圖片對於國中生第二外語字彙習得及故事書閱讀理解之影響

全文

(1)國立臺灣師範大學英語學系碩士論文 Master’s Thesis Department of English National Taiwan Normal University. 圖片對於國中生第二外語字彙習得及故事書閱讀理解之影響 Effects of Pictures on the Incidental Vocabulary Acquisition and Comprehension from Story Book Reading by EFL Junior High Students in Taiwan. 指導教授：朱錫琴 Advisor: Dr. Hsi-chin Chu 研究生：王孟勤. 中華民國一百零四年六月 June 2015.

(2) 摘要本研究旨在探究台灣國中學生在閱讀圖畫故事書時, 圖片對於單字習得及閱讀測驗的效果。圖畫書因為有提供圖片，也許能提供英文初學者額外的資料來推測單字及理解文本內容。本實驗中，六十四位台北市立國中的八年級生在兩個連續的星期內閱讀兩本故事書，而兩本書以兩種模式呈現: 一本有圖片，另一本則無，同時也平衡兩本書的閱讀順序及呈現模式。每讀完一本書，學生們就接受立即的單字選擇題考試以及是非題的閱讀測驗。在實驗結束兩個星期後，學生們接受同一組單字的單字後測，並且填寫關於讀者感受的問卷。 T 檢定的結果顯示，整體而言，提供圖片的模式確實能促進立即性以及兩個星期後的單字的辨識。此外，圖片的提供也可以促進閱讀理解。然而，進一步個別分析兩本故事書，發現圖片的功用只於其中一本關於露營的故事書較顯著; 另一本關於棒球的故事書中，圖片的有無並沒有在單字考試, 閱讀測驗及單字後測上達到顯著的效果。藉由此結果可以推論：關於棒球這本故事書，因為棒球這個題材較為學生熟知且故事內容較淺顯，學生較不需要倚賴圖片。因此，書本題材的不同或許會平緩圖片對國中生在單字認識及閱讀理解的效果。最後，本論文提出對於以圖畫書為閱讀教材的教學建議以及未來相關實驗的建言。. 關鍵字: 字彙習得，閱讀理解，圖畫故事書，讀者感受，外語學習，國中. i.

(3) ABSTRACT This study aims to examine the effects of pictures provided in story reading on incidental vocabulary acquisition and reading comprehension by junior high school EFL readers in Taiwan. Picture books, with the supply of pictures, may provide English beginning readers additional input for meaning making of words and texts. Sixty-four eighth graders in a Taipei municipal junior high school read two story books in two presentation modes, one with pictures and another without, in consecutive two weeks. The presentation sequence of the two topics and the two presentation modes was counterbalanced. After each reading, students took a vocabulary Multiple Choice test and a True-or-False reading comprehension test. Two weeks after each reading session, students took delayed vocabulary tests on the same set of words and filled out a perception questionnaire. T-test results indicate that, overall, supplying pictures for story reading did promote vocabulary acquisition immediately and two weeks after story reading. Moreover, the provision of pictures also facilitates the comprehension in general. However, further examination into the effect of picture on each of the two topics revealed that the picture effects were restricted to one topic, Camping, only; the other topic, Baseball, did not show effects on immediate vocabulary acquisition, delayed vocabulary acquisition, or reading comprehension. It was speculated that Baseball might be more familiar and easier for the students, and thus minimizing students’ resort to pictures. Hence, topic might mediate the effects of picture on the vocabulary acquisition and comprehension from story reading by junior high students. Suggestions for pedagogical practice and for future studies were therefore made.. Key words: incidental vocabulary acquisition, reading comprehension, picture books, reader perception, EFL, junior high students ii.

(4) ACKNOWLEDGEMENTS Getting the master degree was such a tedious and thorny journey; fortunately, because of some precious help, I was lucky to fulfill the mission I had promised myself. First of all, I would like to give all the credits to my professor, Prof. Hsi-chin Chu; she was always there, giving me the best suggestion and the timely help. Without her, I would not have accomplished so much on the way to my destination. I would also like to thank Prof. Wen-ta Tseng and Prof. Chin-kuei Cheng for giving me the advice on revision for my research questions, testing instruments and wording, which made the thesis more mature. I especially appreciate the people who made the two picture books I used in the research; without them, the research would not have come into reality: Curious George Goes Camping, written by Margret and H.A. Rey, illustrated by Vipah Interactive; Curious George at the Baseball Game, written by Laura Driscoll, Margret and H.A. Rey, illustrated by Anna G. Hines. My thanks also go to the publisher and the copyright holder: Houghton Mifflin Harcourt Company for their permission about the reprinting and use of the two picture books. Second, I would like to contribute what I accomplished to my dear classmates for they shared their valuable experiences and ideas with me unselfishly all the time. I would also like to thank my friend, Alice, for her warm encouragement. I am especially thankful to my colleague, Brenda, for helping me out when I felt fretful. Also, I am grateful for Sunny's assistance at the final moment. Last but not the least, there are so many lovely people in my life that I definitely need to mention: my dearest dad and mom, my sister and my husband. They give me the greatest support and care. My babies, Lawrence and Hannah, are both my spiritual comfort during the hard process and the biggest treasure in my life. I love these people so much and I am blessed to have them around me. iii.

(5) TABLE OF CONTENTS CHINESE ABSTRACT............................................................................................i ENGLISH ABSTRACT...........................................................................................ii ACKNOWLEDGEMENTS.....................................................................................iii TABLE OF CONTENTS........................................................................................vii LIST OF TABLES..................................................................................................viii. CHAPTER ONE INTRODUCTION........................................................................1 Background and Motivation......................................................................................1 Purpose of the Study..................................................................................................4 Significance of the Study...........................................................................................4 Definition of Terms....................................................................................................4. CHAPTER TWO LITERATURE REVIEW..............................................................6 Theoretical Background.............................................................................................6 Dual Coding Theory.............................................................................................6 Comprehensible Input Hypothesis and Incidental Learning................................7 Empirical Studies........................................................................................................8 Studies on Visual and Vocabulary Acquisition.....................................................8 L1 Studies on story reading, Vocabulary and Reading Comprehension.............10 L2 Studies on Incidental Learning of Words through Reading...........................12 Studies on Picture Cues and Reading Comprehension........................................15 Picture Books for Reading Comprehension and Vocabulary Acquisition............17. CHAPTER THREE METHODOLOGY....................................................................22 iv.

(6) Pilot Study..................................................................................................................22 Participants...........................................................................................................22 Reading Material..................................................................................................22 Purposes of the Pilot Study...................................................................................23 Results of the Pilot Study......................................................................................23 Main Study..................................................................................................................24 Participants............................................................................................................25 Instruments............................................................................................................25 Reading Materials.............................................................................................25 Word Selection..................................................................................................26 Vocabulary Test Type and Revision..................................................................27 Reading Comprehension Test...........................................................................29 Questionnaire....................................................................................................30 Study Design and Data Collection Procedures......................................................30 Scoring Data...........................................................................................................31 Scoring Vocabulary Test....................................................................................31 Scoring Reading Comprehension Test..............................................................32 Coding Questionnaire........................................................................................32 Data Analysis..........................................................................................................32 Quantitative Analysis........................................................................................32 Qualitative Analysis..........................................................................................33. CHAPTER FOUR FINDINGS....................................................................................34 Effects on Vocabulary Acquisition...............................................................................34 Effects of Reading on Immediate Vocabulary Acquisition...................................34 v.

(7) Effects on Delayed Vocabulary Acquisition.......................................................37 Effects on Reading Comprehension...........................................................................38 Exploratory Analysis..................................................................................................40 Findings from the Questionnaire................................................................................42 Reading Habits and Vocabulary Strategies.........................................................42 Reader Perceptions of the Reading Materials.....................................................44 The Importance of Pictures.................................................................................46 The Future Pedagogical Use of Picture Books....................................................48 Summary of the Findings............................................................................................49. CHAPTER FIVE DISCUSSIONS AND IMPLICATIONS.......................................50 Effects of Story Reading on Vocabulary Acquisition.................................................50 Effects of Pictures on Vocabulary Acquisition...........................................................51 Effects of Pictures on Reading Comprehension.........................................................53 Varied Effects of Pictures for Topics..........................................................................55 Relevance of Pictures to Target Words.............................................................56 Readers' Perception of Story Content...............................................................57 Implications.................................................................................................................58 Implications for Pedagogical Use.....................................................................58 Implications for Future Research......................................................................59 Conclusion....................................................................................................................60. REFERENCES.............................................................................................................62. APPENDICES..............................................................................................................68 vi.

(8) A: Vocabulary Test from Nation: 1000 Level Test.........................................68 B: Sample Pages from Picture Book and a Text for Text Only Version (Camping) ...................................................................................................72 C: Pilot Multiple Choice Test........................................................................76 D: Pilot Chinese Equivalent Production Test................................................78 E: Finalized Multiple Choice Test.................................................................80 F: Reading Comprehension True or False Test..............................................82 G: Questionnaire ...........................................................................................84 H: Questionnaire for Topics...........................................................................86 I: Consent Form............................................................................................88. vii.

(9) LIST OF TABLES Table 1 Target Words after Word Selection. 27. Table 2 Data Collection Procedures. 31. Table 3 Mean Scores, Standard Deviations and Mean Percentage of Immediate. 35. Vocabulary Tests Table 4 T Test Results for Three Tests. 36. Table 5 Mean Scores, Standard Deviations and Mean Percentage of Delayed. 37. Vocabulary Tests Table 6 Mean Scores, Standard Deviations and Mean Percentage of Reading. 39. Comprehension Tests Table 7 Mean Scores and Standard Deviations of Perception of the Two Stories. 41. Table 8 Paired-T Test Results of Perception of the Two Stories. 41. Table 9 Students' Reading Habits and Vocabulary Strategies. 43. Table 10 Frequency of the Rating on the Difficulty Level of the Two Reading. 44. Modes Table 11 Frequency and the Rating on Source of Difficulty of Reading Materials 45. viii.

(10) CHAPTER 1 INTRODUCTION. Background and Motivation For junior high school teachers, it is undeniable that vocabulary plays an essential role when it comes to reading comprehension, which, of course, also affects junior high school students’ academic performance and future career in Taiwan. However, when it comes to the issue about how to teach vocabulary, most of teachers teach vocabulary according to the arrangement of the textbook format. For some teachers, when there is a word list, they would start from the word list. For others, they teach dialogues and short reading passages first and then come to the vocabulary section finally. Some teachers might type some self-made materials for students because of the limited information in their textbooks. It seems that after several years’ teaching, all teachers might end up teaching in the same routine and will be forced to face the same question: is there any other approach for us? What can we do to enhance students’ learning interest? Teaching is an on-going process, and many teachers somehow stagnate at a point without any creativity and refinement. That is the reason for teachers to explore if there are alternative strategies for vocabulary instruction and vocabulary acquisition. Through taking courses in National Normal University, I have found issues on vocabulary acquisition interesting and fascinating. The most surprising thing was that some researchers estimated that a native English-as-first-language high school student needs to know about 75000 words in English and obviously, it is not possible for formal school instruction to cover such large amount (Snow & Kim, 2007). To support school instruction, incidental learning can be a supplement for students’ vocabulary acquisition (Krashen, 1989; Nagy, Anderson, & Herman, 1985, 1987; -1-.

(11) Nation, 2001; Snow & Kim, 2007). But for junior high school students in Taiwan, their ability to acquire incidental learning seems to be ignored and lack evidence. This study was thus conducted to employ children’s picture books as a medium for incidental vocabulary acquisition and to investigate its effects on junior high school students’ reading comprehension, vocabulary acquisition, and response on such reading materials. At schools in Taiwan, junior high school students and teachers usually both face the same teaching routines; teachers try to cram students with a lot of vocabulary, but students sometimes fail to consolidate all the target words into their mind for the limited encounter chances and practices. Students might learn some words hard for a while, and after an exam, they forget them easily. Teachers are always hurried to finish the lessons, forgetting to help develop students’ automatic learning ability. School instruction and lesson hours seem inefficient. If junior high students are provided with picture books, perhaps they can also acquire vocabulary through the help from context and pictures as a visual aid, and the interesting and colorful illustrations might help to enhance their reading comprehension. That is, they might show a certain level of ability to acquire and learn incidentally, which is beyond the reach of formal school instruction and which is able to inspire them to reach more outside reading for English learning is such a long journey. Incidental learning has been proven effective in L1 reading (Krashen 1989; Nagy et al., 1985, 1987; Nation, 2001; Snow & Kim, 2007); and it has been proven to happen to L2 milieu as well (Pitt, White & Krashen, 1989; Day, Omura & Hiramastu, 1991; Hulstijn, Hollander & Greidanus, 1996; Watanabe, 1997: Horst, Cobb, Meara, 1998; Pigada & Schmiss, 2006; Brown, Waring & Donkaewbua, 2008; Sanchez & Schmitt, 2010). Incidental learning might be an optimal aid to L2 leaners’ vocabulary -2-.

(12) acquisition and be a complementary activity to enhance learning. However, so far, to our knowledge, few researches were conducted with picture books and performed on junior high school students, beginning young learners. And no direct researches have proven that junior high students in Taiwan can also achieve incidental vocabulary acquisition when they are given interesting reading materials with pictures as the extra facilitator. Picture books represent the whole human beings' reading experience since childhood. Undoubtedly, picture books are the most original and primary reading source during the early stage of a child’s learning life. That is also how everyone’s learning life begins: children and their parents read together, holding illustrated story books in hands or on laps. This kind of experience is, undeniably, universal, simple and pure, without too much demand on vocabulary. And somehow, children in L1 milieu learn words. Picture books, which are made for young readers originally, without being needed to be modified or revised, might also be a suitable reading material for Taiwanese junior high students to acquire vocabulary and achieve better reading comprehension. Yet, scarce studies on vocabulary acquisition used picture books as reading materials while pictures actually offer additional information, helping readers to infer meanings of vocabulary and to comprehend the storyline for their vividness and concreteness. Picture books might be an alternative learning medium, especially for junior high English learners. With the presence of pictures, beginning learners might form a concrete concept for the target words, achieve better reading comprehension, and have more interest in learning vocabulary. To our knowledge, picture books have been used to inspire university students In Taiwan, and increase learning interest in words (Sun, 2010; Lee, 2007). However, little evidence had been showed that picture -3-.

(13) books promote incidental vocabulary acquisition for younger EFL Learners. Therefore, this thesis intends to fill the void of such studies and aims to provide empirical data.. Purpose of the Study The purposes of the study were three-fold. First, the study aimed to investigate whether picture- plus- story or story- only can facilitate EFL students’ reading comprehension. Second, the immediate and long term effects of vocabulary acquisition were examined and compared between picture-plus-story and story-only situations. Third, this study intended to examine student’s perception of the two versions of reading experience, in terms of difficulty and their interest in English learning. Significance of the Study The study aims to provide empirical evidence for teachers to reconsider the value of picture books on vocabulary acquisition and reading comprehension. The result of the study can help teachers to clarify the role of picture books so as to provide teachers ways to improve reading and vocabulary instruction. If incidental learning indeed occurs from picture book reading, vocabulary might take care of itself and explicit instruction might not be necessary when reading picture books. Then, teaching material developers may incorporate picture books into the curriculum to enhance students’ learning enthusiasm, and bring more vitality to the teaching. Teachers might be able to release some control over their own instruction and students can use pictures as a learning tool of reading comprehension and vocabulary instruction. Definition of Terms Picture Books -4-.

(14) According to Bishop and Hickman (1992), picture books are written with a dual narrative, in which the pictures and the text cooperate closely to tell a story and present content (p.192). In this study, the picture books contain colorful illustrations, showing on each page. Incidental Learning The term "incidental learning" is the opposite of instructional and intentional learning (Nation, 2001; Brown et al., 2008). It means the learning of vocabulary through many daily activities such as outside reading, watching TV or surfing a website; what students learn is not from explicit school education, and their focus is not the vocabulary itself. Learning vocabulary is not the target, but the byproduct during the learning process (Hulstijn et al, 1996). Vocabulary Acquisition According to Nation (2001), knowing a word involves knowing its form, meaning and use. In this study, the definition of vocabulary acquisition is "reception." If participants could recognize the form of the target words and infer the L1 equivalents of the target words (namely, the meaning), they achieved the vocabulary acquisition. Reading Comprehension According to Tompkins (2011), reading comprehension is the ability to understand the knowledge written in the text, and a level of understanding coming from the interaction between the words written and the knowledge triggered outside the text (p.203). In this study, if the participants could understand the plots of the books and give the right answers to True or False questions in L1, then they could be defined as achieving reading comprehension.. -5-.

(15) CHAPTER 2 LITERATURE REVIEW. Theoretical Background This chapter will probe into this study in terms of two related theories: Dual Coding Theory and Comprehensible Input Hypothesis and in terms of empirical studies, hoping to provide the foundation of the present research. Dual Coding Theory Cognitive psychologists contend that human beings are prone to remember images better than words as reflected in "The Dual Coding Theory (DCT), proposed by Paivio (1986). DCT posits that human beings' cognitive processes involve two models: verbal and nonverbal. Verbal model stands for our language system, and nonverbal model means images or pictures, symbolic icons of referents. When the two models work together and interconnect with each other, human beings have more kinds of input resource to recognize and to trace back in memory. Therefore, language as the sole input may not be adequate when it comes to language teaching or learning. According to Paivio (1986), it is essential to learn the second language with appropriate nonverbal referents, while such referents like objects, events, behaviors or emotions, cognitively represent and constitute the knowledge of the world. If L2 learners are provided with more referential interconnections, they will be more equipped to use L2. According to DCT, creating images or pictures is beneficial for vocabulary acquisition and text retention (Paivio, 1986). Rather than having students imagine, the provision of pictures may enhance learners' understanding. Therefore, picture books imight have their values when it comes to L2 learning. The most important reason is -6-.

(16) that a picture book is actually a story composed by illustrations and texts together. When picture books combine two models together: verbal and nonverbal, they might provide students a more direct or stronger interconnection between language and image, which can enhance students' learning efficiency. However, so far, few studies between vocabulary learning and picture books have been done. Most DCT studies are about using the connection between pictures and vocabulary to enhance learning effect in content area (e.g. Mayer &Anderson, 1991, 1992) or implicated the theory as a kind of picture-word mnemonics (e.g. Ellis & Beaton, 1993). There was a huge gap here; picture books might be the medium for vocabulary learning for they provide pictures and words, two models at the same time. Yet few studies mentioned such functions of picture books. Comprehensible Input Hypothesis and Incidental Learning Krashen (1989, 1993) posited that as long as readers can understand reading materials, their reading ability and vocabulary will increase naturally through the power of reading. Even though there will be new words in the reading material, the readers will eventually pick up the meanings of the new words because their strong interest will push them keep on reading, rather than giving up. Namely, through comprehensible input, vocabulary can grow during the reading process as long as the reading material is interesting and motivating. Nagy, Herman, and Anderson (1985) claimed that word learning is an incremental and gradual process, and reading, undeniably, was the main resource for children to acquire vocabulary. If a child were to encounter 10000 unknown words a year, they can acquire about 500-1100 words for the ration to acquire a word through only once or a few exposures was from 0.05 to 0.11. Hence, they proposed that L1 words learned incidentally from contexts were likely to constitute a substantial -7-.

(17) proportion of children's yearly vocabulary growth. And that is the power of incidental learning for we can see that school instruction can not cover so much ground during the language learning process. Incidental learning is often contrasted with instructional and intentional learning (Nation, 1990; Brown et al, 2008). It often refers L1 children's rapid vocabulary growth through input such as outside reading, watching TV or surfing a website in addition to what they obtained from explicit school education. Learning vocabulary is not viewed as the target, by the byproduct during the learning process (Hulstijn et al., 1996). In view of this, L2 researchers also inquired whether or not, the incidental learning experience can be copied to L2. In Taiwan, an EFL milieu, students are trained hard to memorize vocabulary and take tests. However, they might still forget vocabulary no matter how many times these words were taught formally and consolidated through repeating drills and tests. It is noteworthy to investigate whether junior high school students in Taiwan, the beginning language learners, will be able to glean and retain the meaning of vocabulary by themselves and achieve incidental learning to some extent when they are provided with comprehensible input, that is, picture books, as the reading material. Since there had been no direct proof elicited from Taiwanese junior high schools students, we would need an empirical study done on these specific participants and make up the gap in the research field.. Empirical Studies Related empirical studies will be discussed and presented here to see what have been done and what is lacked so far, hoping to fill the gap with the present research. Studies on Visual and Vocabulary Acquisition -8-.

(18) According to the DCT, creating images or pictures is beneficial for vocabulary acquisition and retention. Ellis and Beaton (1993) investigated the effect of generating imagery of words: the keyword strategy to create image, which leads to better vocabulary retention. L2 learners of German were taught the keyword strategy and were found to recall and retain target words than those who were not. However, not every word had the same effect; they found nouns, in all the word classes, were more suitable and feasible to serve as the target words and keywords than other word classes because they contain the most concrete ideas and initiate the most direct connection between the meanings and the images. Oxford and Crookall (1990) also believed that combining pictures and vocabulary could enhance vocabulary learners' vocabulary learning, which has its pedagogical values. Chun and Plass (1996) found that words associating with actual objects and imaginary techniques were learned more easily. They provided university students with vocabulary annotation in the form of text, picture and video. The group with picture plus text could score higher than the other groups in the immediate and delayed vocabulary tests. Still pictures, or images were reported as the most used cues for learners to recall vocabulary. They concluded that pictures have a constant and fixed quality, which allows the learners to develop a mental model of information. Ghlaomi and Abdollahpour (2011) conducted a similar study. They found Iranian students performed much better in vocabulary tests when they were provided with pictorial vocabulary annotations. Compared with the group with word lists only, the group provided with pictures or images could recall vocabulary better because these pictures show students some visual representations and enhance their memory and vocabulary learning. In content-area studies, many researchers attested the DCT as well. For example, -9-.

(19) Mayer and Anderson (1991) once found that university students could understand how pumps worked much better when they were provided words-with-pictures explanation at the same time. The study also proved once again that when learners acquire a skill or some hard techniques, supplying a picture or an image to combine with the text, or the target word, can achieve a better outcome on learning. This is corroborated by Nuttall (1996), who suggested, "an obscure section of text may be clarified by a diagram, or the significance of a graph may become clear from the text. This can be exploited even in an early stage of reading: a learner stumbling over a word is urged to look at the pictures and work out what the word must be"(p53). Following these studies derived from the DCT, we can see that visual aids and pictures indeed help learners to learn better for they represent another channel of information and provide learners with one more model of helpful resource. However, we still lacked a direct study done on vocabulary acquisition for L2 beginning learners. Some studies indicated that pictures and images, as a kind of mnemonics, could help learners to understand or memorize words for older university learners, and some studies focused on content area, not language learning. That is the reason why we needed to testify if the use of pictures, as those in picture books, could have an effect on young L2 learners' vocabulary acquisition in the present research. L1 Studies on Story Reading, Vocabulary and Reading Comprehension According to Krashen (1989), as long as readers can process comprehensible materials, their reading ability and vocabulary will increase. Even though there might be new words during the reading process, the readers will eventually pick up meanings of the words. Competence in spelling and vocabulary might grow through comprehensible input in the form of reading. Elley (1989) asked teachers to read stories loud to elementary school children, - 10 -.

(20) and administered pretests and posttests to measure the extent of new vocabulary those children acquired from reading. Results showed that story reading contributed a lot to those children's vocabulary acquisition. Elley also found vocabulary exposure frequency, the amount of redundancy in the surrounding context and the depiction of the word in the illustration influenced a lot. As long as children liked the story, they obviously could learn some word knowledge from the reading. In 1991, Elley started a program called "Book Flood." In contrast to students under the serious instruction of audio-lingual program, the students who read an extensive range of illustrated story books were found to learn more words, achieve better over all English proficiency and have a more positive attitude towards English books. Elley's long-term studies supported Krashen's comprehensible input theory, but we did not see the detailed statistical data, and his researches were done in Englishas-the-official- language setting, which is quite different from Taiwan. However, despite these empirical evidences generated, a lot of researchers did not concur on the relation between reading and vocabulary acquisition. First, not all language learners are good at the meaning inferring (Jenkins, Matlock, & Slocum, 1989), and sometimes the contexts are not equally helpful, appropriate and effective for language learners to acquire vocabulary through reading (Beck, Mckeown, & McCaslin, 1983). Some researchers did not think reading only could enhance L2 learners' vocabulary acquisition (Mondria, 2003) and "just reading for fun" could not promise any learning effect; readers' reading purpose would influence the learning effect of a target word, especially for lower-achievers, who actually need placid, explicit and direct clues to learn a word (Swanborn & Glopper, 2002). These studies do not show agreement with Krashen's input hypothesis and again, we still lacked concrete data to make a judgment on Taiwanese junior high school students. - 11 -.

(21) After being taught English explicitly and instilled into a lot of vocabulary, Taiwanese junior high school students are usually forced to receive input from the classroom; school textbooks and teachers' instruction are usually their sole input, no matter they are comprehensible or not. We needed an empirical study to examine if Taiwanese beginning L2 learners could show vocabulary acquisition and reading comprehension, especially in the free reading atmosphere without pressure from tests and academic performance. L2 Studies on Incidental Learning of Words through Reading In a classical L2 study, Saragi, Nation and Meister (1978) had university students read a novel called "A Clockwork Orange" to check their gain on Russian words in the novel. They found that the participants could identify the meaning of the Russian words correctly in an unexpected multiple choice test. Following their study, many other studies on incidental learning were done with participants of different mother tongues, age groups, reading materials, input modes, reading tasks, test forms, aspects of word knowledge and time interval. The results varied, yet a different rate of success has been found. Pitts, White, and Krashen (1989) replicated the "Clockwork Orange" study and found ESL language learners could also acquire Russian words through reading the novel. One or two words could be acquired among every 30 words. The study showed small but significant gain for ESL language learners. Day, Omura, and Hiramastu (1991) asked Japanese high school and university students to read a shortened story, and checked how much they learned about 17 target words in a multiple choice test. As for the control group, they did not read the reading material in advance. Of course, there was great difference between treatment and control group, and the study proved that Japanese EFL learners had the ability to - 12 -.

(22) recognize the target words through reading the story; incidental learning could happen to EFL learners. Hulstijn, Hollander, and Greidanus (1996) had advanced students in Holland read a French article and checked their 16 target word acquisition. However, the 16 target words were grouped according to the exposure frequency. Eight of them showed up once, and the other eight showed up more than three times. The researchers also designed two treatment groups: the marginal gloss group and the dictionary use group. They found the exposure frequency was important for a word to be learned, and the marginal gloss group performed better than the dictionary use group. Besides the exposure to words in reading, the factor of exposure frequency might have an effect. Horst, Cobb and Meara (1998) found that exposure frequency had to be more than 8 times for a word to be learned, and they proposed that using a longer text as the reading material could help the readers to meet the target words more in their study. It seems that incidental learning indeed happens, but happens at a lower rate than the previous researches claimed. Therefore, some following studies focused on how to enhance the better effect of incidental learning. Pigada and Schmitt (2006) increased the reading time of participants to a month, that is, extensive reading, and believed that extensive reading could enhance the happening rate of incidental vocabulary acquisition. Brown, Waring, and Donkaewubua (2008) proposed that the target word showing more than 15 times was more likely to be learned; and adding one more input mode (reading and listening at the same time) could achieve vocabulary learning effect. Swanborn and Glopper (2002) believed that reading purpose was an essential factor deciding whether incidental learning happens or not; reading for fun does not lead to vocabulary acquisition. As for Webb (2008), he claimed that the quality of context would - 13 -.

(23) influence if a target word could be leaned or not. Some contexts were just not informative enough for learners to acquire a new word. Then, Sanchez and Schmitt (2010) double checked the cutting point of exposure frequency for a target word to be learned: 8 times above, and they found that actually word form (spelling) seemed to be harder for participants to learn than word meaning recognition. From these studies, several implications are shown. First, incidental learning of vocabulary indeed happens to L2 learners, but the rate varies a lot. Most of incidental studies focused on university or high school students, who actually have more mature cognitive ability than junior high students. Hence, we could not be sure if the same successful experience of incidental learning could occur to junior high students as well when they are just beginning L2 learners. Second, to enhance the learning effect of incidental learning, most studies prolonged the reading time to months. Few studies focused on increasing the input resource except Brown, Waring and Donkaewubua's (2008), who found that combining reading and listening modes achieves the best vocabulary learning effect. But no researches used pictures as another mode of input resource, which is worth further investigation and exploration. Thirdly, reading purpose seems to influence the outcome of incidental learning effect. Students with an academic purpose or a task to fulfill seem to achieve better learning performance. But the Taiwanese junior high students have a learning purpose all the time: they learn English for future academic studies. Then why do they still acquire vocabulary ineffectively? Other reading options for these students should be reconsidered. Fourth, some researchers believed that incidental learning was an overestimated - 14 -.

(24) phenomenon, which seldom happens to lower-achievers. But no previous study checked how placid and clear the context should be to be helpful for the lower achievers. With pictures as a kind of visual and immediate aid, perhaps lower achievers can acquire vocabulary through context as well. Studies on Picture Cues and Reading Comprehension Pictures play an important role in studies about reading and text comprehension and the value of picture cues to enhance readers' reading comprehension is proven. For example, Holmes (1987) conducted a research on 116 elementary school children's ability to answer inferential questions from print, pictures, and print with pictures. The materials were 15 photographs and 15 short passages of 150 words. These children were divided into three groups: print-only, picture-only and print-withpicture, and read passages and answered inferential questions. She found these children performed much better in picture-only and print-with-picture conditions than in print-only condition, which proved that pictures offer important clues for meaning during the reading process. Gambrell and Jawits (1993) assigned 120 subjects to four treatment groups: (1) subjects read an illustrated text and were told to create a mental image, (2) subjects read a non-illustrated text but were told to imagine, (3) subjects read an illustrated text, (4) subjects read a non-illustrated text; then all the four groups took a free recall and reading comprehension test with 16 questions. They found the first group attained the best performance in the recall and reading comprehension test and proposed that images and illustrations independently enhance reading comprehension. With the help from illustrations, readers could create mental images easily, which could induce their understanding and memory of the text. Liu (2004) used comic strips as a kind of visual aid, and examined whether - 15 -.

(25) presenting the text with or without comic strips could generate different results on L2 university students in the USA. In his research, he grouped these university students according to their language proficiency level (higher and lower achievers), and he provided them with four kinds of texts: (1) low-level text with comics, (2) low-level text without comics, (3) high-level text with comics, and (4) high level text without comics. He found that the lower achievers, with pictures, scored much better in an immediate recall test than other three groups. As for higher-achievers, the existence of comic strips did not seem to influence their performance since they were already proficient language learners; but for the lower-achievers, the existence of comic strips indeed helped them a lot for them comprehend a more complicated text. Guo (2010) demonstrated the value of picture cues in her study by asking 42 middle school students in Taiwan to read two short stories: one with a picture cue and the other without a picture cue. These students were told to write a free recall about the two texts and fill out a text reception questionnaire, and the result showed that using a picture cue could make a difference: the picture cue could help the students recall and the picture cue was perceived to decrease the difficulty level of the text but increased the clarity and amusement of the text, which made the students hold positive attitude towards picture cues in English reading. Some researchers pointed out that language alone may not promise the full understanding of a text; only when other modes of input are taken into consideration can the meaning of a text be fully represented (Kress, 2000), and images basically extend and enhance the communication between readers and the text, which contributes to comprehension (Glenberg, 1992; Mcvicker, 2007). These studies indicated that visual aids promote reading comprehension and incidental vocabulary learning indeed happens. Picture books, which contain both text - 16 -.

(26) and pictures extensively, might even have a better effect on reading comprehension and vocabulary acquisition, but they were often ignored in related researches. Thus, a research needed to be done on L2 junior high learners because if they could get exposure to both text and pictures as a kind of visual aid, it might trigger their vocabulary acquisition and contribute to their reading comprehension. Picture Books for Reading Comprehension and Vocabulary Acquisition As we can see, in studies on incidental vocabulary learning and reading comprehension, very few researchers have thought about adopting picture books; however, picture books, being used as the reading material, might have a better effect on readers' reading comprehension and incidental vocabulary learning for they provide both context and concrete images. Given the commonly reported view by students that English textbooks are boring, difficult and dull, helping students acquire a single positive reading experience can have a profound effect on students' attitude towards English reading and learning (Cho & Krashen, 2001). Nation (2001) has pointed out for second and foreign language learners, reading might be the only input resource for them out of the classroom (p.155). It is therefore suggested that providing students with interesting story books to read might be a remedy. Providing story books with pictures might perhaps reignite students' motivation to read, especially for slow readers. Children's picture books, which are short, simple, and usually come with colorful illustrations, seem to be an alternative when it comes to English reading. Picture books, according to Bishop and Hickman (1992), are written with a dual narrative, in which the pictures and the text cooperate closely to tell a story or present content (p.192). In L1 setting, some researchers have mentioned that the use of picture books can be powerful and inspirational to preschool education at home (Winn, 2002). - 17 -.

(27) Roberts (2008) found a positive relationship between home storybook reading and preschool children's vocabulary knowledge. She found that, whether in L1 or L2, picture book reading from care-givers to children at home enhanced children's vocabulary knowledge and helped children foster more interest in reading. Nicholas (2007) made emergent L1 readers (first-graders) join a regular reading programs in a local library in the USA, and she found children who saw illustrations when a book was being read demonstrated better performance in overall story comprehension and retelling ability. As for classrooms and schools, English language learners and struggling learners could also acquire some unfamiliar vocabulary because the messages in picture books are conveyed through both the text and illustrations (Carr, Buchanan, Wentz, Weiss& Brant, 2001; Hibbing& Rankin-Erickson, 2003; Linder, 2007). Sun (2010) proposed eight inherent features, which are also the strengths of picture books: (1) brevity, (2) writing style, (3) story, (4) illustrations, (5) rich vocabulary, (6) reading aloud function, (7) universal theme, and (8) rich culture resource. Picture books are short and concise; therefore, the small amount of text and vocabulary will not end up being a huge burden for language beginning learners. Meanwhile, illustrations supply visual aid to enhance students' reading comprehension and they can be viewed as a kind of reinforcement and extension of text. Picture books are written with authentic language and everyday vocabulary; with the help of illustrations, students are likely to pick up new words if they are shown vividly and visually because learning is built on the experience of connecting a word to a new perception of a visual image. Children's picture books are "multimodal texts." The color, line, shape in the images are all designed with a level of meaning and can be a source of interpretation - 18 -.

(28) (Beckman & Diamond, 1984; Hassett & Curwood, 2009). To our students, teenagers, who are usually "consumers of many types of visual media, (e.g., television, movies, video games, magazines)" picture books can be more motivating and present in a visual format (Linder, 2007). Even in China, an EFL environment, Liu (2006) found that setting up a classroom library with western picture books could have a positive influence on elementary school children's vocabulary gain, story-retelling ability, and their attitudes to English learning. Sheu (2008) investigated EFL primary school teachers' view on picture books and suggested three main values of picture books: (1) linguistic value, (2) the value of story, (3) the value of picture. Yet, it was a report with interviews and questionnaires from teachers' perspective; no direct and empirical studies were conducted on students. Lee (2007) investigated EFL college vocabulary acquisition through reading picture books and found 15 EFL college female students could achieve from 10 % to 14 % word gain after picture book reading. The main factors for a word to be learned were the exposure frequency, the familiarity of the topic and the words with word parts which can be analyzed. Some Taiwanese university students were found to achieve a high ratio of vocabulary acquisition improvement after reading picture books, which was about 20% among 50 target words no matter if teachers provided explanation or not (Sun, 2010). Lee's and Sun's studies focused on Taiwanese college students, who have longer English learning experience and both studies adopted Vocabulary Knowledge Scale test (VKS), which depends on the participant's self report. Because of the time interval between reading and testing, these students might have chances to look up the new words in dictionaries or discuss with others, which might weaken the validity of findings. Thus, a study on less mature L2 learners' reading comprehension and - 19 -.

(29) incidental learning of vocabulary immediately after reading, like the present study, was called for. In sum, the comprehensible input hypothesis points to the function of picture book reading for vocabulary learning; the DCT postulates that the concrete and vivid in picture books reinforces input; incidental learning model depicts the potential of picture book in promoting the acquisition of target words. Yet, no direct related research had been done on junior high school students in Taiwan before the present study. It is therefore hypothesized that reading story books with pictures appended could help EFL junior high school students achieve more vocabulary gain on the immediate and delayed vocabulary tests and facilitate their reading comprehension more than reading texts only. Besides, EFL junior high school students might be more interested in reading with pictures than reading texts only because pictures might make a story more comprehensible and a new word easier to guess. Hence, in the present study, the researcher hopes to answer the following research questions: (1) Does story reading have an effect on EFL junior high students' immediate and delayed vocabulary acquisition? (2) Does supplying pictures in story reading have an effect on EFL junior high students' immediate vocabulary acquisition? Does such effect vary according to story topics? (3) Does supplying pictures in story reading have an effect on EFL junior high students' delayed vocabulary acquisition? Does such effect vary according to story topics? (4) Does supplying pictures in story reading have an effect on EFL junior high students' reading comprehension? Does such effect vary according to story topics? - 20 -.

(30) (5) What are EFL students' perceptions of picture effects in terms of material (comprehensibility, vocabulary difficulty, interest level), use (when and how to use pictures) and the impact on reading outcome? The present study hopes to answer these questions and fill up the gap in the research area by providing specific data and proof.. - 21 -.

(31) CHAPTER 3 METHODOLOGY The pilot test and the main study will be presented with details and procedures in this chapter. Pilot Study Participants Prior to the main study, an informal inquiry was done with 12 students from an eighth grade class in a Taipei municipal junior high school, the school where the researcher worked and aimed to conduct the main study. The twelve students would not take part in the main study and they belonged to another class. Reading Material Three picture books were selected by the researcher; the titles of the three books are Curious George at the Baseball Game, Curious George Goes Camping and Curious George Goes to the Beach. They are from the same reading series, Curious George series, written for L1 Children, with the same illustration qualities and number of pages. This series is popular in the USA, written for American children to develop early literacy, and even got adapted to a cartoon on TV. Besides, it is full of colorful images and thus is suitable for exploring the effects from pictures during reading. What's more, the contents are concise and brief enough for the participants to finish reading in a class period but not enough to allow time for consulting dictionaries. As such, there’s no need for continuing reading at home. In addition, the series has not been published in Taiwan, which helps to make sure that the books had not been read by the participants in the present study. Meanwhile, the function of the illustrations in the three books was decorative in nature; no additional information beyond the text content was presented in the - 22 -.

(32) illustrations. This was made sure by two experienced teachers who closely examined the pictures against text content. For comparison, nine reading lessons from students' current English textbook, published by Nan-Yi publisher, were checked for readability. We found that the Flesch-Kincaid Reading Level of the nine lessons is 5.0 on average, ranging from 4.2 to 6.1. The three picture books come with 4.5, 4.1 and 4.6 respectively, which are within the readability range of these students' current textbook. Purposes of the Pilot Study The 12 students were asked to read the three picture books to help solicit information in order to help determine (1) time for reading each book respectively with or without pictures, and time for completing each vocabulary test, (2) selection of target words for vocabulary test, (3) type of vocabulary test to adopt, and (4) the suitability of the content in terms of students' content familiarity and affective response. The 12 students were asked to read either the three picture books or the corresponding texts without pictures silently during their lunch break. They were told to circle the words they did not know. Six of them were given multiple choice test while the other six took Chinese equivalent production test after they finished their reading. At this time, the testing items were decided by the research conductor and her colleague. The words that were not covered in the textbooks studied in the previous year were chosen. After taking the test, the 12 students were asked to talk about their feelings and opinions about the reading materials and tests. Results of the Pilot Study Based on the results from the informal pilot test, the following decisions were made: first, 20 minutes was adequate for them to finish their reading picture books or - 23 -.

(33) texts; 10 minutes was enough for them to finish the vocabulary test. Second, Curious George Goes to the Beach was ruled out of the study for the participants circled only about a mean of five new words, which is obviously lower than the other two. Therefore, it was deleted from the list and the other two books Curious George at the Baseball Game and Curious George Goes Camping were finalized. In the two books, 28 words were circled by all the 12 pilot students as unfamiliar ones in Curious George Goes Camping and 24 words in Curious George at the Baseball Game. These words were kept as the potential target words for the main study, waiting for further selection and deletion. Third, for the multiple choice test with 10 items, the correct answer got one point. The six students reached 7.4 mean scores on average; but in the Chinese equivalent production test containing 10 items, in which the correct answer got one point as well, the other six students got 3.9 points only. In brief, the Chinese equivalent production test seems too hard for eighth graders, given that students in Taiwan are more used to multiple choice test format, a meaning-recognition test. However, option items were in need of modification because some students still admitted that they made a guess in the test. For the Chinese equivalent production test, the researcher decided to use the original sentences in the books as the stem, hoping to improve students' performance. Last, for the content familiarity and students' interest in the two picture books, all the 12 pilot students agreed that the two books were interesting and challenging to them and the reading experience was pleasant.. Main Study Based on the preliminary results from the pilot test, the instruments and - 24 -.

(34) procedures for the main study were set. Participants The participants of the study were 64 eighth graders in two intact classes from a Taipei municipal junior high school. One class is composed of 32 students (15 females, 17 males) and another, 32 students (14 females, 18 males). A third class, with 31 students (15 females, 16 males), was served as a pilot group for the second-round pilot test. Their age ranged from thirteen to fourteen. About half of them have had formal English learning since they were in elementary schools; and half of them started to learn English since kindergarten. Besides, diversity in socio-economic background yielded a variety of additional exposure to English learning outside of school. Hence, to ensure comparability, the students' vocabulary level was measured prior to the main study, using a version of Nation's 1000 vocabulary level test (see Appendix A: Vocabulary Test from Nation: 1000 Level Test), which contains 39 items. Class 1, the pilot group, averaged a mean score of 27.90 (s.d. = 5.50). Class 2 was 27.12 (s.d. =8.35) and Class 3 was 26.52 (s.d. = 9.54); the mean scores of the two classes met the standard of Levene test of homogeneity, F=0.99, p=0.32>.05, and there was no significant difference between the two main classes' vocabulary proficiency level through the analysis of T test, t (63) = .27, ns. Class 1 was treated as the pilot group for further revision of vocabulary test; Class 2 and Class 3 participated in the main study. Each student in class 2 and 3 received two kinds of reading treatment (one picture book and one text only), and their grades on vocabulary tests and the reading comprehension tests were the main data in the study. Instruments Reading Material. In the pilot study, two books were finalized: Curious George at the Baseball Game (Driscoll, Hines, Rey Margret & Rey H.A., 2006) and Curious - 25 -.

(35) George Goes Camping (Rey Margret, Rey H.A., & Vipah Interactive, 1999). Both of the two books contain colored pictures, have 24 pages, and are from the same series. The series is popular with L1 kids in the USA. According to the 12 pilot students, the two books were interesting and challenging, and they would like to read more similar books in the future. Both books center on the same main character, George, a young naughty monkey. Curious George at the Baseball Game is mainly about George's adventure at a baseball game: George was invited to a baseball game, and then he had trouble. In the end, he still solved the problems and became a star on that day. Curious George Goes Camping is about how George got into trouble at a campsite but finally saved everyone by putting out a fire. Both of them contain 24 pages and have at least one picture per page. There are 604 words in Curious George at the baseball Game, and 860 words in Curious George Goes Camping. For the text-only version, some moderations were necessary. Without pictures, some sentences needed to be revised. For example, "This is George. He was a good little monkey." Because there is no picture as the referent, the researcher changed the sentences into one sentence such as "George was a good little monkey." The text-only versions contained words only, and all the illustrations were deleted. But they still looked like a book format without pictures (see Appendix B: Sample Pages from Picture Book and a Text for Text Only Version). Word Selection. The 12 pilot students identified some words unfamiliar or unknown to them in common. But further deletion was based on the following criterion: (1) words and derived forms were excluded from the target words list because they are in the 2000 word list for junior high students announced by MOE, such as tents and bucket (2) words that could be guessed from word-parts, such as campsite, unpack - 26 -.

(36) and scoreboard (3) words are too difficult for participants to infer their meanings due to the limited life experience of junior high school students or cultural differences, such as dugout (4) words that happened to appear in the student's mid-term exam at school, such as the phrase: get rid of, and it is hard to extract rid from the phrase in other contexts. The finalized 28 target words are shown in the following table: Table 1. Target Words after Word Selection Title of the Book Curious George Goes Camping Nouns: gear, marshmallows, pump, lizard, skunk, emergency, hiss (N=7) Verbs: wander, warned, poured, chase, peeking splashed, scrubbed (N=7) Curious George at the Baseball Game Nouns: stadium, signals, heels, cart, lens, sigh, sniff (N=7) Verbs: groaned, tugged, rubbed, scratched, distracted, crept, shrugged (N=7) These words would be our vocabulary test items, and the numbers of nouns and verbs happened to be the same. Vocabulary Test Type and Revision. In the informal inquiry done previously, the 12 students did not perform well on Chinese equivalent production test; however, they admitted that they guessed a lot on the multiple choice test. Hence, after target words selection, the revision of vocabulary tests were necessary. The researcher decided to include the original sentence in the story as the stem sentence, so as to help participants relate to the words. For instance, in the multiple choice test, the original sentence "At the campsite, the man with the yellow hat unpacked their gear" would be the stem, and the options would be shown in Chinese like this: A. 設備. B. 車輪. C. 引擎. D. 地毯 E. 我不知道. The researcher decided to add the last option: "I don't know" to eliminate students' - 27 -.

(37) guessing probability. The other three distractors had to belong to the same semantic category, the same word class, and had to be in a similar length as the correct answer (see Appendix C: Pilot Multiple Choice Test). For the Chinese equivalent production test, the researcher also used the original sentence as the stem. Therefore, the students would see a test item like the following example: 請寫出下列句子劃線部分的中文意思: At the campsite, the man with the yellow hat unpacked his gear. (See Appendix D: Pilot Chinese Equivalent Production Test). After revision, a second-round pilot was set to decide which test form would be appropriate for the main study. Class 1 was divided into two groups: 16 of them took the Chinese equivalent production test, and 15 of them took the multiple choice test after reading the story: Curious George Goes Camping. In both tests, the correct answer could get 1 point. It was found that the Chinese equivalent production subgroup reached the mean score of 2.81(s.d.=2.68), and the multiple choice test subgroup reached the mean score of 7.53 (s.d.=1.35). There was still a huge gap between the two kinds of testing types, Levene test= 9.04, p<.05. Besides, because the researcher had provided the Chinese equivalents as the options, which somehow gave the pilot students a hint, they were all willing to make a guess by making an inference or comparison between the stem sentence and the answer options. Hence, the researcher decided to choose multiple choice test as the finalized version of testing type, but the stem sentences needed to be revised again. For example, when the students saw the word: campsite, they would presume that "設備" was the most related to "camping" among the four. The researcher followed Nations’ principle: cutting short of the stem sentences and providing the least clues. For example, in the original stem sentence: “At campsite, - 28 -.

(38) the man with the yellow hat unpacked his gear", “at campsite” was deleted for it might be too informative and could easily lead participants to infer the meaning of “gear”. If the stem sentences were impossible to be shortened, then all the distractors needed to be distracting to the same extent. For instance, in the item “He splashed in the creek” (Camping, 8), the researcher provided the participants with all four options related to “water”. The stem sentences basically were modified and constructed with the syntax order: Subject+ Verb+ (Object). The finalized multiple choice test is shown as Appendix E: Finalized Multiple Choice Test; there are 7 nouns and 7 verbs for each test altogether. For delayed vocabulary test, the sequence of the four options and the sequence of the 14 items were altered, so that the influence of memory for answers was minimized. Reading Comprehension Test. The researcher would like to know if the existence of illustrations could help the participants to understand the main ideas of the books, so the researcher designed the reading comprehension tests in the form of short-answer factual questions at first. Basically, there was a question for every 2 or 3 pages (namely, an episode); the reading comprehension tests contained 8 short-answer questions written in Chinese. Students could answer either in English or Chinese, and as long as students could provide short answers corresponding to the plots, they could get a point for each question. For instance, in the question, "What did George and its master see first at the campsite? 到了營地時, George 和主人先看到了甚麼? (Camping, 1)," students could get a point if they answered "tents" or "帳篷". Class 1, the second-round pilot group, took the reading comprehension tests after reading the camping story. The picture-book reading subgroup performed much better than the text-only reading subgroup. The picture-book group got a mean score of 7.53, and the text-only group got only 5.44, Levene test =11.18, p <0.05, when the perfect - 29 -.

(39) score was 8. However, the short-answer test format containing only 8 questions might lead to a biased result because it did not provide enough test items for data analysis, and the questions were not precise or detailed enough, which would make the test hard to be scored. After talking with the professional, the researcher decided to adopt the True or False format and designed a reading comprehension test with 14 questions, hoping to provide enough test items and concrete answers as the scoring standard, avoiding any scoring controversy (see Appendix F: Reading Comprehension True or False Questions). Questionnaire. In order to obtain more qualitative data about these participants, a questionnaire was also designed to help the researcher to understand more about the participants' reading and L2 learning habits (see Appendix G: Questionnaire). The researcher would like to see if the participants felt greatly different when they had illustrations as the visual aid during reading, and the researcher wanted to know if they thought pictures had a great influence on their vocabulary test and reading comprehension. The researcher was also interested in knowing the participants’ perceptions and interest level about picture books reading. Study Design and Data Collection Procedures Class 2 and Class 3 participated in the main study. Each class was divided into two sub-groups: Group A and Group B. Both read the intact picture books, and read texts only in a counter-balanced sequence in two sessions in consequential two weeks. In the first week, after signing the Consent Form (see Appendix I), students first read the first text, either with or without pictures, in 20 minutes. Then, with materials collected, they performed an immediate multiple-choice vocabulary test in 10 minutes followed by a true-false reading comprehension test in another 10 minutes. The same - 30 -.

(40) procedure was followed for the second reading in the second week. In the fourth week, a delayed vocabulary test on the same words from the two readings was administered. Additionally, a questionnaire survey on perceptions of pictures and stories was given. The following table shows the outline of the data collection procedures: Table 2. Data Collection Procedures Title Week 1. Class 2: Camping Class 3: Baseball. Procedures Group A (N=16) Read picture books (20 min.) → Vo. test (10 min.) and RC test (10 min.) Group B (N=16) Read text-only versions (20 min.) → Vo. test (10 min.) and RC test (10 min.). Week 2. Class 2: Baseball Class 3: Camping. Group A (N=16) Read text-only versions (20 min.) → Vo. test (10 min.) and RC test (10 min.) Group B ( N=16) Read picture books (20 min.) → Vo. test (10 min.) and RC test (10 min.). Week 4. Delayed Vo. test (15 min.) and Questionnaire (10 min.). Scoring Data Scoring Vocabulary Test. The multiple choice test format was used. There were 14 items in a multiple choice test, and the right answer got one point. The perfect score was 14 points. The wrong answers and “I don’t know” did not get any point. Each multiple choice test had 7 verbs and 7 nouns. So there were 14 verbs and 14 nouns in total for the two multiple choice tests. The same scoring principle was applied to the immediate vocabulary tests and the delayed vocabulary tests. - 31 -.

(41) Scoring Reading Comprehension Test. In the True or False test, participants could get one point if they could judge the sentence was true or false correctly according to the story they read. For example, the students could get one point if they answered "T" to the description: "George's friend asked George to get some water. (George 的朋友要求 George 去幫忙取水。) The total score of each reading comprehension test was 14 points. Coding Questionnaire. The first two questions were about the participants' basic personal information. Question 3 to Question 6 were set to ask about students' learning habits and strategies. In Question 3 and Question 4, students were asked to check if they had habits of reading picture books and using language learning tools. For Question 5, set for students' learning strategies, item frequency was calculated. Question 6 asked students to name their specific English learning medium. Question 7 and Question 8 had students indicate the difficulty level of the two reading experiences. Question 9 to Question 11 solicited the participants' perceptions of the two picture books. Question 12 was set to detect the importance of pictures to the students. Question 13 and 14 were set to make students provide specific examples. Question 15 and Question 16 were set to help understand if students would like to read this kind of reading materials more in the future. So Question 3, 4, 5, 7, 8, 9, 10, 11, 12, 15, 16 were determined by total numbers and frequency, and Question 6, 13, 14 would be coded. Data Analysis Quantitative Analysis. Two round ANOVA repeated measure was first used after the data collection was completed. One round was conducted on immediate vocabulary multiple choice test, and the other round was done on delayed vocabulary multiple choice. The independent variable was input modes: text only versus picture book - 32 -.

(42) reading. However, the finding violated Levene Test of homogeneity and T tests (Paired-T test and Independent T test) were used at last. As for the reading comprehension, T tests were used as well. On three tests, Paired-T test was conducted to examine the overall effects of pictures and Independent T test was adopted to scrutinize the effects of pictures across different book topics. Qualitative Analysis. The data resource was mainly from the questionnaire. These students' reading habits and learning strategies were scrutinized, and the researcher would like to see if pictures really helped them more to achieve reading comprehension and vocabulary acquisition from their points of view. The students' perceptions of the reading materials were also shown from the questionnaire.. - 33 -.

(43) CHAPTER 4 FINDINGS This study investigates the effects of pictures on EFL junior high school students' vocabulary acquisition and reading comprehension in book reading. In this chapter, the researcher presents the findings in four sections in response to the five research questions in the last section.. Effects on Vocabulary Acquisition In this part, two vocabulary tests (immediate vocabulary test and delayed test) were both analyzed through descriptive statistics and Paired-T test to see the impacts of picture book reading on junior high school students' vocabulary acquisition, followed by Independent T test to check the varied effects of pictures across different topics.. Effects of Reading on Immediate Vocabulary Acquisition First of all, based on the pilot test, all the 28 target words were claimed unfamiliar by the participants before reading. However, as shown in Table 3, overall, 51% of the target words (M=7.24, s.d.=3.45) were still indentified by the students through reading books with or without pictures. It means that the students could recognize about half of the target words, and infer the Chinese equivalents of them no matter which reading mode they were under.. - 34 -.

(44) Table 3. Mean Scores, Standard Deviations and Mean Percentage of Immediate Vocabulary Tests Mode. With Pictures (s.d.). Text Only (s.d.). Overall (s.d.). Camping. 9.37 (3.03) 66%. 6.24 (3.55) 45%. 7.95 (3.61) 56%. Baseball. 6.45 (2.90) 46%. 6.60 (3.40) 47%. 6.53 (3.16) 47%. Overall. 8.05 (3.29) 58%. 6.44 (3.45) 46%. 7.24 (3.45) 51%. N= 64; Standard deviation in parenthesis; total number of words: 28. Second, it is noteworthy that when these students read picture book versions, they obtained even more than half of the words, about 58%, (M= 8.05, s.d.= 3.29), more than the Text-only mode, 46%, (M= 6.44, s.d.=3.45). However, Lavene Test of homogeneity showed a significant result, F=2.79, p<.05; therefore, ANOVA assumption of homogeneity was not met. Paired-T test was thus used to compare two mean scores. The Paired-T test on the two immediate vocabulary tests showed that "Mode" (with or without pictures) influenced these students' performance, t (63) = -4.35, p<.001 (see Table 4). Next, across the two reading topics, Independent T test was used to analyze the two books respectively. Since Lavene test results showed that the homogeneity assumption for ANOVA tests was violated, it would not be possible to examine the interaction effects between mode and topic using ANOVA. Therefore, the effect of mode was further examined for each of the two topics on the immediate vocabulary test, using two Independent T tests. Hence, within each topic, different groups of students read a text in different modes, yielding mode as the only betweensubject factor. - 35 -.