailed information (either in Thai or in English) rather than short responses. Students completed the questionnaires anonymously.

Interviews: process data

After the intervention, semi-structured group interviews with students in each group were conducted on the university premises to investigate the main idea processes they had employed in determining the main idea of a passage and how those processes had been used. Interviews lasted between 15 and 20 minutes per group. Interview conversations were tape-recorded.


Research Question 1

Paired samples t-tests were conducted to compare the mean scores of each teaching condition, between the pre-test and the post-test. The level of significance was set at 0.05. The results (as shown in Table 1) indicated significant differences between the mean scores on the pre-test and the post-test of both the experimental groups (the LC and the LCM) and the control group (the ST). That is, students of all groups were successful in enhancing their main idea performance.

Research Question 2

To test the statistical significance of group differences, One-Way ANOVAs were implemented in three different phases for: 1) scores on the selection responses (items 1-10); 2) scores on production responses (items 11-15); and 3) the overall scores (items 1-15). Results from all the analyses revealed no statistically significant differences of mean scores among the three instructional groups (p > .05). In other words, there are no distinctions in the effects of the three teaching models. Tables 2-4 presents the mean difference scores and standard deviations for the main idea comprehension performance on test items 1-10, 11-15 and 1-15, respectively.

Research Question 3

Regarding the students' perceptions towards each instructional method, the descriptive analysis of the close-type questionnaire showed that there were slight distinctions among the groups. The LCM participants tended to more support the training they had received (M = 3.18, SD = 0.20) than their LC (M = 3.04, SD = 0.35) and ST (M = 3.16, SD = 0.30) peers. Responses to the open-ended questions were qualitatively analyzed using the content analysis procedure. Students responded positively to the main idea instructional methods. The LC and LCM students agreed that the lexical cohesion model could help them understand important ideas of texts, although it was not an easy task. Many of them indicated their intent to apply it to their future studies. The ST condition comprised students of differing views; some participants were satisfied with the instruction while others judged that the approach they experienced was something they had already known and, to some students, it did not really improve their main idea comprehension.

Research Question 4

The qualitative analysis of interview responses suggested that the three instructional groups differed in terms of main idea processes. The pattern of the process in the LC condition was similar to that of the LCM group--their main idea process was based on the observation of lexically connected items. The recognition of word connections were claimed to make text information more explicit and provide them with a clearer picture of what the central theme of the text could be. The complaints were about their inadequate vocabulary and the lengthy period of time required in the word connections recognition process. Regarding the LCM instruction, the data suggested that for the deletion rule, many students could distinguish between important and unimportant details more easily, with the assistance of lexical connections. The construction rule was more difficult than deletion and generalization because a good understanding of the text content and a higher level of linguistic knowledge are required. Many LCM students accepted that the simultaneous integration of lexical cohesion and macrorules could facilitate their main idea comprehension and helped them create main idea sentences. The approach, however, according to their responses, was sometimes inadequate for the understanding of the text's main point. Accordingly, it was necessary for them to use other reading skills such as finding meaning of words, understanding the syntactic structure and retrieving background knowledge. Choosing the first sentence of the passage as a main idea statement was also regarded as a last resort, when all attempts had failed.

The interview data also revealed the ST group's application of various techniques during reading. Many ST students focused their attention on finding the meaning of individual words, understanding the grammatical structure of a clause or sentence, translating text information into Thai and making use of the titles or headings of the reading passages. They were guided towards locating stated main ideas in expected positions of a paragraph. That is, the ST participants learned to determine the main idea by selecting either the first or the last sentence of the reading material. Besides, they used 'key words' in understanding the text content. These key words, as they reported, were only word repetitions (i.e. bear--bear, bird--bird) or the vocabulary frequently mentioned by the author. The observation of key words in this teaching condition is distinct from the model used in the LC and LCM classes where students used Hoey's taxonomy of lexical cohesion to find lexically related items.

Discussion and Conclusions

Two observations can be made based on the study's findings. First, the application of lexical cohesion and macrorules alone, was shown to be inadequate for the understanding of macrostructures. Main idea comprehension is a complex task that involves a variety of reading components and knowledge sources (Afflerbach, 1990; Pressley, 1998; Grabe, 2009). General findings can be summed up to show that the formation of main idea requires readers to integrate text-processing skills other than those used in each of the three instructional approaches. The significant increases in the main idea performance of the participants could be the direct consequences of both the LC, LCM and ST treatments and other various reading processes they had employed. Based on the research data, it is clear that none of those processes by themselves were sufficient to facilitate the comprehension of the central theme. Even in the LCM instruction where students were guided towards the combination of lower-order word recognition skills and higher-order comprehension skills, it seemed that other text processing skills were also necessary for the participants to construct text meaning. As revealed by the interview responses, the LCM subjects also utilized other reading skills to compensate for the approaches that seemed ineffective in certain situations. These skills might also positively affect the readers' formulation of main ideas. This finding may provide support for the claim that the determination of main ideas primarily depends on a series of reading techniques--a reader cannot solely rely upon a single approach in order to achieve macrostructures.

Second, some students did not successfully applied the procedures they have practiced (i.e. using lexical cohesive links and the macrorules) while processing texts at the post-test. There are some possible factors that may explain the ineffective use of the procedures. During the intervention, the LC and LCM students were strongly encouraged to apply the reading processes which had been recommended. However, it is probable that the techniques they had practiced were not properly carried out in a real situation, for instance, the post-test. For the LC and LCM conditions, the ineffective use (by some participants) of lexical connections seemed to be a result of their limitations of vocabulary knowledge. Students' inadequate vocabulary discouraged them from practicing and becoming familiar with the procedures. The interpretation of the text is also hindered by this language problem (Perry & MacDonald, 2001; Hirsch, 2003). If the LC and LCM subjects had had a wider vocabulary and improved their skills in recognizing lexical connections, then they should have been able to perform significantly better than the control group at the post-test. Regarding the LCM approach, whereas the participants were capable of applying the deletion and generalization rules, they found the construction rule much more difficult to operate. This is due to the fact that the construction rule required them not only to focus attention to the information contained in the text but also to integrate ideas from several sentences. Some LCM students, as a result, could not operate the rule effectively.

Limitations of the study

There are several limitations of the current research which should be taken into account in future studies. First, regarding the approaches emphasized in the LC and LCM groups, the most obvious features of the instructional approaches were the use of word connections and the macrorules. This was to limit the scope of the present study. However, it does not mean that these processes are the only recommended reading components which the students can utilize to approach text information more efficiently. In fact, as revealed by the interview data, there were many other types of reading techniques employed by the subjects. These techniques might also positively affect the readers' formation of text meaning.

The second limitation lies in the use of lexical cohesive relations. The findings suggest that Thai EFL learners with inadequate command of English vocabulary were found to have greater difficulties in performing the task. The technique of using word connections would have been more efficient if the subjects had possessed higher proficiency in word knowledge. Future research, therefore, should look more closely at the relationship between vocabulary knowledge of skilled and less-skilled EFL readers and their ability to recognize lexical relations.

Another factor which could have affected the research findings was the limit to the amount of time for main idea instruction. It seems essential for the subjects to receive a longer period of instruction, as well as more extensive training in necessary strategies and skills for main idea comprehension, including the recognition of lexical relations and the application of macrorules. Longer practice in the procedures may have assisted the students in performing the task more properly, thereby producing more conclusive results.


Appendix A

Questionnaire on Student's Perception of Instructional Procedure

The survey is specifically for Kasetsart University students taking 01355201: Fundamental English Reading, during the summer session, 2009. It aims to investigate the students' attitudes towards instructional methods in relation to main idea determination.

Your survey responses will be strictly confidential. Your information will be coded and will also remain confidential. Thank you very much for your time and support.
Part A:

Please check the option that applies to you. Choose one answer only.

No.   Attitudes                Strongly   Agree   Disagree   Strongly
                                 Agree                       Disagree
1     The approach you have
      learned help you read

2     Those processes help
      you grasp the main
      idea of a text more

3     You think you will
      score higher on the
      post-test after
      practicing those main
      idea determination

4     You will use those
      processes to help you
      decide the main idea
      of the text you may
      encounter in the

5     You enjoy practicing
      those processes.

6     Those processes make
      you more anxious when
      trying to determine
      main ideas from texts.

7     The main idea processes
      you have practiced are
      useless and waste
      your time.

8     You find it hard to
      perform those processes
      while reading.

9     Those processes give
      you clearer guidelines
      for finding the main
      points in a text.

10    You think English
      reading instructors
      should continue
      teaching those main
      idea determination

11    You think the
      processes you have
      practiced do not
      develop your
      main idea

12    You think you have the
      ability to capture the
      main idea of the text
      you read, regardless
      of those main idea

13    The processes you have
      practiced in class
      motivate you to think
      logically and
      reconsider your

Part B:

Answer the questions about what you do to help yourself in main idea determination.

1. Prior to this course (355201), what approach(es) did you usually use in English reading to grasp the main idea of a text?

2. Do you think the techniques you have practised in class make main idea identification or construction an easier task? In what ways?

3. If you have to read English texts outside the class, will you make use of the strategies you have learned from this course? If not, what technique(s) will you employ in order to help you determine the most important content?

4. What do you think of the classroom atmosphere during main idea comprehension practice?

Appendix B

I. Illustration of the Lexical Cohesion Approach

When a garden warbler sings from trees or bushed, no one can see it. The colours of this songbird match the colours of the leaves. When an animal blends in with its surroundings, its enemies can't see it either. This kind of protection is called camouflage.

Birds must protect themselves from their enemies. Sometimes this means having to fight. Sometimes it means fooling the enemy. Sometimes it means being able to escape. Birds must also protect their eggs and their young. Cats, rats, and foxes love eggs for breakfast. They prowl around looking for eggs and young chicks to eat. How can birds defend themselves against such enemies?

Each species has its own way of defending itself. Birds called common terns fight with their beaks and claws. In a swarm, they peck and scratch at anyone who come too close to their nests. Ostriches protect themselves by escaping. They can't fly, but they can run very fast on their long, muscular legs. These birds can reach speeds of up to forty miles per hour. How fast is that? Well, if the wind blows this hard, it can rip huge branches from trees.

A bird called a killdeer has a lot of courage. It cares very much for its young. It would rather die than see its eggs eaten by a fox. If a fox wanders towards the nest, the killdeer pretends to be hurt. Dragging one wing, it hops away from the nest and draws the hungry fox after it.

* warbler--this song bird--animal--birds--chicks--common terns--ostriches--killdeer

* enemies--enemy

* protection--protect--defend--defending

II. Illustration of the Macrorules Model

1. Deletion

From the reading passage above, students traced lexical links in order to obtain important information. During this process, irrelevant details were eliminated.

2. Generalization

Lexical links were replaced by a superordinate term.

warbler--this song bird--animal birds--chicks-- [right arrow] BIRDS common terns--ostriches--killdeer

3. Construction

Several lexical links (birds, enemies, protect) are taken together as one overall concept and replaced by an appropriate proposition that signifies a global fact. The main idea of the passage above is "Birds have many ways of protecting themselves."


Afflerbach, P.P. (1990). The Influence of Prior Knowledge on Expert Readers' Main Idea Construction Strategies. Reading Research Quarterly, XXV/1, 31-46.

Brit, M.A., & Sommer, J. (2004). Facilitating textual integration with macrostructure focusing tasks. Reading Psychology, 25, 313-339.

Celce-Murcia, M., & Olshtain, E. (2000). Discourse and Context in Language Teaching. Cambridge: Cambridge University Press.

Cohen, H. G., Rosenbaum-Cohen, P. R., Ferrara, J., & Fine, J. (1988). Reading English for specialized purposes: discourse analysis and the use of student informants. In P. L. Carrell, J. Devine, & D. E. Eskey (Eds.), Interactive Approaches to Second Language Reading. Cambridge: Cambridge University Press.

Dhieb-Henia, N. (2003). Evaluating the Effectiveness of Metacognitive Strategy Training for Reading Research Articles in an ESP Context. English for Specific Purposes, 22, 387-417.

Friend, R. (2001). Effects of strategy instruction on summary writing of college students. Contemporary Educational Psychology, 26(1), 3-24.

Fung, I.Y.Y., Wilkinson, I.A.G., & Moore, D.W. (2003). L1-Assisted Reciprocal Teaching to Improve ESL Students' Comprehension of English Expository Text. Learning and Instruction, 13, 1-31.

Gallini, J.K., & Spires, H.A. (1995). Macro-based, Micro-based, and combined strategies in text processing. Reading Psychology: An International Quarterly, 16, 21-41.

Grabe, W. (2002). Foundations for L2 reading instruction. The Language Teacher Online, 26.07, July. Retrieved July 21, 2009, from 02/07/grabe.

Grabe, W. (2009). Reading in a Second Language: Moving from Theory to Practice. Cambridge: Cambridge University Press.

Grabe, W., & Stoller, F.L. (2002). Teaching and Researching Reading. Harlow: Pearson Education.

Hirsch, E.D., Jr. (2003). Reading Compresion Requires Knowledge--of Words and the World. American Educator, Spring, 10-29.

Hoey, M. (1991). Patterns of Lexis in Text. Oxford: Oxford University Press.

Hoey, M. (1994). Pattern of lexis in narrative: A preliminary study. In S-K. Tanskanen, & B. Warvik (Eds.), Topics and Comments: Papers from the Discourse Project, 13, 1-39.

Hoey, M. (2001). Textual Interaction. London: Routledge.

Jitendra, A.K., Hoppes, M.K., & Xin, Y.P. (2000). Enhancing Main Idea comprehension for Students with Learning Problems: The Role of a Summarization Strategy and Self-Monitoring Instruction. The Journal of Special Education, 34 (3), 127-139.

Kintsch, W. (2002). On the notions of theme and topic in psychological process models of text comprehension. In M. Louwerse, & W. van Peer (Eds.), Thematics: Interdisciplinary Studies (pp. 157-170). Amsterdam: Benjamins.

Kintsch, W. (2004). The Construction-Integration Model of Text Comprehension and Its Implications for Instruction. In R. Ruddell, & N. Unrau (Eds.), Theoretical Models and Processes of Reading (pp. 1270-1328). Newark, DE: International Reading Association.

Kintsch, W. (2005). An overview of top-down and bottom-up effects in comprehension: The CI perspective. Discourse Processes, 39 (2 & 3), 125-128.

Kintsch, W., & van Dijk, T. (1978). Toward a model of text comprehension. Psychological Review, 85, 363-394.

Landry, K.L. (2002). Schemata in Second Language Reading. The Reading Matrix, 2(3), 1-14.

Liu, M., & Braine, G. (2005). Cohesive features in argumentative writing produced by Chinese undergraduates. System, 33, 623-636.

Lin, L. (2008). The study of English learners' synthesizing process while reading. Asian EFL Journal, 10 (1), 1-13.

Mikulecky, B.S., & Jeffries, L. (2003). More Reading Power: Reading for Pleasure, Comprehension Skills, Thinking Skills, Reading Faster (Second Edition). New York: Addison-Wesley Longman.

Perfetti, C.A., Marron, M.A., & Foltz, P.T. (1996). Sources of Comprehension Failure: Theoretical Perspectives and Case Studies. In C. Cornoldi, & J. Oakhill (Eds.), Reading Comprehension Difficulties: Processes and Intervention. New Jersey: Lawrence Erlbaum Associates.

Perry, D., & MacDonald, P. (2001). Word Knowledge and Vocabulary Instruction. International Conference on Engineering Education. Oslo, Norway. August 6-10. Retrieved November 4, 2009, from 200 l/Proceedings/papers/201.pdf

Pressley, M. (1998). Reading Instruction That Works: The Case for Balanced Teaching. New York: The Guildford Press.

Simmons, J., & Palmer, B. (1994). Reading by Doing: An Introduction to Effective Reading. Illinois: National Textbook Company.

Spiak, D.S. (1999). Reciprocal Reading and Main Idea Identification. Teaching and Change, 6(2), 212-219.

Stevens, R.J., Slavin, R.E., &Farnish, A.M. (1991). The Effects of Cooperative Learning and Direct Instruction in Reading Comprehension Strategies on Main Idea Identification. Journal of Educational Psychology, 83(1), 8-16.

Van den Broek, P., Lynch, J.S., Naslund, J., levers-Landis, C.E., & Verduin, K. (2003). The Development of Comprehension of Main Ideas in Narratives: Evidence from the Selection of Titles. Journal of Educational Psychology, 95 (4), 707-718.

Van Dijk, T.A. (1980). Macrostructures: An Interdisciplinary Study of Global Structures in Discourse, Interaction, and Cognition. New Jersey: Lawrence Erlbaum Associates.

Van Dijk, T.A., & Kintsch, W. (1983). Strategies of Discourse Comprehension. London: Academic Press.

Vidal-Abarca, E. & Sanjose, V. (1998). Levels of Comprehension of Scientific Prose: The Role of Text Variables. Learning and Instruction, 8(3), 215-233.

Walter, P. (2000). Six-Way Paragraph: Middle Level. Illinois: Jamestown Publishers.

Wang, D. (2009). Factors affecting the comprehension of global and local main idea. Journal of College Reading and Learning, 39(2), 34-52.

Wilawan, S. (2006). Lexical Cohesion and Metacognitive Strategy Training: An Integrated Approach to Main Idea Comprehension. Unpublished PhD Thesis, University of Leeds, UK.

Yamada, K. (2009). Lexical patterns in L2 textual gist identification assessment. Language Testing, 26(1), 101-122.

Yang, Y. (2002). Reassessing Readers' Comprehension Monitoring. Reading in a Foreign Language, 14(1), 18-42.


Department of Foreign Languages

Faculty of Humanities, Kasetsart University

50 Phahonyotin Rd., Chatuchak, Bangkok 10900, Thailand
Table 1: Pre- and Pot-test Mean Scores of the Experimental
and Control Groups

Groups   Measure     N    Mean   SD     t       Sig.

LC       Pre-test    36   5.17   1.65   -7.36   0.00
         Post-test   36   7.78   1.93

LCM      Pre-test    34   4.28   1.76   -7.72   0.00
         Post-test   34   7.24   1.60

ST       Pre-test    36   4.72   2.20   -4.94   0.00
         Post-test   36   6.94   1.94

Note: paired-samples statistics

Table 2: Means and Standard Deviations for Experimental
and Control Groups (Items 1-10)

  Groups               N        Mean      SD

  LC                   36       0.66      1.94
  LCM                  34       0.70      1.93
  ST                   36       0.25      2.56

                   SS      df    MS     F     Sig.

Between groups     4.54     2   2.27   0.48   0.61
Within groups    485.80   103   4.71
Total            490.34   105

Note: one-way ANOVA

Table 3: Means and Standard Deviations for Experimental and
Control Groups (Items 11-15)

  Groups               N        Mean      SD

  LC                   36       1.94      1.21
  LCM                  34       2.25      1.10
  ST                   36       1.97      1.02

                   SS      df    MS     F     Sig.

Between groups     1.97     2   0.98   0.79   0.45
Within groups    128.98   103   1.25
Total            130.96   105

Note: one-way ANOVA

Table 4: Means and Standard Deviations for Experimental and
Control Groups (Items 1-15)

  Groups               N        Mean      SD

  LC                   36       2.61      2.12
  LCM                  34       2.95      2.23
  ST                   36       2.22      2.69

                   SS      df    MS     F     Sig.

Between groups     9.43     2   4.71   0.84   0.43
Within groups    576.96   103   5.60

Total            586.39   105

Note: one-way ANOVA
COPYRIGHT 2011 Project Innovation (Alabama)
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:English as a Foreign Language
Author:Wilawan, Sujunya
Publication:Reading Improvement
Article Type:Report
Geographic Code:1USA
Date:Jun 22, 2011
Previous Article:Exploring two interventions to promote graduate education majors' dispositions toward culturally responsive teaching: taking action to address my...
Next Article:Teachers' leisure reading habits and knowledge of children's books: do they relate to the teaching practices of elementary school teachers?

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |