토익 시험에서의 한국 대학생들의 문항 유형 및 난이도에 따른 독해전략

Abstract

The purpose of this study was to explore Korean university students’ use of reading strategies by item types and item difficulty on the TOEIC. The participants were 6 university students taking a TOEIC class at a university near Busan. To better obtain participants’ use of reading strategies in the process of understanding texts and reaching right answers to the test items in the TOEIC, a think-aloud method, in-depth interviews, and observations were used. Qualitative data was first transcribed, and then analyzed by means of line by line coding using NVivo 12 in order to explore the strategies utilized by the participants. The differences in reading strategies used were then compared and analyzed according to the participants’ proficiency levels, item types, and item difficulty. The results were as follows: first, participants in this study were seen using a variety of strategies to comprehend texts, including general strategies, local strategies, meta-cognitive strategies and other strategies. Participants tended to use local strategies that focused on the language of the text a little more frequently than general strategies focusing on understanding the overall text, while the use of other strategies and meta-cognitive strategies was quite limited compared to local and general strategies. An analysis of the top six reading strategies used to reach answers to test items revealed that participants had a strong tendency to select an answer based on their understanding of the text or evidence found in the text, regardless of item type. Second, all of the reading strategies used to take the TOEIC were reanalyzed by item type, including factual information items, inferencing items, and insertion items. The results show that it is effective to go back to the text and find evidence for a judgment, or for test-takers to take advantage of their textual understanding in the case of factual information items and inferencing items. This supports the results of prior research by Anderson et al. (1991) and Lee and Gu (2005). However, in the case of inferencing items, excessive reliance on the text may lead to greater difficulties due to the characteristics of the item type in which the basis of the correct answer is not clearly revealed in the text, making it necessary to compensate for the lack of clear evidence in the text by using various strategies, such as utilizing learners’ background knowledge, or choosing the correct answer by process of eliminating all of the incorrect options. An examination of reading strategies frequently used for the insertion items showed that all strategies were used only in the insertion items, resulting in the use of strategies different from factual information items and inferencing items. This can be seen as different from the results of Cohen and Upton (2006) which showed that certain item types do not induce the use of specific strategies. In addition, it was found that understanding the overall text and understanding the content of the insertion sentence were prerequisites for reaching correct answers to the insertion items. In other words, without participants' understanding of the overall text and content of the insertion sentences, it can be difficult to expect the effects of any strategy used. Third, all of the reading strategies used to take the TOEIC were reanalyzed, according to participants’ proficiency levels and item difficulty. As a result, it was shown that there were differences in the use of the strategies utilized to understand a text and reach answers by participants’ proficiency levels. In the process of understanding a text, low proficiency groups used strategies the most, and high proficiency groups used them the least. It was supposed that high proficiency students do not need to use strategies in order to understand a text, while low proficiency students have a greater need for using strategies to overcome difficulty caused by their limited language knowledge, which is consistent with the findings in Oxford et al. (2004). In addition, looking at the kinds of strategies used by the proficiency groups, it was shown that high proficiency students preferred to use general strategies and meta-cognitive strategies, while middle and low proficiency students preferred local strategies. This study identified differences in the use of strategies based on participants' language proficiency, which was in line with several previous studies (Block, 1986; Kim, 2019; Phakiti, 2003; Shin, 2003). In the process of choosing answers, high proficiency students used strategies most frequently, followed by the mid proficiency and the low proficiency groups. It was found that high proficiency students tended to recheck their judgment even when they were sure about the answer. This tendency of high proficiency groups, which was distinct from the other groups, can lead to frequent use of strategies in choosing answers. Furthermore, all of the reading strategies were analyzed according to objective difficulty and subjective difficulty. As a result, it was found that objective difficulty differed from the actual difficulty felt by participants. However, based on subjective difficulty, it was confirmed that there was a difference in the use of strategies depending on the level of difficulty of the items felt by participants. It was shown that all of the participants increased the frequency and kinds of strategy use in the items that they felt were challenging. In other words, all groups can control their strategy use depending on the level of perceived difficulty. This can be seen as different from the results of Lim Hye-jung (2009) who showed that only high proficiency students are proficient in controlling their strategy use based on difficulty. However, in the case of items that were too difficult to find the correct answers, participants tended to give up without making any effort. Comprehensive results of qualitative analyses in this study showed that differences in strategy use could be attributed to by learners’ proficiency levels and the levels of difficulty that learners feel when understanding a text and reaching correct answers. Pedagogical implications and further suggestions for future studies are provided.