Scholarly article on topic 'A Study of Students’ Reading Performance in Two Test Formats of Summary Writing and Open-ended Questions'

A Study of Students’ Reading Performance in Two Test Formats of Summary Writing and Open-ended Questions Academic research paper on "Educational sciences"

CC BY-NC-ND
0
0
Share paper
OECD Field of science
Keywords
{"Reading comprehension" / "reading assessment" / "reading proficiency" / "summary writing" / "open-ended questions"}

Abstract of research paper on Educational sciences, author of scientific article — Leila Hassani, Tengku Nor Rizan Tengku Mohamad Maasum

Abstract The purpose of this research is to investigate the impact of two test formats summary writing and open-ended questions on students’ reading performance and its relationship to students’ English language proficiency. An expository passage was used for two test formats. In addition a reading component of TOEFL test was applied to investigate students’ English language proficiency on 35 postgraduate students. Based on descriptive and inferential statistics, the findings were revealed that there was no significant difference of the test takers’ reading performance on the two test methods of summary writing and open-ended questions. In addition, results of t-test for the two test methods was significant (p<.05). Consequently the test takers had a better performance in summary writing than open-ended questions.

Academic research paper on topic "A Study of Students’ Reading Performance in Two Test Formats of Summary Writing and Open-ended Questions"

Available online at www.sciencedirect.com

SciVerse ScienceDirect PfOCSCl ¡0

Procedia - Social and Behavioral Sciences 69 (2012) 915 - 923

International Conference on Education and Educational Psychology (ICEEPSY 2012)

A Study of Students' Reading Performance in Two Test Formats of Summary Writing and Open-ended Questions

Leila Hassani, Tengku Nor Rizan Tengku Mohamad Maasum *

Universiti Kebangsaan Malaysia, Bangi, Selangor,43600, Malaysia Universiti Kebangsaan Malaysia, Bangi, Selangor, 43600, Malaysia

Abstract

The purpose of this research is to investigate the impact of two test formats summary writing and open-ended questions on students' reading performance and its relationship to students' English language proficiency. An expository passage was used for two test formats. In addition a reading component of TOEFL test was applied to investigate students' English language proficiency on 35 postgraduate students. Based on descriptive and inferential statistics, the findings were revealed that there was no significant difference of the test takers' reading performance on the two test methods of summary writing and open-ended questions. In addition, results of t-test for the two test methods was significant (p<.05). Consequently the test takers had a better performance in summary writing than open-ended questions.

© 2012 The Authors. Published by Elsevier Ltd.

Selectionandpeer-reviewunder responsibilityofDr.Zafer Bekirogullari ofCognitive - Counselling,Research&Conference Services C-crcs.

Keywords: reading comprehension; reading assessment; reading proficiency; summary writing; open-ended questions

1. Introduction

Assessment was undertaken for different purposes. Teachers used it as an important tool to provide a wealth of information to guide classroom practice and to manage learning and learners. It could be used to improve instruction and help students take control of their learning (Bostwick & Gakuen, 1995).

Testing was a form of assessment which was typically used at end of a stage of instruction to measure students' achievement (Hedge, 1988). According to Hedge (1998), testing was referred to the specific procedures that teachers and examiners employed to try to measure learners' ability in the language. Hedge (1998) believed that good tests provided the opportunity for learners to show how much they knew about language structure and vocabulary, as well as how they were able to use these formal linguistic features to convey meaning in classroom language activities as

* Leila Hassani. Tel.: +60-173-958-870; fax: +98-242-522-3684. E-mail address: hassani.leila320@gmail.com.

1877-0428 © 2012 The Authors. Published by Elsevier Ltd.

Selection and peer-review under responsibility of Dr. Zafer Bekirogullari of Cognitive - Counselling, Research & Conference

Services C-crcs.

doi: 10.1016/j.sbspro.2012.12.016

listening, speaking, writing and reading. Tests could be used as part of an integrated assessment system and applied to measure learners' language ability.

Reading assessment was an essential part of teaching-learning process. Students' performance could be measured by different procedures. Assessment for reading comprehension in ESL/EFL was necessary in order to have a working knowledge of what that process entails (Cohen, 1993). There were many available language tests to assess students' reading performance based on their individuals' language abilities (Bachman & Palmer, 1996). However the students' language ability affected their reading performance on language tests and eventually the quality of the information obtained from their tests (Bachman & Palmer, 1996).

The types of the test formats used in language tests were important as one test task could be more appropriate for a given purpose than another. Multiple-choice questions, as one of the test formats, were categorized under the select-type tests. They were a common device to test students' reading comprehension that allowed test constructors to control the range of possible answers to comprehension questions. In this test format, students should select the correct response from the available alternatives prepared for each question. The most important advantage of all select-type tests unlike other test formats was not time-consuming in administration and grading the papers. But the most significant disadvantage of all select-type tests was the chance of guessing and cheating to find the correct response (Alderson, 1990).

Pearson and Hamm (2005) believed that most of the traditional methods of assessing reading comprehension with standardized tests and multiple-choice questions were the most frequent types of assessment. They were used in commercial reading tests, state-mandated achievement tests and tests used to compare reading proficiency among students. In order to improve the effectiveness of the measurement of reading performance, researchers and test constructers should examine and construct appropriate test format to draw a valid estimation of the readers' comprehension of the text.

Besides considering the type of test formats in the assessment of reading, the test constructers or users needed to give careful thought to the assessment method. These variable test formats, which included summary writing, open-ended (short-answer) questions, multiple-choice questions (MCQ), oral recall and written recall (Ozuru et al., 2007; Hudson, 2007; Meyer, 2003; Alderson, 2000), could be applied to assess reading performance on students of a text. Each format had its own advantages and disadvantages but for the purpose of this paper only two test formats that were, summary writing and open ended questions were taken into consideration.

In summary writing, a reader was asked to prepare a summary from a reading passage in his own words. Through this test format, the researcher was able to assess students' ability in identifying the main ideas in the passage and the way reader put these ideas together in his own words to create a coherent summary of the passage. Sometimes, there were situations whereby the students copied the text and transferred it to the summary section. In such cases, the student would lose marks on summary writing as he had not developed a summary in his own words in a synthesized manner. The advantage of summary writing in assessing reading comprehension was that it assessed reading comprehension as a whole and it did not break reading into parts. Another advantage of summary writing in assessing reading was that it was a supply-type test format and there was minimal chance for cheating and guessing the correct answer without reading the passage. In this test method, the reader should read the text thoroughly, comprehend the gist of the passage and identify the important parts that carried the core meaning to enable the students to develop an effective summary. Therefore if the summary could relate to a real-world task, the adequacy of the response would be easier to establish (Alderson, 2000).

Another test format in assessing reading was open-ended or short-answer questions. These two terminologies open- ended and short-answer questions were used interchangeably in the reading literature. Open-ended questions like summary writing were categorized under the supply-type tests. Open-ended questions could be used to check the literal knowledge (the answers exist in the reading text), the global knowledge (reader should combine his world knowledge with the text information to answer the question) and the overall comprehension of the reading passage. The advantage of this test format was that students could read the passage and comprehend it completely as there was no chance of guessing the correct answer without comprehending the passage. This was especially true for the global questions and overall comprehension questions in which there was no direct answer to be found in the passage. The readers should be able to process the text and infer the answers based on their own understanding.

Reading assessment was not a one-dimensional skill that could be evaluated by a single test format. Different test formats measured reading comprehension differently. It was recommended to use more than one test format to have a better understanding of the readers' comprehension ability. One of the most important issues in the field of reading

assessment that needed to be researched more in depth was applying different test formats to investigate the potential of each test method in assessing reading comprehension of the test takers (Alderson, 2000; Hudson, 2007; Ozuru et al., 2007). It was conceivable that different testing techniques permitted the measurement of different aspects of the reading construct. Therefore it was important to consider what techniques/test formats that were suitable for assessing, as well as what they might typically assess (Alderson, 2000). It was inadequate to measure the understanding of text with only one method by test takers. Different test methods should be applied to gauge students' reading performance based on their proficiency.

This paper described the use of two different test methods in measuring similar ability of the test takers. It might yield quite different results. An individual's performance on a language test may vary due to the influence of both his language ability and test method facets. In the other words, it was assumed that students at different level of proficiency performed differently across the two test formats. More specifically, this paper attempted to identify the differences in terms of students' reading performance in two test formats; summary writing and open-ended questions.

2. Literature Review

The theoretical assumption in this paper was based on two theories; facets theory and schema theory that were discussed in following section.

2.1 Theoretical framework

There were variations in theories such as facets theory and schema theory and their application to the EFL/ESL language reading comprehension are discussed in this research.

a. Facets Theory

Bachman and Palmer (1996) presented a theory of language testing that contained not only different aspects of language ability but also the methods and other factors involved in the measurement of this ability. Bachman (1996) presented a model of language ability which included test method facets in the discussion of language ability and drawn attention to a range of factors which could affect test performance. Test method facets were divided into five categories that could affect test performance (Bachman, 1996). They were (a) testing environment (b) test rubrics (c) the nature of the input (d) the nature of the expected response and (e) the interaction between the input and the response.

It was important for test takers to be aware of their influences and, if possible, minimize them. This study focused on the third and fourth categories of these facets. The nature of the input (the materials presented to test takers) was chosen as the main variable of this study because the choice of materials was very important factor in reading comprehension tests. Background knowledge, for example, was a well-researched area.

Following Bachman's (1990, 1995) language ability model, Kobayashi (2004) argued that there was a systematic relationship between the students' test performance and the effects of the nature of the input and expected response while the nature of input in testing referred to test organization and test method characteristics.

Bachman and Palmer (1996, p. 62) posited the importance of method facets, which they now term as task characteristics, as follows:

"...Language use involves complex and multiple interactions among the various individual characteristics of language users, on the one hand, and between these characteristics and the characteristics of the language use or testing situation, on the other. Because of the complexity of these interactions, we believe that language ability must be considered within an interactional framework of language use".

Facets theory or test method facets were represented to find out whether students' performance on language test, could be affected by facts such as familiarity of the place, requirement time of testing, physical condition and also individual attributes such as age, sex, educational background, general levels of language ability. Therefore using test method facets could help testers to find out the relationship between learners' characteristics and their performance. Test constructors would be able to make inferences about test takers ability to use language to

perform tasks in a particular domain. This theory could be useful to help test constructors to decide which type of test to use (Bachman & Palmer, 1996).

It was extremely important for language test constructors, or anyone who made judgments on the basis of test results, pay attention to the test methods used when they produced their assessment instruments or interpret test scores.

b. Schema Theory

Basically, schema theory originated in cognitive psychology but it had its applications in a majority of educational fields. The term "schema" (plural form is schemata or schemas) could be traced back to Plato who proposed the invisible ideal types of knowledge existing in the mind. Kant (1781, cited in Nassaji, 2007) was the first man to use the term "schema" in the literature. According to Kant (1781 cited in Nassaji, 2007) schema theory was the background knowledge for reading comprehension. When subjects read a story with an unfamiliar topic, they would modify the original version of the story according to their previous knowledge (Kant, 1781). Many researchers had studied the influence of background knowledge and the organization of texts on reading comprehension from psychology (Rumelhart & Ortony, 1977; Spiro, 1977; Rumelhart, 1980, 1984), psycholinguistics (Goodman, 1967), linguistics (Fillmore, 1982; Chafe, 1977a, 1977b; Tannen, 1978, 1979) and artificial intelligence (Schank & Abelson, 1975).

Rumelhart (1980) gave a definition of schemata as data structures representing concepts stored in memory. Schemata were fundamental elements on which information processing depended (Rumelhart, 1975, cited in Hudson, 2007). A schema was generalized knowledge about a sequence of events (Rumelhart, 1977). Pearson and Spiro (1982) posited schema as an abstraction of an experience that one was constantly fine-tuning and restructuring according to new information one receives. Reid (1993) indicated that a schema was the previously acquired background knowledge structures.

Nassaji (2007) argued that among the information that was embedded in a reading passage, only information that had relationship to the reader's background knowledge could be stored and then retrieved. In other words, if the reader possessed some topical knowledge related to the text that test taker was reading, those new information from the text would hang on to the already existing pegs in the reader's mind. They could then be memorized and recalled faster than the information without previous knowledge (schema) for them. Since ESL readers encoded meaning in ways that were different from native English speakers, Carrell (1984) suggested that teaching ESL readers the text structures of academic prose facilitated reading comprehension. When form and content were familiar, reading and writing were relatively easy. But when one or both were unfamiliar, efficiency, effectiveness and success were problematic.

Rumelhart (1980, p. 33-58) listed the characteristics of schema theory in reading as (a) skimming---get general ideas of text called gist (b) scanning---look for specific items (c) Emphasizing ideas and generalizations rather than isolated facts (d) summarizing and interpreting (e) global idea, done to get general idea of each paragraph (f) topic sentences in correct order (g) coping techniques to extract meaning (h) advance organizers (i) encourage students not to translate or to get hung up in individual words. A reader first skimmed the reading passage to get an overall idea of the whole text, and then scanned the passage looking for more specific details and ideas in the text. At the next stage, the reader tried to extract the most important ideas from the text and trimmed the unnecessary information from the reading passage. Then test taker was able to summarize the text into a short passage that were contained all important and necessary information. Finally, the reader used advance organizers to extract meaning from the reading passage by using his background knowledge to unfold the new information in his mind (Rumelhart, 1980).

2.2 Testing in reading skills

Reading was probably one of the most challenging skills to test. Testing reading had difficulties as there are many issues that test constructers had to take into consideration. Test methods which measured student' reading performance on a language test, may vary due to the influence of both his language ability and test method. One of the factors influencing test performance was test method facets. According to Bachman (1990), test method was related to the characteristics of methods used to elicit test performance. He believed that the methods researchers used to assess test-takers might yield quite different results although they intended to measure the same ability (Bachman, 1990). Following Bachman's (1990) language ability model, test developers could better understand

specific characteristics could be varied and used as a valuable tool for specific groups to perform at their best and affected students test performance. The background knowledge as one of the crucial issues in testing reading played an important role in reading comprehension. Therefore when a reader read the passage he was able to comprehend it more easily. A reader was also able to make more accurate inferences in passage/s from his background knowledge. On the other hand, testing in reading skill may look easy at first glance, but the possible difficulties should not be ignored such as the test methods should be carefully considered.

2.2.1 Open-ended Questions

Open-ended questions were a kind of constructed-response test method and they were supposed to involve higher order thinking skill as test taker needed to construct the response in his own word (Zheng, Cheng & Klinger, 2007). In open-ended questions, respondents needed to extract their background knowledge to integrate with the information in the text. The background knowledge could be used to answer the inferential and application questions. Majority of open-ended questions were inferential and application types as these kind of questions checked the test takers' comprehension of the text. However, literal questions were also used to deal with the lower order thinking skills.

2.2.2 Summary Writing

Summarizing helped readers to focus on main ideas or other key concepts. It was a complex activity that required the student to paraphrase and reorganize text. It also encouraged a deeper engagement with a text through rereading as students construct a summary (Kamil, 2004).

According to facets theory or test method facets from Bachman and Palmer (1996), paper aimed to examine whether students' performance on language test could be affected by test methods. Therefore using test method facets could help/enable testers to find out the relationship between learners' characteristics and their performance to make inferences about test takers ability to use language to perform tasks in a particular domain. This theory could be useful to help testers to decide which type of test to use (Bachman & Palmer, 1996).

3. Methodology

In this paper, a quantitative research design was used. Scores of the two test formats, summary writing and open-ended questions were assessed by two raters to ensure the reliability of scores.

3.1 Nature of the participants

There were 35 postgraduate students who participated in the present research. This study was carried out at Universiti Kebangsaan Malaysia (UKM).

3.2 Instruments

Three instruments were used to collect data for the research which investigated students' reading performance based on two test formats, summary writing and open-ended questions.

3.2.1 Reading Component of TOEFL Test

In this research a standard reading proficiency test; the reading component of test of English as a Foreign Language (TOEFL) was applied to identify the students' level of proficiency on the two test formats, summary writing and open-ended questions. This TOEFL test included 5 short reading passages that each of them consists of 10 Multiple-choice questions tests (MCQ). This test was carried out as first phase of this research.

3.2.2 Expository Reading Passage for two test formats, summary writing and open-ended questions

In this study, an expository reading passage was used for the two test formats, summary writing and open-ended questions. The title of the passage was "Shyness". In order to assess the suitability of the reading passage, the chosen passage was subjected to a readability index. For this passage, the Flesch Reading Ease scored was 46.2 and the number of words was 928. In assessing reading, the cut-off score for the Flesch Reading Ease was below 60 to be appropriate for the university students (Meyer, 2003). Summary writing test was done in the second phase and open-ended questions test was carried out in the third phase of this study.

4. Data Analysis

In this research, the TOEFL, summary writing and open-ended questions scores was analyzed using descriptive and inferential statistics by using SPSS version 18. For the inferential statistics, a paired samples T-test was used.

4.1 Results

The findings of the reading component of TOEFL test were analyzed by using descriptive statistics to gauge the students' level of proficiency. The scores of the two test methods were also analyzed descriptively.

Based on the previous researches in this field such as Meyer (2003) and Ozuru et al (2007) the score interval 80100 was considered high-achievers, the interval score 60-79 was categorized as intermediate-achievers, and the interval score 40-59 was ranked as low-achievers. Table 1 showed that there were 23 (65.71%) students who were placed in the Intermediate-achievers group of proficiency while only 12 (34.28%) students were ranked as low-achievers in the sample of the present study. Thus, most of the respondents belonged to the intermediate-achievers based on reading component of TOEFL test. Table 1 illustrated the classification of students according to their level of reading proficiency.

Table 1: Classification of students according to their reading proficiency

(TOEFL score)

Groups of proficiency TOEFL scores N %

Intermediate-achievers 60-79 23 65.71%

Low-achievers 40-59 12 34.28%

Overall 35 100.00%

Based on Table 2, between summary writing and open-ended questions, the results showed that students obtained higher mean scores in summary writing (M=71.42; SD=15.90) in comparison to open-ended questions (M=56.48; SD=15.26). This showed that students were able to perform better in summary writing compared to open-ended questions. The students faced more challenges in open-ended questions as they were not familiar with the question types (e.g. literal, application and evaluation). Grabe and Stoller (2002) stated that majority of the students were familiar with answering the literal questions in which the answer directly could be found in the text. They also were not familiar with other types of questions such as inferential and application questions. As a result, they failed to answer these types of questions correctly.

In addition, the results demonstrated that the performance of the intermediate achievers in summary writing (M=74.96; SD=15.14) was higher than their performance in open-ended questions (M=59.34; SD= 15.40). Similarly, the results showed that the performance of low achievers in summary writing (M= 64.66; SD=15.70) was also higher than open-ended questions (M=51.00; SD= 14.00). The difference between the mean scores for intermediate-achievers (74.96) and low-achievers (64.66) was 10 point. For the open-ended questions, there was a difference of 8 point between the mean scores for intermediate-achievers (59.34) and low-achievers (51.00). The findings illustrated that students could perform better in summary writing than open-ended questions as illustrated in Table 2.

Table 2: Students' performance on two test formats

Test methods Reading proficiency Mean Std. Deviation N

Summary- Intermediate-achievers 74.96 15.14 23

writing Low-achievers 64.66 15.70 12

1Overall 71.42 15.90 35

Open-ended Intermediate-achievers 59.34 15.40 23

questions Low-achievers 51.00 14.00 12

Overall 56.48 15.26 35

A paired samples t-test was conducted to evaluate the impact of two test formats on students' scores in summary writing and open-ended questions test. There was a statistically significant difference in the two test formats of summary writing (M= 71.42, SD= 15.90) to open-ended questions (M= 56.48, SD= 15.26), t (35) = 5.78, p<0.005 (two-tailed). The mean difference in the test formats was 14.94 with a 95% confidence interval ranging from 9.69 to 20.19. The eta squared statistics (.49) indicated a large effect size. Thus, the findings showed that students performed better in summary writing than open-ended questions. It was revealed that the students were more capable of answering the literal questions in which their answers could be found in the reading passage. In addition, the test takers were less capable of answering inferential questions corresponding to the reading passage. Based on mean scores in summary writing (71.42%) and mean scores in open-ended questions test (56.48%) the students performed better in summary writing comprised to open-ended questions test. The researcher found a significant difference between the two test methods. Table 3 illustrated the results of the findings.

Table 3: Paired sample t-test for two test methods Paired differences

95% confidence Interval of the Difference

Mean Std. Error lower upper t df Sig.

_Mean_(2-tailed)

Two Test

methods 14.94 2.58 9.69 20.19 5.78 34 .000

*a=.05 (level of significance)

The findings of this paper were in line with some of the previous studies such as Sawaki (2003), Vongpumivitch (2004) and Yu (2008) who found significant difference between these two test methods. These findings indicated that there was the potential difference in test taker performance in different test formats. The students got better scores for summary writing than answering the open-ended questions. The results reported that students faced more challenges to do summary writing than to open-ended questions. Based on the students' results, summary writing was considered to be easy while in open-ended questions. The students were unfamiliar to the different type of questions (Sawaki, 2003; Vongpumivitch, 2004, Yu, 2008).

Moreover, the findings were in line with Yu's (2007) and Sawaki's (2003) studies whereby test takers performed better in summary writing test. Students produced good quality summaries regardless of the language they had to

use in summary and manage to get the main idea/s from the passage. As a result, students achieved higher scores in summary writing test compared to open-ended questions test.

5. Conclusion

In this paper the two groups of proficiency, intermediate-achievers and low-achievers had better performance in summary writing while the same two groups had a poor performance in open-ended questions. The results showed that a significant difference was found between the test takers' performance in summary writing and open-ended questions. It was revealed that test takers produced better summaries compared to their performance in open-ended questions whereby they had to answer the literal and inferential questions. It could be interpreted that task difficulty on the part of the test takers was not affected by their level of proficiency. It meant that if a test method was difficult to complete, it was difficult for every individual student regardless of their level of proficiency. Based on two test formats scores, the students had achieved higher scores in summary writing compared to open-ended questions tests.

This paper was an investigation of the variances in students' performance in different test formats. The results of this study would be useful for assessing reading performance in the field of measurement. A suitable assessment tool could be considered as an opportunity for students to assess their reading performance. Students could demonstrate their language abilities as a rich view of progress, achievement, effort and strategies in reading assessment.

In addition language testers and second language researches made judgments on the basis of test results wherever there was a systematic relationship between the students' reading performance and two test formats. They tried to pay attention to the test methods used when they produced their assessment instruments or interpreted test scores.

This paper also had clearly highlighted that test constructers and classroom teachers would benefit from the present research in a way that teachers would know about the two test formats (summary writing and open-ended questions) usefulness in assessing reading ability of their students in the reading classes. It would be informative for the test constructers and textbook developers. They would know if these two test formats measure students' reading ability equally or differently in reading classes at schools. Hence, the most suitable test format would be used in testing reading formally and informally. Test results may be used to decide whether a student should be admitted to university, whether a prospective employee should be hired, or whether a project should continue or not. As Bachman and Palmer (1996) believed language testing could enable the constructor to make inferences about test takers ability to use language to perform tasks in a particular domain. The results of this paper could highlight some implications for assessment of English for Academic Purposes (EAP) and in English for Specific Purposes (ESP) especially in the domains of international politics, political news and literary translations (Bachman & Palmer, 1996).

References

Alderson, J. C. (2000). Assessing Reading. Cambridge, UK: Cambridg University Press. Alderson, J. C. (1990). Testing reading comprehension skills. Reading in a Foreign Language 6(2), 425-438 Bachman, L. F. (1990). Fundamental Considerations in Language Testing. Oxford University Press. Bachman, L. F. and Palmer, A. S. (1996). Language Testing in Practice. Oxford: Oxford University Press. Carrell, P. L. (1984). The effects of rhetorical organization on ESL readers. TESOL Quarterly. 18(2), 441-469. Chafe, W. L. (1977a). The recall and verbalization of past experience. In R. W. Cole (Eds.), Current Issues in

Linguistic Theory Bloomington. IndianaUniversity Press. Chafe, W. L. (1977b). Creativity in verbalization and its implication for the nature of stored knowledge. In R. O. Freedle (Eds.)

Cohen, A. D. (1993). The role of instructions in testing summarizing ability. In D. Douglas & C. Chapelle (Eds.),

A new decade of language testing research (pp. 132-160). Arlington, VA: TESOL Fillmore, C. J. (1982). An alternative to checklist theories of meaning. Proceedings of the First Annual Meeting of the Berkeley Linguistics Society University of California.

Goodman, K.S. (1967). The psycholinguistic nature of reading process, in Goodman, K.s. (Eds.), the

Psycholinguistic Nature of Reading Process (pp. 15-26). Detroit, MI: Wayne State University Press.

Grabe, W. & Stoller, F. (2002). Teaching and Researching Reading. New York: Longman.

Hedge, T. (1988). Writing. Oxford: Oxford University Press.

Hedge, T. (1998). Managing developmental evaluation activities in teacher education in P.Rea-Dickins and K. Germaaine (Eds.), Managing Evaluation and Innovation in Language Teaching (pp. 132- 58). Harlow: Addison-Wesley Longman.

Hudson, T. (2007). Teaching Second Language Reading. Oxford: Oxford University Press.

Kamil, M. L. (2004), September. Comprehension: What do we know from research? PowerPoint presentation, PREL Focus on Comprehension Forum, New York.

Kant, I. (1781). Critique of Pure reason. Translated to English by N. K. Smith 1963, Macmillan Publishing Company.

Meyer, B. J. F. (2003). Text coherence and readability. Top Lang Disorders. 23(3), 204-224.

Nassaji, H. (2007). Schema theory and knowledge-based processes in second language reading comprehension: a need for alternative perspectives. Language Learning, 57(1), 79-113.

Ozuru, Y., Best, R., Bell, C., Witherspoon, A., & McNamara, D.S. (2007). Influence of Question Format and Text Availability on the Assessment of Expository Text Comprehension. Cognition and Instruction, 25(4), 399- 438.

Pearson, P.D., & Spiro, R. (1982). The new buzz word in reading is schema. Instructor, 46-48.

Pearson, P.D. & Hamm, D.N. (2005). The assessment of reading comprehension: A review of the practices- past, present, and future (pp. 13-69). In S.G. Paris, & S.A. Stahl (Eds.), Children's reading comprehension and assessment. Mahwah, NJ: Lawrence Erlbaum Associates Publishers.

Reid, J. (1993). Reading and spoken language: the nature of the links. In R. Beard (Eds.), Teaching Literacy: Balancing Perspectives. London, Hodder and Stoughton.

Rumelhart, D. E. (1984). Understanding. In J. Flood (Eds.), Understanding Reading Comprehension. Newark, Delaware: International reading Association.

Rumelhart, D. E. (1980). On evaluating story grammars. Cognitive Science. 4(3), 313-316.

Rumelhart, D.E. (1977). Understanding and summarizing stories. In D. LaBerge, & S. J. Samuels (Eds.), Basic Processes in Reading: Perception and Comprehension (pp. 256- 303). Hillsdale, N.J: Lawrence Erlbaum Associates. .

Rumelhart, D. E. & Ortony, A. (1977). The representation of knowledge in memory. In R. C. Anderson, & R. J. Spiro (Eds.), 8tW E. Montague.

Sawaki, Y. (2003). A Comparison of Summarization and Free Recall as Reading Comprehension Tasks in Web-Based Assessment of Japanese as a Foreign Language. Unpublished PhD dissertation, University of California, Los Angeles.

Schank, R. C. & Abelson, R. (1975).Scripts, Plans, Goals and Understanding Hillsdale, N.J.: Lawrence Erlbaum

Spiro, R.J. (1977).Theoretical Issues in Reading Comprehension. Hillsdale, NJ: Erlbaum.

Vongpumivitch, V. (2004). Measuring the knowledge of text structure in academic English as a Second language (ESL). Unpublished PhD dissertation, University of California, Los Angeles (UCLA).

Yu, G. (2008). Reading to summarize in English and Chinese: A tale of two languages?Language Testing, 25(4), 521-551.

Zheng, Y., Cheng, L, & Klinger, D. (2007). Do test formats in reading comprehension affect English as a second language students' test performance differently? TESL Canada Journal, 25, 65-80.