Citation
Developing English Determiners through Internet Chat: An Experiment with Korean EFL Students

Material Information

Title:
Developing English Determiners through Internet Chat: An Experiment with Korean EFL Students
Creator:
HA, JONGBUM ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Conversation ( jstor )
English language learners ( jstor )
Foreign language learning ( jstor )
Function words ( jstor )
Language ( jstor )
Learning ( jstor )
Nouns ( jstor )
Reading comprehension ( jstor )
Second language acquisition ( jstor )
Students ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Jongbum Ha. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
2/28/2006
Resource Identifier:
436098811 ( OCLC )

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

DEVELOPING ENGLISH DETERMINERS THROUGH INTERNET CHAT: AN EXPERIMENT WITH KOREAN EFL STUDENTS By JONGBUM HA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Jongbum Ha

PAGE 3

This document is dedicated with love and gratitude to my parents and my wife. Without their love, support and blind faith in me, this dissertation would not have been possible.

PAGE 4

ACKNOWLEDGMENTS I would like to express heartfelt thanks to my supervisory committee chair (Dr. Roger M. Thompson), for inspiring and nurturing my passion for second language acquisition (SLA) and teaching English as a second language (TESL). His direction and influence were instrumental in my development as an SLA researcher. I will always remember him as an exceptional ESL teacher trainer. I thank Dr. Theresa Antes (one of my supervisory committee members) for her constant support and guidance through my graduate studies at University of Florida. I admire her excellent teaching skills, and I appreciate her thorough knowledge and valuable comments on my dissertation. I thank Dr. Gillian Lord (another committee member) for her critical review of each chapter of my dissertation. Her thorough and sometimes harsh feedback contributed to the development of my academic writing skill. Also, I thank my external committee member (Dr. Greg Ulmer) for sharing my research interests and for his balanced comments on my dissertation. I am indebted to the following people for their assistance and guidance with data collection and analysis: Cheol-Gi Yoo, Kwang-Cheol Park, Sang-Hyeon Seo, Yeon-Ok Cheon, Kirsten Benkert and Carolina Von Vajna. I am grateful to my Korean colleagues in the Linguistics program at the University of Florida (Duk-Young Kim, Chang-Yong Yang, Sang-Hee Yeon, Sun-Hee Kwon, Hee-Nam Park, Kyeong-Ok Paek, Jong-Kuk Lee and Yoon-Pil Jeong) for their friendship during my graduate study at University of iv

PAGE 5

Florida. I am also grateful to Seung-Man Kang for his encouragement during my dissertation writing. Special thanks go to Phillip Monahan for his extensive help with my dissertation and my study at University of Florida and for his friendship with me and my family. Special thanks also go to Ji-Woon Yang, Il-Ho Kang, Hyun Paek, Kwang-Pyo Kim and other members of Gainesville Korea Catholic Community (GKCC) for being my friends and enriching my life. I would like to thank my mentors in Korea (Won Park, Jae-Sil Shin, Seung-Chull Lee, Byeong-Choon Lee and Moon-Sub Han) for their encouragement. I want to say “thanks” to my two brothers (Jong-Soo Ha and Jong-Woo Ha) for their lifelong love and concern. Finally, I thank my parents, Yeong-Joon Ha and Yeong-Hee Ko; and my wife, Eun-Sook Oh, for their enduring support and love and for their blind faith in me. v

PAGE 6

TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................iv LIST OF TABLES ...........................................................................................................viii LIST OF FIGURES ...........................................................................................................xi ABSTRACT .....................................................................................................................xiii CHAPTER 1 INTRODUCTION........................................................................................................1 Introduction...................................................................................................................1 The Determiner System................................................................................................4 Theoretical Explanation of the English Determiner System.................................4 Second Language Acquisition of the English Determiner....................................7 Possible reasons for difficulty in acquiring the English article system..........8 Previous research on the acquisition of the English article (determiner) system.....................................................................................................12 2 NOTICING IN SLA...................................................................................................17 Three Perspectives on Noticing..................................................................................18 Modality as a Variable for Noticing...........................................................................21 The Effect of Textual Enhancement on Noticing.......................................................24 The Role of Output in Noticing..................................................................................29 Use of Chat for Noticing L2 Function Words............................................................34 Characteristics of Chat........................................................................................34 Potential of Chat for Developing L2 Function Words........................................36 Research Questions.....................................................................................................39 General Research Question.................................................................................39 Research Question 1............................................................................................40 Research Question 2............................................................................................40 3 METHODOLOGY.....................................................................................................42 Participants.................................................................................................................42 Target Structure..........................................................................................................43 vi

PAGE 7

Instruments and Materials...........................................................................................44 Treatment and Procedures..........................................................................................46 Data Collection and Analyses.....................................................................................48 4 RESULTS...................................................................................................................52 Research Question 1...................................................................................................52 Analyses of Test Scores.......................................................................................53 Analyses of Error Types in the Test....................................................................58 Research Question 2...................................................................................................63 The Use of Determiners in Convergent and Divergent Tasks.............................64 Analyses of Errors to Total Head Nouns.............................................................68 Analyses of Deletions to Total Errors.................................................................72 Qualitative Data..........................................................................................................78 5 DISCUSSION AND CONCLUSION........................................................................83 Discussion...................................................................................................................83 Research Question 1............................................................................................83 Research Question 2............................................................................................88 General Research Question.................................................................................92 Implications................................................................................................................93 Limitations and Suggestions for Future Research......................................................94 Conclusions.................................................................................................................96 APPENDIX A INFORMED CONSENT............................................................................................98 B QUESTIONNAIRE..................................................................................................100 C PRETEST AND POSTTEST...................................................................................104 D MATERIALS FOR GROUP E.................................................................................108 E MATERIALS FOR GROUP C................................................................................124 F MATERIALS FOR GROUP EC..............................................................................136 G SAMPLE CHAT TRANSCRIPTS...........................................................................141 LIST OF REFERENCES.................................................................................................146 BIOGRAPHICAL SKETCH...........................................................................................152 vii

PAGE 8

LIST OF TABLES Table page 1-1 The order of English determiners in a NP..................................................................5 1-2 Bickerton’s Dynamic Paradigm.................................................................................7 3-1 The participants’ demographic information.............................................................43 3-2 Target structures of study.........................................................................................44 3-3 The procedures of the experiment............................................................................46 3-4 Types of errors with English core-determiners........................................................49 4-1 Group C scores for pretest and posttest (N=16).......................................................53 4-2 Group E scores for pretest and posttest (N=16).......................................................53 4-3 Group EC scores for pretest and posttest (N=16)....................................................53 4-4 Control group scores for pretest and posttest (N=16)..............................................54 4-5 Attention-E scores for pretest and posttest (N=7)....................................................54 4-6 No attention-E scores for pretest and posttest (N=25).............................................55 4-7 Control group scores for pretest and posttest (N=16)..............................................55 4-8 Statistical analysis on the test scores........................................................................55 4-9 Regression of “pretest”.............................................................................................56 4-10 Regression of “pretest,” “treatment type,” and “attention-E”..................................57 4-11 Attention-C scores for pretest and posttest (N=15)..................................................57 4-12 No attention-C scores for pretest and posttest (N=17).............................................57 4-13 Control group scores for pretest and posttest (N=16)..............................................58 4-14 Attention-E error types in pretest and posttest (N=7)..............................................59 viii

PAGE 9

4-15 No attention-E error types in pretest and posttest (N=25).......................................59 4-16 Control group error types in pretest and posttest (N=16).........................................60 4-17 Statistical analyses on error types in the pretest and posttest...................................61 4-18 Total head nouns to total words (N/W) in chat convergent and divergent tasks......64 4-19 Ratio of errors to total head nouns (Err/N) in chat convergent and divergent tasks..........................................................................................................................64 4-20 Ratio of deletions to total errors (Del/Err) in chat convergent and divergent tasks.65 4-21 Ratio of retentions to total errors (Ret/Err) in chat convergent and divergent tasks..........................................................................................................................65 4-22 Ratio of overuse of “the” to total errors (The/Err) in chat convergent and divergent tasks..........................................................................................................65 4-23 Ratio of overuse of “a(n)” to total errors (A/Err) in chat convergent and divergent tasks..........................................................................................................65 4-24 Statistical analyses on N/W, Err/N, Del/Err, Ret/Err, The/Err and A/Err in chat convergent and divergent tasks................................................................................66 4-25 Correlations of test scores and Err/N for convergent tasks......................................67 4-26 Correlations of test scores and Err/N for divergent tasks.........................................67 4-27 Attention-E Err/N in chat convergent tasks (N=5)...................................................68 4-28 No attention-E in chat convergent tasks (N=27)......................................................68 4-29 Attention-E Err/N in chat divergent tasks (N=5).....................................................69 4-30 No attention-E Err/N in chat divergent tasks (N=27)..............................................70 4-31 Attention-C Err/N in chat convergent tasks (N=15)................................................70 4-32 No attention-C Err/N in chat convergent tasks (N=17)............................................71 4-33 Attention-C Err/N in chat divergent tasks (N=15)...................................................72 4-34 No attention-C Err/N in chat divergent tasks (N=17)..............................................72 4-35 Attention-E Del/Err in chat convergent tasks (N=5)................................................73 4-36 No attention-E Del/Err in chat divergent tasks (N=27)............................................73 ix

PAGE 10

4-37 Attention-E Del/Err in chat divergent tasks (N=5)..................................................74 4-38 No attention-E in chat divergent tasks (N=27).........................................................74 4-39 Attention-C Del/Err in chat convergent tasks (N=15).............................................75 4-40 No attention-C Del/Err in chat divergent tasks (N=17)...........................................75 4-41 Statistical analyses on Del/Err during chat convergent tasks...................................76 4-42 Attention-C Del/Err in chat divergent tasks (N=15)................................................76 4-43 No attention-C Del/Err in chat divergent tasks (N=17)...........................................76 4-44 Statistical analyses on Del/Err during chat divergent tasks.....................................77 4-45 Comparison of the two students’ determiner use in the test and chat......................81 x

PAGE 11

LIST OF FIGURES Figure page 4-1 Scores for pretest and posttest according to treatment type.....................................54 4-2 Scores for pretest and posttest according to meta-awareness during input enhancement.............................................................................................................55 4-3 Scores for pretest and posttest according to meta-awareness during chat...............58 4-4 Del/Err in pretest and posttest according to meta-awareness during input enhancement.............................................................................................................60 4-5 Ret/Err in pretest and posttest according to meta-awareness during input enhancement.............................................................................................................60 4-6 The/Err in pretest and posttest according to meta-awareness during input enhancement.............................................................................................................61 4-7 A/Err in pretest and posttest according to meta-awareness during input enhancement.............................................................................................................61 4-8 N/W, Err/N, Del/Err, Ret/Err, The/Err and A/Err shown in chat convergent and divergent tasks..........................................................................................................66 4-9 Err/N in chat convergent tasks according to meta-awareness during input enhancement.............................................................................................................69 4-10 Err/N in chat divergent tasks according to meta-awareness during input enhancement.............................................................................................................70 4-11 Err/N in chat convergent tasks according to meta-awareness during chat...............71 4-12 Err/N in chat divergent tasks according to meta-awareness during chat.................72 4-13 Del/Err in chat convergent tasks according to meta-awareness during input enhancement.............................................................................................................73 4-14 Del/Err in chat divergent tasks according to meta-awareness during input enhancement.............................................................................................................74 xi

PAGE 12

4-15 Del/Err in chat convergent tasks according to meta-awareness during chat............76 4-16 Del/Err in chat divergent tasks according to meta-awareness during chat...............77 xii

PAGE 13

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy DEVELOPING ENGLISH DETERMINERS THROUGH INTERNET CHAT: AN EXPERIMENT WITH KOREAN EFL STUDENTS By Jongbum Ha August 2005 Chair: Roger M. Thompson Major Department: Linguistics My study examined whether Internet chat could be a useful tool to develop L2 function words, such as English determiners. In chat conversations, frequent omissions of grammatical features of low communicative value (i.e., function words) are prevalent in order to maximize time and effort. For this reason, a minimal level of meta-awareness during chat conversations is necessary for students to use targeted items properly. My general research question was whether it is possible to encourage students to focus on and use determiners so that students will learn targeted forms during a chat activity. My study compared the performance of four groups of university students in Korea (N=16) who experienced different activities (textual enhancement, chat, textual enhancement and chat, and control) for 4 weeks. For textual enhancement, students were required to read a textually manipulated passage in which all core-determiners and following nouns in the passages were highlighted and each of the core-determiners was underlined. For chat activities, students were paired and assigned a certain task in MSN xiii

PAGE 14

Messenger. For textual enhancement and chat, students experienced both tasks. The effects of these treatments were investigated in terms of the gains of scores on a posttest and the ratio of deletion and retention of determiners and the ratio of overuse of “the” and “a(n)” in the preand posttests and in students’ chat transcripts. Results show that none of the activities significantly affected students’ performance. However, when students’ meta-awareness was considered a variable, textual enhancement helped students perform better in the targeted forms (reflected by higher scores and change of error type), given meta-awareness. On the other hand, chat did not affect students’ performance, even with meta-awareness. Although meta-awareness during chat helped students significantly decrease the ratio of deletion of determiners in chat conversations, deletions of determiners were still prevalent throughout chat. This is because the need to focus on articles was not there and, accordingly, students did not have many chances to notice the targeted forms. In answering the general research question, chat is unlikely to be an appropriate forum for asking students to focus on and use determiners, particularly articles . This is due to the nature of chat. On the other hand, considering that students in my study were at the intermediate (or lower) level of English, it was too difficult for them to focus on less-urgent grammatical features (i.e., articles), when participants had to pay attention to more-urgent features (such as prepositions, pronouns, and verb tense). xiv

PAGE 15

CHAPTER 1 INTRODUCTION Introduction Chat is an increasingly popular form of communication throughout the world. Chat was defined in my study as synchronous text-based online chat as contrasted with asynchronous interaction (e.g., electronic mail, bulletin board) or visual interaction (e.g., video conference) through Internet. Because of its unique nature (i.e., written yet synchronous/interactive), its application in second language acquisition (SLA) was recently pursued. Research shows that chat can reduce learners’ anxiety, encouraging them to produce more target language output and helping them obtain greater fluency (Bump, 1990; Kern, 1995; Oliva & Pollastrini, 1995; Warschauer, 1998). Furthermore, research shows that task-based synchronous network communication (such as chat) could also foster negotiation of meaning and thus improve grammatical competence (Blake, 2000; Kitade, 2000; Lee, 2002; Pellettieri, 2000; Saleberry, 2000). This is because, during negotiation, learners are given corrective feedback, which leads second-language (L2) learners to incorporate target language forms into subsequent output. Yet, few studies have examined chat for the development of function words, such as English determiners. A function word is one with grammatical meaning as apposed to a content word, which has a lexical meaning. The acquisition of English determiners by Korean speakers and the role of chat in their process are the focus of my study. Among English determiners, the English article is extremely difficult to learn and master for speakers of languages without an article system. Articles are the most 1

PAGE 16

2 commonly used words in English, but their use is complex. Explicit teaching of English articles alone may be relatively ineffective for these learners because of the semantic complexity of the articles’ distribution. Also, too much focus on the explicit teaching method could impair acquisition of communicative aspects of the language. Furthermore, even though current second language pedagogy strongly favors more-communicative methods (while not eliminating the explicit way of teaching), these methods present inherent problems for acquisition of articles because of the lack of salience of articles in speech. Given such problems, chat presents a possibly greater potential. The written nature of chat may reduce inherent problems of communicative methods for learning L2 function words, since it enables students to notice, monitor and edit in a self-paced learning environment (Lee 2001). As a result, students can attend to words that are not phonologically salient in speech. Moreover, the synchronous, interactive nature of chat allows learners to integrate into the discourse level and to develop awareness of variations of article use in communicative contexts (Pica, 1983). This is important because the linguistic information that the student needs to use and interpret English articles is often discourse-related, and semantically too complex to learn through explicit teaching methods alone. Despite this potential, however, chat has some problems for students acquiring target language (TL) function words. Given that, in chat, there is frequent omission of function words in order to save time and effort, students seem to need a minimal level of meta-awareness during chat conversation to use the targeted items properly. Another concern is the lack of negotiation of meaning of articles. Unfortunately, learners’ misuse of articles seldom results in the communication breakdowns (which might trigger

PAGE 17

3 corrective feedback). For these reasons, my study used an additional session (i.e., input enhancement) to promote students’ meta-awareness during chat and give them models of target forms through this input session. We combined input enhancement and chat sessions to improve students’ ability to notice the targeted forms. My study was based on the assumption that it is easier for students to concentrate on both meaning and form in written mode (i.e., chat) than in face-to-face conversation. In this case, “meaning” refers to real-world, referential concepts (e.g., time, person, aspect, etc.); and “form” refers to grammatical surface features of language (e.g., verb endings for person, number, tense, mood, gender and number) as well as functional items (e.g., prepositions, articles and other non-content words) (VanPatten, 1996). If learners are better able to attend to determiners during chat conversation than in oral interaction, students may have more chances to learn the complex semantic distribution of determiners implicitly, as contrasted with explicitly. The purpose of my study was to investigate if Internet chat could be a useful tool to develop L2 function words, such as English determiners. Given the potential problems with chat (i.e., the frequent omission of function words), however, we needed to find out if it was possible to encourage students to focus on and use determiners so that they can notice the targeted forms during chat activity. I hoped that my study would help second and foreign language teachers evaluating the usefulness of chat for developing students’ L2 competence. In addition, if we could find positive effects of chat for learning English articles, my study would serve as a point of departure for future research on developing L2 function words through chat. With Internet access becoming more global, this form of

PAGE 18

4 acquisition will become a more accessible supplement to classroom teaching for everyone learning an L2. The Determiner System This section gives a theoretical explanation of the English determiner system and a review of its acquisition by L2 learners. It then presents the Korean equivalent for English core-determiners, which would contribute to a better understanding of the English determiner system and its acquisition by Korean learners. Theoretical Explanation of the English Determiner System The Concise Oxford Dictionary of Linguistics (Matthews, 1997) defines a determiner as “any of class of grammatical units characterized by ones that are seen as limiting the potential referent of a noun phrase.” Celce-Murcia & Larsen-Freeman (1999) used “determiner” to mean “the special class of words that limit the nouns that follow them” (p.19), and included various types of words in this category, such as articles, demonstratives, possessives, quantifiers, multipliers (e.g., twice, three times), and cardinal/ordinal numbers. Their classification separates the determiner from the adjective, while older grammars incorporate them into the adjective word class (Jespersen, 1933). In fact, in English, determiners precede an adjective if one is present; otherwise, they precede a noun. Radford (1997:46-47) distinguished determiners from adjectives using four properties: The same type of adjective may be used recursively (the eloquent, articulate man), but the same type of determiner cannot (e.g., *the, this man), while some determiners may occur together (e.g., all my friends; my many friends). Syntactically, determiners occupy a separate slot (i.e., the specifier position) – and must precede all adjectives (e.g., the bright sunny day; *bright the sunny day).

PAGE 19

5 A singular count noun cannot stand without a determiner, though it can stand without an adjective (e.g., the boy; *tall boy). Most determiners can only modify nouns with specific count properties, but no adjectives are similarly constrained (e.g., *a men; *these building; *many water; red men; a gray building; blue water). Simons (2001) more plausibly defines English determiners as “function words with little or no lexical meaning that modify a noun and carry overt or covert person, number, gender, and case properties” (p. 8). This definition distinguishes the determiner from the adjective, which is a content word with one (or more than one) lexical meanings. Additionally, Simons noted that determiners have “the grammatical function that determines the referential or quantificational properties of the noun they accompany” (p.4). Various types of words fall into this category: for example, articles, possessives, demonstratives, quantifiers, ordinal/cardinal numbers, and multipliers. English determiners are generally divided into three types in SLA literature: pre-, core-, and post-determiners. Celce-Murcia &` Larsen-Freeman (1999:335) suggest the order of determiners in a noun phrase (NP) (Table 1-1). Table 1-1. The order of English determiners in a NP Pre-determiners Core-determiners Post-determiners Quantifiers: all, both, half Multipliers: twice, three times, etc. Articles: , a(n), the Possessives: my, your, his, her, its, their, our, -’s Demonstratives: this, that, these, those Quantifiers: some, any, no, each, every, either, neither, enough Cardinal numbers: one, two, three, etc Ordinal numbers: first, second, third, etc. General ordinals: next, last, another, etc. Quantifiers: many, much, (a) few, (a) little, several, more, less, most, least Phrasal quantifiers: a great deal of, a lot of, a good number of, etc. Note: Nouns with no salient core-determiner are considered to follow the zero article ().

PAGE 20

6 Accordingly, articles can optionally be preceded by one pre-determiner and followed by one or two post-determiners. In other words, as Celce-Murcia & Larsen-Freeman (1999) note, “it is possible to sequence determiners in an English noun phrase by picking one determiner from the pre-determiner column, one from the core determiner column, and one or more from the post-determiner column” (p. 335), as shown by these examples (Celce-Murcia & Larsen-Freeman 1999:335): All our many hopes were kept alive by her encouraging words. (1-1) pre core post These next two weeks before school starts will be hectic. core post post Each of the core-determiners is mutually exclusive in English, though not necessarily so in other languages. In English, therefore, no more than one core-determiner can occur in a noun phrase (NP) as shown by these examples. the big car (1-2) some other cases my two sisters *the no person *my the book *the these houses *those some eggs. In Example 1-2, the first three noun phrases are grammatically correct, because they have only one core-determiner (the, some, and my), each of which is followed by an adjective (big) or a post-determiner (other/two). The last four noun phrases are ungrammatical, because they have more than one core-determiner (e.g., the, no, my, the, these, those, and some). Among the core-determiner category in English, articles hold a unique position: They have little lexical meaning but are semantically more complex than other

PAGE 21

7 determiners in English. An English article has no salient meaning in itself, but has complex semantic distribution in discourse. The Concise Oxford Dictionary of Linguistics (Matthews, 1997) defines the article as “a determiner whose basic role is to mark noun phrases as either definite or indefinite.” For example, “the” is definite in “the car” and “a” is indefinite in “a car.” The use of English articles pertains to discourse referentiality (they refer or point to something in discourse). Bickerton (1981) provides a systematic approach to the analysis of English article usage. According to Bickerton, the use of the English articles – “a(n),” “the,” and “” is governed by the semantic function of the noun phrase (NP) in discourse. The classification of the semantic function of an NP is determined by two binary discourse features: (a) whether a noun has a specific referent ( + SR); and (b) whether the hearer knows the referent ( + HK). Based on such an analysis, NPs fall into four major semantic types (Table 1-2). Table 1-2. Bickerton’s Dynamic Paradigm Type 1 -Specific referent +Assumed known to hearer Type 2 + Specific referent + Assumed known to hearer Type 3 + Specific referent Assumed known to hearer Type 4 Specific referent Assumed known to hearer Type 1 [-SR] [+HK]: e.g., We need water to live. / A lion is brave. / The Germans are diligent. Type 2 [+SR] [+HK]: e.g., Can I drive the car ? Type 3 [+SR] [-HK]: e.g., I saw a funny looking dog today. Type 4 [-SR] [-HK]: e.g., I don’t have a car . Second Language Acquisition of the English Determiner Much of the research on acquisition of determiners in English focuses on articles specifically, because the rules of the English article system are typically more difficult to acquire than those of the other determiners. To better understand the acquisition of

PAGE 22

8 English determiners, this section examines why it is difficult and reviews previous research. Possible reasons for difficulty in acquiring the English article system The English article system is difficult to learn for at least three possible reasons: L1 transfer, the lack of salience of articles in speech, and the semantic complexity of the article system. The first reason is L1 transfer. Many languages with no article system use a topic-comment system; in contrast, English uses a combination of indefinite and definite articles. The topic provides given (or old) information (already familiar to the hearer by a previous utterance or some other means). The comment provides new information (new to the hearer and is being introduced into the discourse for the first time). The English article system is particularly difficult for learners of English whose L1 lacks articles. Articles are placed in the core-determiner category in English, so comparing the Korean equivalent for English core-determiners shows better why Koreans have more difficulty in learning English articles than learning the other core-determiners. English core-determiners have four components: demonstratives, possessives, quantifiers, and articles. The Korean equivalents are as follows: First, Korean has a three-way demonstrative system (Example 1-3). a. i -saram “this person” (1-3) b. jeo -saram “that person” c. geu -saram “the person over there who we already mentioned; or the person we already mentioned” First, i “this” marks a noun as definite and near the speaker; jeo “that” marks a noun as definite and farther from the speaker; and geu marks a noun as definite and even farther away, or a noun already mentioned in the discourse (though not present). All three

PAGE 23

9 of the Korean demonstratives precede their nouns and are dependent on them (In other words, they are clitics); and cannot occur independently as pronouns, as the demonstratives in English can (e.g., “That person is my teacher”; “That is my bag”). Korean has no equivalent for such English plural demonstratives as “these” and “those.” Instead, in Korean, the plurality marker -deul is added after a noun (e.g., i-saram-deul “these people”). Second, Korean has the genitive particle eui that combines with a (pro)noun (Example 1-4). a. i-sarameui chaek (1-4) this-person-genitive particle book “this person’s book” b. jeoeui chaek (= je-chaek) I-genitive particle book “my book” In Example 1-4a, the genitive particle eui combines with the common noun saram giving the meaning “person’s.” In Example 1-4b, the genitive particle eui combines with the pronoun jeo, meaning “my,” and this combination can contract to form the possessive pronouns (e.g., jeo contracts with eui to form je). These pronouns precede their nouns and function similarly to their counterparts in English. Third, Korean quantifiers can appear either before or after a noun phrase, whereas English equivalents can appear only before a noun phrase (Example 1-5). a. John-i jogeum -eui chaek-eul satta. (1-5) John-[NOM] some-genitive particle book-[ACC] buy-past-indicative “John bought some books.” b. John-i chaek-eul jogeum satta. John-[NOM] book-[ACC] some buy-past-indicative “John bought some books.” Note: NOM= nominative case marker; ACC= accusative case marker

PAGE 24

10 Example 1-5a is similar to English except for the genitive particle eui. Example 1-5b is somewhat different, given that the quantifier jogeum is placed after the noun phrase as an adverbial quantifier. Korean has no equivalent for the English articles. English is the subject-predicate language; Korean is mainly a topic-comment language. Therefore, it may be useful to see how Korean performs the semantic and pragmatic functions that English performs by using articles. A major job of articles in English is to indicate new and old information, and thereby achieve cohesion in contexts. One way Korean does this is by using particles. In Korean, grammatical function is normally indicated by postpositional particles, the topic marker–(n)eun and the nominative case marker –i/ga. Consider the Example 1-6. a. Dodukeun bujeongjik-hada. (topic) (1-6) thief [TOP] dishonest be-[indicative] “ A thief, (s)he is dishonest.” b. Doduki geu-don-eul humchyeotta. (subject) thief [NOM] that money [ACC] steal-[past-indicative] “The/A thief stole the money.” Note: TOP= topic marker In Example 1-6a, -eun does not depend on content because its main function is to convey a generic statement. In other words, it can be used to describe the general nature of a thief. Example 1-6b (which includes –i) tells who stole the money. It can be translated into either the definite or indefinite article in English, based on the speaker’s intention. Thus, –i is more of a discourse-pragmatic marker than a grammatical marker. Although no Korean equivalent exists for English articles, one way for Koreans to represent English definite articles is to use the demonstrative “that” where either “that” or “the” is used in English (although this strategy is not applicable in every case).

PAGE 25

11 a. na-neun geu -saram-eul mannatta. (1-7) I [TOP] that person [Acc] meet-[past-indicative] “I met the (or that) person.” b. geu -saram-i geu don-eul humchi-n saram-ida. that person [SUB] that money [Acc] stole [ADJ] person be-[past-indicative] “That person is the person who stole that money.” Note: SUB= subject case marker; ADJ=adjective marker In Example 1-7a, the demonstrative geu “that” can be used for either “that” or “the” in English. However, this is not always the case in Korean. In Example 1-7b, no demonstrative is used before the second saram “person” in Korean, while “the” is required in English. Overall, Korean parallels exist for English demonstratives, possessives and quantifiers; but no Korean parallels exist for English articles. This makes it particularly difficult for Korean students of English as a foreign language (EFL) to learn the English article system. The second difficulty in learning articles is lack of communicative value in discourse, which leads to their lack of salience in speech. Because they have little or no lexical meaning, articles are rarely stressed in spoken English unlike many stress-timed languages; so listeners/interlocutors may not always attend to native speakers’ utterance of articles. For example, in the sentence, “I need a student to help me out” or “I need water to quench my thirst,” the articles are not crucial for understanding the situation and are thus often unstressed. Since it is difficult for learners of English to notice the form of articles in speech, earners may not observe models of article use. This is why it would be difficult for a non-native speaker (NNS) to learn the English article system through interaction alone.

PAGE 26

12 Finally, the semantic complexity of articles’ distribution in discourse makes it difficult for NNS to learn them. As we have seen, the use of the English articles – “a(n),” “the,” and “” – is governed by the semantic function of the noun phrase (NP) in discourse. The choice of articles is conditioned by following head nouns and complex semantic rules in discourse. Learning these choices is quite difficult for L2 learners. This semantic complexity may be why explicit teaching of English articles alone is relatively ineffective. Previous research on the acquisition of the English article (determiner) system Acquisition of English determiners, particularly articles has been studied. Hakuta (1976), in one of the few early studies on ESL acquisition of articles, investigated the presence or absence of articles in obligatory contexts while observing the acquisition of English morphemes by a Japanese child (e.g., past-tense marking and third-person singular indicative marking) who stayed in the United States for 2 years. The study was a 60 week longitudinal and non-experimental study when the child was 5 to 6 years old. Spontaneous speech was recorded every other week for at least 2 hours and later transcribed in orthography. Hakuta considered two types of errors in the participants’ article usage, termed “omission” (no use of an article in an obligatory context) and “commission” (the unnecessary use of “the” and “a”. Early stages of English learning showed frequent deletions of articles, and limited uses of the “the” and “a”. The participant did not have full control of the articles until much later in development. Huebner (1983) initiated a new direction of research on ESL article acquisition by using Bickerton’s (1981) noun classification system (discussed in the previous section). Using Bickerton’s two binary features, Huebner classified the semantic functions of the NPs. This classification of nouns made it possible to examine the article(s) that an ESL

PAGE 27

13 learner uses with each type of noun, and thereby understand the learner’s use of articles in the semantic context. Huebner’s (1983, 1985) findings suggest that his adult L2 learner, who was a native speaker of Hmong, initially overused the definite article “the” with almost all nouns in his speech. The amount of overuse gradually decreased in [-SR, -HK] situations, and his participant began to use “the” almost exclusively in [+SR, +HK] and [-SR, +HK] cases. However, because Hakuta’s (1976) findings were based on only one participant, we cannot generalize this progression of article development. While Hakuta (1976) and Huebner (1983) studied only one participant, Master (1988) analyzed the spoken interlanguage of speakers of five different native languages: three with no article system (Chinese, Japanese, and Russian), and two with article systems (Spanish and German). Data were elicited through informal interviews with four speakers of each language for a pseudo-longitudinal analysis of article usage. Participants whose L1 has no article system tended to overuse “the” in the [+SR, +HK] and [-SR, +HK] environments but not in [-SR, -HK] environments. Appropriate use of “a(n)” was delayed, compared with “the.” Based on these results, Master suggested that L2 learners might initially associate “the” with the feature of [+HK]. Thomas (1989) studied 30 adult L2 learners (24-46 years old) representing nine native languages, with the largest subgroups being speakers of Japanese and Chinese. Thomas analyzed article usage in the oral production of learners of English. Eight pairs of photographs were used for a picture description task: the pairs contained various scenes at various levels of complexity. Thomas’ findings led him to three observations regarding L2 article use: Whereas L1 children showed accurate use of “a” in [-SR, -HK] contexts at an early stage, the accurate use of “a” by adult L2 learners was delayed.

PAGE 28

14 The most common errors by L2 learners across proficiency levels was the overgeneralizaton of zero articles (i.e., deletion of articles). Both L1 child acquirers and L2 learners overgeneralized “the” in first-mention contexts ([+SR, -HK]) but not in [-SR, -HK] contexts. Trademan (2002) examining the impact of language transfer in acquiring English articles, analyzed the written language of two groups of 25 international students (at the university level) from two different language backgrounds: Spanish (with articles similar to English articles) and Japanese (which lacks articles). Data were collected from three academic essays over 6 to 8 months. A detailed discourse analysis considered the semantic features of hearer knowledge ( + HK) and specific reference ( + SR). He examined use, misuse, and deletions of English articles, and the role of genre and essay topic selection. The two groups showed some similarities in acquiring English articles: Integrating the definite article before the indefinite article A higher accuracy rate in sentence-level (structural) use of the definite article compared to discourse-level (textual) use of the definite article (e.g., “the” used with a first-mention noun versus “the” used with a noun already mentioned or related to the noun already mentioned) Generic use of English articles as contrasted with their specific use (e.g., “the Germans are diligent” versus “the book is mine”). The two groups also showed numerous differences in acquiring English articles, because of differences between Spanish and Japanese. This was particularly noticeable in generic use. Nevertheless, most Spanish speakers frequently used articles with low error in their final essay; Japanese speakers lagged behind, both in frequency and correct usage. Overall, Spanish speakers acquired the English article system more quickly than did the Japanese speakers, suggesting that L1 transfer is a major contributing factor in L2 acquisition of English articles.

PAGE 29

15 Simons (2001) was one of the few researchers to focus on acquisition of the entire determiner category. Data for this study were elicited from 30 native speakers of Korean and 20 native speakers of Spanish at four levels of L2 English proficiency. For data collection, a set of eight cartoon pictures and a jigsaw map task were used as instruments to elicit the participants’ oral production, and each participant’s performance was audio-recorded and then transcribed. After identifying all obligatory occasions from these data, all determiners (e.g., articles, demonstratives, possessives, and quantifiers) were examined to determine if there was a sequence of acquisition in the students’ interlanguage, and to determine the order of acquisition of the target language forms. Then the data from the Korean and the Spanish speakers were compared to identify potential influences of the L1 in the acquisition of the determiners. According to Simons (p. 33), the sequence of acquisition refers to the step-by-step acquisition of interlanguage structures (e.g., deletions of articles precede the overgeneralization of “the”); and the order of acquisition refers to the acquisition of final target language structures (e.g., the acquisition of possessives precedes that of articles). The sequence of acquisition for both L1 backgrounds had four general patterns: Deletion of determiners by the lowest proficiency learners Retention of determiners in inappropriate contexts Overgeneralization of “the” in all environments Appropriate use of “a(n)” in specific, referential, and indefinite contexts with “the” withdrawn from inappropriate contexts His proposed order of acquisition, which was based on the mean accuracy order, shows that determiners are acquired in the following order for both L1 backgrounds: 1. WH-determiners and demonstratives

PAGE 30

16 2. possessives 3. quantifiers 4. grammatically bare common nouns 5. null articles with proper nouns 6. cardinal noun combinations 7. the definite article 8. the indefinite article In each of these areas, the performance of the Koreans was worse than that of the Spanish speakers. This suggests some language transfer effects for the definite, indefinite, and null article as well as for cardinal noun combinations. In summary, deletion and retention of articles and overuse of “the” and “a(n)” were the primary measurements to determine L2 learners’ development of articles. Research showed that there may be an order of acquisition and a sequence of acquisition for English determiners. Specifically, simpler forms and those that exist in learners’ L1 may be learned earlier than those that do not. Additionally, although few differences were observed in these studies, given the variety of methods for data collection, type of data (e.g., essay writing, multiple-choice, oral or written interactional conversation) should be considered as variables in future research.

PAGE 31

CHAPTER 2 NOTICING IN SLA Noticing is an important concept in L2 research. Many L2 researchers claim that communicative instruction should involve systematic treatments to draw L2 learners’ attention to linguistic forms in order to develop communicative competence (Doughty & Williams, 1998; Izumi, 2003; Overstreet, 2003; Sharwood-Smith, 1981, 1991, 1993, 2000; Swain, 1985, 1993, 1995, 1998). Here is one difference between L1 and L2 learning: for L1 acquirers, detecting targeted features occurs seemingly without conscious awareness and is mainly guided by what Sharwood-Smith (1991) calls “internally generated” input enhancement; but for L2 learners, detecting them requires externally induced input enhancement. Internally generated input enhancement refers to automatically paying attention to targeted grammatical features by (L1) learners’ learning mechanisms (Sharwood-Smith, 1991). On the other hand, externally induced input enhancement refers to any act of mental activities for better noticing, such as textual enhancement and conscious awareness (or meta-awareness). Textual enhancement is, according to Doughty & Williams (1998), an implicit and unobtrusive means to draw the learners’ attention to forms contained in the written input, as a form of input enhancement. According to Sharwood-Smith (1991), L2 learning can be maximized by using externally induced input enhancement, whereas L1 learning or acquisition appears impervious to instruction. For this reason, it is important for SLA researchers and teachers to investigate ways to make 17

PAGE 32

18 targeted L2 features more salient for L2 learners (which, otherwise, would not be perceptually salient for them), by using explicitly induced input enhancement. My study dealt with externally induced input enhancement to investigate how it can be used to help L2 learners better notice and learn particular L2 grammatical features. Particularly, I examined effect of textual enhancement, chat activities, and meta-awareness of the targeted forms. As addressed in Chapter 1, the written nature of chat may enable students to process input (i.e., notice) and monitor and edit output through a self-paced learning environment (Lee 2001), while allowing learners to integrate into the discourse level and to develop awareness of variations of article use within communicative contexts thanks to its synchronous, interactive nature (Pica, 1983). Therefore, the next section discusses three perspectives of noticing. Then I discuss three issues relevant to noticing: modality (e.g., oral versus written), textual enhancement, and the role of output in noticing. I then discuss how chat can be used for noticing L2 function words. Research questions were proposed based on this review of literature at the end of the chapter. Three Perspectives on Noticing I reviewed three perspectives on noticing by Schmidt (1990, 1993, 1994, 1995, 2001), Tomlin and Villa (1994), and Robinson (1995) to help to understand the role of attention in language acquisition. Schmidt (1990, 1993, 1994, 1995, 2001) and Schmidt & Frota (1986) proposed the Noticing Hypothesis, which claims that “noticing is the necessary and sufficient condition for the conversion of input to intake for learning” (Schmidt, 1994:17). In my study, intake was defined as “that part of the input that has been attended to by L2 learners while processing the input” (Leow, 1993:334). Schmidt (1995) claimed that subliminal language learning is impossible, saying subliminal effects

PAGE 33

19 are shown only when very familiar stimuli are presented, activating already established mental structures (p. 24). Schmidt added that intake is what learners consciously notice, and the requirement of noticing applies equally to all aspects of language: lexicon, phonology, syntax, and pragmatics. For this reason, “in order to acquire phonology, one must attend to phonology; in order to acquire pragmatics, one must notice both linguistic forms and the relevant contextual features” (Schmidt, 1993:209). His position gives potential implications for future research and L2 classroom instruction in that L2 learning effect can differ by whether L2 learners consciously attend to L2 forms or not. Tomlin & Villa (1994), on the other hand, argued that conscious attention is not a necessary component of attentional processes in SLA. They defined awareness as “the subjective experience of any cognitive content or external stimulus” (p.194), while attention is “the process of selecting critical information for further processing” (p.187), and attention has three principal components, namely alertness, orientation and detection. Tomlin & Villa (1994) defined “alertness” as an overall, general readiness to deal with incoming stimuli or data. They claimed that as alertness increases, the rate at which information is selected for further processing also increases. Alertness is of general import to SLA in that learners usually have to be ready to process information. “Orientation” was defined as “the attentional process responsible for directing attentional resources to some type or class of sensory information to the exclusion of others” (p.191). They claimed that this component of attention is of greater importance to SLA than alertness, in that when attention is directed to a particular piece of information, detection, which was defined as “the cognitive registration of sensory stimuli and the

PAGE 34

20 process that selects or engages a particular and specific bit of information” (p.192), is generally facilitated. Regarding detection, the third function, Tomlin & Villa (1994) stated that once information is detected, further processing of that information is possible. Thus, detection is the necessary and sufficient condition for intake. However, contrary to Schmidt’s view that learners must notice or be consciously aware of features of input in order for intake to occur, they postulated that awareness can be dissociated from attention. They gave an example from cognitive psychology research in which subliminally presented items can cause semantic priming. In this case, priming is that activation of one lexical items spreads to related items (e.g. phonologically or semantically related), assuming the mental lexicon is a network of related lexical items. They claimed that attention, or more specifically the attentional function of detection, is necessary for SLA, while awareness is not. As for the third perspective, Robinson (1995) attempted to reconcile Schmidt’s and Tomlin & Villa’s positions. Robinson assigned detection to a stage in the learning process that occurs ahead of noticing. Schmidt’s concept of noticing was redefined in Robinson’s model as “detection plus rehearsal in short term memory, prior to encoding in long-term memory” (p.296). According to Robinson, this process of detection plus rehearsal is the one that does require some level of awareness in order for learning to take place. Although these three perspectives are generally accepted as the main perspectives for SLA research, more research may need to be conducted on attention and awareness in SLA toward more practical directions. Simard & Wong (2001) noted that what is more

PAGE 35

21 fruitful for research is to examine how different levels of attention/awareness may impact learning rather than to ask whether attention/awareness are necessary or not for SLA. Like their suggestion, future research needs to design conditions and tasks that engage different levels of learners’ attention and awareness for this purpose, although there are difficulties in operationalizing attention and awareness. Modality as a Variable for Noticing Modality was considered a variable for noticing of L1 or L2 forms by many researchers (e.g., Danks, 1980; Greenslade et al., 1999; Johnson, 1992; Johnson & Newport, 1989; Leow, 1993, 1995; Lund, 1991; Murphy, 1997; VanPatten, 1990; Wong, 2001). L1 research investigated the differences between reading and listening to texts, and this line of research often suggests advantages for the written mode over the aural mode in terms of noticing of particular L2 features. As compared with listeners, readers have complete control over the amount and rate at which they process visual input, because written input is constantly available for reprocessing. Listeners, on the other hand, do not have this control because speech is continuous and auditory signals fade rapidly (assuming that they are processing aural input from natural speech rather than from audio recordings) (Danks, 1980). In this section, we discuss these differences with respect to findings from cognitive psychology research, and we also examine the effect of such differences in the context of L2 learning. This difference in the two modalities can be seen as related to attention capacity. Much of the discussion in SLA on attention capacity was initiated by Kahneman’s (1973) capacity model of attention. According to Kahneman, capacity theories of attention are based on the fact that a human being’s capacity is limited while performing mental work, and that capacity is drawn from a single resource pool, and requires various degrees of

PAGE 36

22 attention or effort. When it is necessary to perform more than one task, the success of dual-task performance depends on the specific task demands of the two activities. If the demands exceed available capacity, interference occurs. This interference results from competition between task demands for limited attentional resources. Kahneman’s model was extended by Wickens (1980, 1984, 1989), who proposes multiple resource pools within three systems: “(1) perceptual/cognitive activities versus response activities, (2) analog/spatial versus verbal activities, and (3) auditory versus visual activities (or vocal versus manual responses)” (Wickens, 1989:82). During dual task performance, task demands, which are globally similar (i.e., that both engage the same system), compete for the available processing resources of that system and one or both of the tasks suffer interference. For example, listening to music is easier while looking at a picture than while listening to radio news. This is because, in the latter, similar task demands (i.e., auditory) compete with each other. Modality is a factor that can decrease the demand on resources. For example, in Johnson (1992), the issue of modality was addressed as a factor that might influence native Chinese and Korean ESL participants’ performances on grammaticality judgment tasks. Johnson’s study was originally intended to examine if there would be any relationship between a critical period effect and modality effects. While Johnson & Newport (1989) used an aural grammaticality judgment task, Johnson (1992) used the same participants but employed a written grammaticality judgment task. Johnson found that the written results for the adults were significantly higher than their previous aural results. From this finding, she speculates that adult L2 learners may have grammatical knowledge that is only available in performing written tasks.

PAGE 37

23 Leow (1993, 1995) investigated the effects of simplification in two different modalities (i.e., written input (1993) versus aural input (1995)). In both studies, the participants were exposed to one of the following four conditions: a simplified or unsimplified reading (or listening) passage with the present perfect tense form or a simplified or unsimplified reading (or listening) passage with the present subjunctive form. Leow found that L2 learners of Spanish receiving aural input are not able to process the present subjunctive as well as those receiving the same written input, and noted the importance of the effect of modality on comprehension and input processing. More recently, Wong (2001) conducted an experiment to investigate how processing input for meaning and forms of low communicative value varied in aural and written modes. In his study, while trying to comprehend the main meaning of a passage, the listeners and readers were required to attend to a function word (i.e., the definite article) or a lexical item which was crucial for comprehending the referential meaning of the passage. Recall protocols were employed for scoring based on the number of idea units recalled, and mean recall scores were computed for each group. The results showed that attending to a lexical item had little negative effect on comprehension in either mode. On the other hand, attending to a function word had a detrimental effect on comprehension in the aural mode, whereas it had none in the written mode. Wong suggests that, in the written mode, attending to function words was not very difficult, because written input was neatly segmented and readers had the advantage of assuming greater control over the input (i.e., rereading, skimming, and focusing on one specific feature of input over others). Conversely, in the aural mode, form and meaning compete

PAGE 38

24 for learners’ limited attentional resources during input processing, which results in learners’ preference for processing meaning before form. However, although these studies suggested the advantage of the written mode over the aural mode for noticing targeted features, other studies show different results. For example, Greenslade et al., (1999) conducted research in the written mode with the same design as Wong (2001). His findings showed that when learners’ attention was directed to a lexical item, there was no negative effect on comprehension of the reading passage, but when directed to attend to the forms of low communicative value, students had difficulty in comprehending the referential meaning of the reading passage. Given the nature of reading, where the main purpose is normally comprehension of content, it may be hard for the reader to attend to specific forms unless intentional attention is placed on these forms. For this reason, textual enhancement was often employed in SLA research in order to draw the reader’s attention to form during reading, though this may comes at the expense of comprehension. The Effect of Textual Enhancement on Noticing As discussed above, “externally induced” input enhancement refers to any attempt by instructors to focus the learners’ attention on the formal properties of language by deliberately making those properties salient. Sharwood-Smith argued that externally created salience can facilitate the development of L2 knowledge, although it is possible that it does not have any consequent effect on learning for other reasons, such as developmental readiness. According to Pienemann (1984, 1998), some grammatical features are developmental, that is, learners acquire those features in a clear developmental sequence, and the order of the sequence cannot be altered or skipped by instruction. Thus, instruction can only successfully affect the acquisition of

PAGE 39

25 developmental features when a learner is approaching the final stage of the sequence and is psycholinguistically ready for acquiring them. Nevertheless, Sharwood-Smith believed that externally created salience could increase learners’ chances of detecting the target form or structure in the input to which they are exposed, and hence may increase their chance to acquire the form or structure. As defined earlier, textual enhancement is a form of input enhancement, and an implicit and unobtrusive means of drawing the learners’ attention to form contained in the written input (Doughty & Williams, 1998). The goal of textual enhancement is to provide salient targeted features of written input that would otherwise not be perceptually salient. The idea behind this technique is that by increasing the perceptual salience of particular features of L2 written input using typographical cues, learners will pay attention to these features and select them for intake. The method of textual enhancement is simply increasing the perceptual salience of the target form via combinations of various formatting techniques (e.g., bolding, capitalizing, or underlining), which may sometimes be accompanied by an explicit mention to the learners to attend to the highlighted form (Izumi, 2002). Empirical studies demonstrated that increasing the perceptual salience of specific features of language through textual enhancement may be beneficial to learners in that it may help them to pay more attention to form, which, in turn, can lead to more learning. For example, Rosa & O’Neil (1999) investigated how intake was affected by both awareness and the conditions under which a problem-solving task was preformed as means of textual enhancement. Think-aloud protocols were employed to assess level of awareness. The participants gave an indication that focal attention was being directed

PAGE 40

26 toward the target item. Intake of the target structure was assessed through a multiple-choice recognition task in which learners had very limited time to decide which of four verb forms best completed each sentence; the 22 items included in the pretest and in the posttest had to be completed in a maximum of 6 minutes. Results indicate that the higher the level of awareness students demonstrated, the stronger the effect on intake was. More recently, Overstreet (2003) examines the effects of textual enhancement on L2 learner reading comprehension and form recognition based on two primary variables: the amount of enhancement and the relative communicative value of the target. While his previous study (Overstreet, 1998) enhanced the entire verb, Overstreet (2003) adapted two enhancement conditions: whole word and morphology (i.e., inflective marker) only. As for the communicative value of the target, the progressive and the imperfect subjunctive were used as high communicative value and low communicative value respectively. A total of five conditions were created: no enhancement; morphology enhancement /progressive; whole word enhancement/progressive; morphology enhancement/imperfect subjunctive; and whole word enhancement /imperfect subjunctive. Then, each of 109 participants in third-semester Spanish read a text under one of these five conditions, and then completed a free recall task and a form recognition task. Results of the free recall task suggest that neither of the main variables (i.e., the amount of enhancement and the communicative value of the target) had an effect on global recall, while enhancement of the entire word increased recall of target sentences. Results of the form recognition task suggested that amount of enhancement had no effect

PAGE 41

27 on form recognition, whereas relative communicative value did. The items of higher communicative value were better detected and identified by learners than the items of lower communicative value. These results showed that textual enhancement of a form of higher communicative value makes it possible to process both form and meaning, while that of lower communicative value limits learners to processing form only, or primarily. Leow (2001) and Leow et al., (2003) designed empirical studies to test the premise that textual enhancement induces learners to attend to the target item, which in turn promotes further processing of the target form. The targeted linguistic forms were the Spanish present subjunctive and present perfect (Leow, 2001) and the formal imperative/command in Spanish (Leow et al, 2003).They collected concurrent think-aloud protocol data from the learners while they were exposed to either enhanced or unenhanced input. To measure participants’ intake of targeted forms, a multiple-choice recognition task was employed. The results show that there were no significant benefits of enhanced input over unenhanced input for amounts of reported noticing of the Spanish formal imperatives, readers’ comprehension of text content, and readers’ intake as measured by the recognition task. To explain these differing results from Rosa & O’Neill (1999), Leow (2001) and Leow et al., (2003) suggested that several variables could be responsible, such as task type, type of targeted linguistic items, participants’ proficiency level in the target language, number of participants in each cell, etc. As for task type, Leow (2001) and Leow et al., (2003) employed reading tasks rather than problem-solving tasks as in Rosa & O’Neill (1999), who reported higher levels of noticing during tasks. According to Leow et al., (2003), in a problem-solving puzzle task, learners generally deal with the

PAGE 42

28 targeted forms item by item in order to complete the task, while in the reading passage used in Leow (2001) and Leow et al., (2003), the learners processed the targeted forms more holistically at a sentential, paragraph, or even discourse level rather than item by item. This is because of the nature of reading, whose main purpose is comprehension for content rather than attending to form. For this line of research, Izumi (2002) suggested three factors helpful in identifying directions for future research on input enhancement. First, the employment of a noticing measurement is critical to future studies, given the importance of attention in learning but the uncertainty regarding the link between input enhancement, noticing, and learning. Second, whether input enhancement alone can induce desired learning effects needs to be confirmed, excluding intervening variables. Finally, Izumi emphasized the learnability of forms because the treatment effect could be constrained by the learner’s developmental readiness. Accordingly, it is important to be sure that the learners are indeed psycholinguistically capable of learning the targeted form. In sum, through textual enhancement, learners are induced to focus on the formal properties of an L2. Simultaneously, however, various intervening factors (e.g., task types, targeted items, and individually different levels of meta-awareness) are involved in this line of research, making it difficult to make direct comparisons among them. Particularly, given the nature of reading, where the main purpose is comprehension of content rather than attention to form, merely being exposed to enhanced material may not guarantee that the learner attends to form. For this reason, as Leow (2003) suggests, “meta-awareness is necessary to promote further written production of the targeted forms, and noticing in discourse may not necessarily contribute to more profound processing of

PAGE 43

29 grammatical information, beyond intake, unless accompanied minimally by a level of meta-awareness” (p. 505). One of the best ways to promote L2 learners’ meta-awareness may be the act of output, which researchers consider helpful for noticing L2 forms, as discussed in the following section. The Role of Output in Noticing Without denying the role of input, research suggests that there is also a need for output in SLA in learning an L2 form (e.g., Erlam, 2003; Izumi, 2002, 2003; Swain, 1985, 1993, 1995, 1998; Swain & Lapkin, 1995). Swain (1985) initiated the issue of output in SLA, arguing that one possible important cause of L2 grammatical errors among immersion learners may be their insufficient language production. Swain’s research suggested that the teacher-oriented interactions in French immersion classrooms permit few students to produce utterances longer than a single clause and, consequently, to acquire complete knowledge of the language. This is because they do not have to produce accurately; in other words, they are rarely pushed to be more accurate. From these observations, Swain concludes that comprehensible input, although invaluable to the acquisition process, is by itself insufficient for learners to fully develop their L2 proficiency, and that comprehensible output is additionally necessary if learners are to be accurate as well as fluent in the target language. The focal point of comprehensible output is that when learners face communication difficulties, they will be pushed into making their output more precise, coherent, and accurate. Swain (1985) claimed that “producing the target language may be the trigger that forces the learner to pay attention to the means of expression needed in order to successfully convey his or her own intended meaning” (p. 249). Swain claimed in her Output Hypothesis that output contributes to second language acquisition in several ways

PAGE 44

30 (Swain, 1993, 1995, 1998). First, fluency can be enhanced through practice. Practice helps the learner reach automatization of the target forms, which results in fluency. Second, output promotes the “noticing/triggering” function. This function of output is consistent with the ideas expressed in Schmidt’s Noticing Hypothesis, which claims that acquisition of a target-like form is possible if and only if it is consciously noticed. The role of output, then, is to facilitate the process of noticing problems in one’s interlanguage output as well as the relevant features in the input (i.e., the target language). Third, producing output could be a way of testing target language hypotheses. By producing output, learners can be more accurate in their production by being pushed in the process of negotiation of meaning. In this way, output makes it possible to test whether one’s interlanguage is comprehensible or linguistically well formed. Lastly, through their reflection on their own language, learners can control and internalize linguistic knowledge, which serves a metalinguistic function. Output could induce the learner to engage in syntactic processing in addition to comprehension of input. This syntactic processing may be followed by modified output, which may lead to language acquisition. Swain (1995) proposes a Vygotskian perspective on language learning in her discussion of the functions of output as follows: “According to Vygotksy, cognitive processes arise from the interaction that occurs between individuals. That is, cognitive development, including presumably language development, originates on the inter-psychological plane. Through a process of appropriation, what originated in the social sphere comes to be represented intra-psychologically, that is, within the individual.” (p. 135) In other words, the act of output (i.e., interpersonal interaction) facilitates the process of intrapersonal interaction. This perspective differs radically from Krashen (1994), who proposed the Input Hypothesis to account for the fundamental similarities

PAGE 45

31 between children’s L1 acquisition and adults’ L2 learning and who seems to miss the essence of the comprehensible output hypothesis (although his was not to evaluate output). Swain’s output hypothesis was empirically investigated in a number of studies. Izumi (2002) investigated if output and visual input enhancement can facilitate the acquisition of English relativization by adult ESL learners. With the systematic grouping of output (O) and input enhancement (IE) (i.e., output only group, output plus input enhancement group, input enhancement only group, and control group), a text reconstruction task was used for the presentation of target input materials, since the text reconstruction task can promote comparisons between interlanguage output and target language input. Four different kinds of written tests were used. A sentence combination test and a picture-cued sentence-completion test examined the participants’ productive knowledge, while an interpretation test and a grammaticality judgment test were used to investigate their receptive knowledge. Results of the study indicated that the output plus input enhancement group outperformed the input enhancement only group; and the input enhancement only group did not evidence measurable differences in their learning. These findings suggested that visual input enhancement alone may be insufficient to promote grammar learning, especially for a complex structure like relative clauses, and pushed output helps learners to detect and learn targeted forms. In addition, Erlam (2003) investigated the relative effects of structured-input instruction and output-based instruction on the acquisition of direct object pronouns in French (excluding en and reflexives). Three classes of 70 secondary school students were assigned to three groups: structured-input instruction, output-based instruction, and

PAGE 46

32 control. The structured-input group received explicit instruction regarding the key grammatical items (e.g., direct object noun and direct object pronoun), and the output-based instruction group received additional form-focused activities that allowed students time to think and apply rules they have learned. Then, students were assessed on listening comprehension, reading comprehension, written production, and oral production tasks. The results showed that the output-based instruction group gained more overall in the tests. However, the results of some studies did not support these findings. Koyanagi (1998) examined the effects of three types of classroom learning tasks, comparing focus on form versus focus on forms, pushed output versus input processing, and mechanical-drill output versus task-based meaningful output. The subjects were 30 learners of Japanese as a foreign language at the college level, and the target structure was the conditional expression to. There were three experimental groups: Input Group (performed input processing tasks); Output Group (performed both input processing tasks and output processing tasks); Drill Group (engaged in audiolingual pattern practice). They attended six 50-minute instructional sessions, and the Control Group received no instruction. All of the participants took a pretest and two posttests, which consisted of grammaticality judgment, listening comprehension, oral production, and written production tests. The results revealed that the Input and Output Groups outperformed the Control Group; the Input and Output Groups outperformed the Drill Group only on the delayed posttest; and there was no difference between Input and Output Groups or between Control and Drill Groups. The results confirmed that communicatively meaningful tasks

PAGE 47

33 that direct learners’ attention to form are beneficial in learning, although, in close examination of the amount of pushed output, Koyanagi found that it was quite limited. While she did not find direct support for the hypothesis, Koyanagi (1998) argued for the importance of output in order to develop automatized access to metalinguistic knowledge. Horibe (2004) also investigated the validity of Swain’s output hypothesis, but did not find a significant advantage for output. Horibe focused on the acquisition of Japanese temporal subordinate conjunctions by American adult learners of Japanese as a foreign language, and compared two types of instructional treatments: input only and input plus output conditions. The conditions were compared in order to determine the effects of opportunities for output on the acquisition of the target forms. Three intact classes were assigned to one of three conditions: an input only group, an input plus output group, and a control group. After a brief grammar explanation, the learners in the input only group engaged in comprehension activities of reading and listening materials containing the target structures, while the learners in the input plus output group received opportunities to produce the target structures both in written and oral modes, in addition to comprehension of the input, as in the input only group. The control group worked on materials unrelated to the target forms. The results of the tests indicated that there was no statistically significant difference between the input only group and input plus output group in terms of the acquisition of the target forms, although both experimental groups outperformed the control group. Horibe suggests that the non-significant difference between the experimental groups may be explained by a cognitive overload for the learners in the input plus output group, who had to produce the target structures in addition to processing the input.

PAGE 48

34 Through the review of research on output in SLA, it appears that the evidence in favor of output is not entirely convincing. One possible explanation is that these studies employed different modalities (e.g., oral or written), task type and targeted grammatical items. Izumi (2002) employed the a reconstruction task for students’ written production targeting English relativization, whereas Horibe (2004) used both written and oral production targeting temporal subordinate conjunctions. These variables (i.e., modality, task type, and targeted item) should be considered as variables for future research. In the following section, a review on the characteristics of chat is presented, and how chat can be useful for noticing and producing L2 function words (e.g., English determiners) is discussed. Given its unique nature, it seems worth conducting an empirical study on chat for noticing L2 grammatical features. Use of Chat for Noticing L2 Function Words Characteristics of Chat Chat is a kind of synchronous communication in which messages are exchanged in real time. Chat can be considered as sharing characteristics of both spoken and written languages, because it is conducted in the written mode while it also has a synchronous, interactive nature like spoken interaction, and it is often conducted in colloquial language/registers. However, despite its resemblance with oral discourse and written texts, chat also has its own unique properties, which rarely appear in either spoken or written forms. Chat is characterized as more discursive in nature than regular conversation, in that no overlaps and interruptions are possible while an utterance is being produced (i.e., typed). For this reason, successive and independent speech acts are simply juxtaposed, and different topics can be interwoven. Specifically, Negretti (1999) noted that a chat

PAGE 49

35 discussion does not have to follow the linear and sequential pattern of a normal face-to-face conversation. In a chat session, one person can receive replies to a question even after the conversation has shifted to a new topic. Depending on the number of people involved, the conversation can develop into multiple strands, with various interactions being carried out simultaneously. From a pedagogical standpoint, this feature of chat can be detrimental to interaction, as in the case where many mini-conversations develop at once, leaving out one of the participants, who in frustration chooses to leave the chat session or stops participating. In addition, there is an abundant use of addressivity and abbreviation. In the chat conversation in which many people participate, it is not always easy know to whom a message is aimed because of the absence of gestures, facial expressions, eye contact, etc. For this reason, chatters have a greater tendency to use people’s name (or user ID) when the message is sent. In other words, participants use a greater degree of addressivity to avoid ambiguity and discontinuity of exchanges or turn-takings. The frequent use of abbreviation is due to a combination of spatial, temporal and social constraints, such as screen size, average typing speed, minimal response time, and competition for attention (Werry, 1996). While many of these characteristics of chat could also occur in face-to-face conversation, they are undoubtedly much less likely than in chat mode. Recent research investigated how chat can build learners’ communicative and grammatical competences (e.g., Blake, 2000; Kitade, 2000; Lee, 2002; Pellettieri, 2000; Salaberry, 2000). These studies suggested that the chat session could be a useful tool for the second language classroom. During chat sessions, learners have comparatively more time to reflect on the target language while composing a message than during oral

PAGE 50

36 interactions, and there is less pressure to perform in front of people. Nevertheless, because of its synchronized, real-time nature, the learner is forced to respond in a timely fashion. Perhaps this fact – it is synchronous but with slightly more time – is why chat interactions tend to contain more negotiation of meaning than oral communication, as shown by Pellettieri (2000). Potential of Chat for Developing L2 Function Words As discussed earlier, the lack of salience of articles in speech and the semantic complexity of their rule systems in English can pose a learning problem for NNS. English articles might not be phonologically salient enough for NNS to receive models from their interlocutor in spoken interaction, and the semantic complexity of their rule systems seems to make it difficult for learners to understand and hold a control over the usage of the system. Therefore, both oral communicative methods and explicit teaching methods might entail limitations for learning the English article system. Given these problems with oral communicative methods and explicit teaching methods, the written yet synchronous, interactive nature of chat may help to overcome these limitations. The written nature of chat might reduce inherent problems of communicative methods for learning function words. Thanks to the visual (i.e., written) property of the chat mode, comparatively more time than in face-to-face conversation enables students to process input and monitor and edit output through a self-paced learning environment (Lee, 2001, 2002). As discussed earlier, the written mode seems to have an advantage in processing input because written input is constantly available for reprocessing (Danks, 1980). Bump (1990) reported that the visual/written nature of his program (i.e., Interchange) was praised by his participants, one of whom said “on the whole, it also makes us think more and analyze more and it leaves us up to our own writing abilities”

PAGE 51

37 (p.56). Additionally, research indicated that the written mode likely has more advantages in attending to input than does spoken language (e.g., Chapelle, 2001; Skehan, 1998b). Chapelle suggested that “one of the potential contributors to manipulation of learners’ attention in learning tasks is modality (i.e., spoken or written), and written communication typically affords more opportunity for attention to form, whereas spoken language often occurs to achieve fluency” (p. 49). Chapelle’s position was based on the cognitive advantage of written input, as suggested in Skehan (1998b). The synchronous, interactive nature of chat allows learners to be integrated into the discourse level. Pica (1983) claimed that it is impossible to master the English article system through textbook and classroom practice alone, because the linguistic information that the student needs to use and interpret English articles is often discourse-related. As a result, “a key to ESL students' attaining proficiency in their use should be through developing awareness of variations of article use within communicative contexts” (p. 231). Compared with simple output practice, as in essay writing or oral presentation, chat conversation might provide the learner with various contexts where they can practice determiners, and, accordingly, could make it possible to learn this semantically complex system in an implicit way. In addition to these potential benefits from chat, we need to be reminded of the second function of output as discussed earlier: output promotes the “noticing” function (Swain, 1993, 1995, 1998). In accordance with Schmidt’s Noticing Hypothesis, the act of output during chat could help a student facilitate the process of noticing problems in his or her interlanguage output as well as the relevant features in target language input.

PAGE 52

38 Despite these positive aspects of chat, however, there exist some potential problems with chat for acquiring target language function words. Given that the main purpose of chat in classrooms is often no more than communication itself, chatters may want to save time and effort by minimizing the amount of typing, and thus often omit less crucial words for communication, like function words. For this reason, students’ meta-awareness and task types need to be considered in order to maximize the effect of chat. First, students need to be encouraged to attend to and correctly use the target items during a chat conversation. As Leow (2001) suggests, a minimal level of meta-awareness may be additionally required to contribute to more profound processing of grammatical information. Additionally, Schmidt (1993) defined attention as one form of consciousness and explicitly claims that learners must consciously pay attention to or notice input in order for L2 data to become intake. His position gives potential implications for my study in that L2 learners’ noticing during chat, and the consequent effect, can differ according to whether L2 learners consciously attend to L2 forms or not. Students’ meta-awareness of (or conscious attention to) targeted L2 forms may be possible during a chat activity because, as shown in previous research (e.g., Johnson 1992; Lund 1991; Murphy 1997; Wong 2001), attending to function words in the written mode would not negatively affect the communication as much as in oral conversation. In other words, students need to be aware of the difference between chat conversations for educational purposes, as in my study, and casual chat, perhaps at the expense of the authenticity of this tool. Second, the types of tasks used in the chat activity also need to be considered, given that task type was considered a variable in SLA and CALL research (e.g., Pellettieri, 2000; Pica, 1989, 1994; Skehan, 1996, 1998a, 2003). For example, Pellettieri

PAGE 53

39 (2000) suggests that the communication task is an important factor for the quantity and type of negotiation, and that it needs to be goal-oriented so that possible outcomes can be minimized. According to her, this minimized outcome causes form-focused interaction, and promotes the noticing of problematic linguistic structures, and thus plays a significant role in the development of grammatical competence. We reviewed that when combined with proper input sessions, output of the target language may help learners notice the gap between the TL and their interlanguage, although not all research supports this claim. As discussed in this section, chat has a unique nature for L2 learning practice, and its uniqueness might lead to the potential of L2 learning. Therefore, it is worthwhile attempting to investigate if chat can be an appropriate place for L2 output, and to determine the effect of the combination of textual enhancement and chat on noticing L2 targeted forms. Research Questions General Research Question Is chat an appropriate forum to ask learners to focus on and use determiners and other function words? Through the review of literature, we learned that chat has many positive aspects for developing L2 function words because of its written while synchronous, interactive nature. At the same time, given its other characteristics (i.e., it is telegraphic in form; it contains frequent omission of function words, etc.), there are concerns about its utility for this purpose. One way to optimize the effect of chat might be to induce students to use determiners rather than merely omitting them. Therefore, by answering this general research question, we will learn whether or not chat, when it is appropriately used for educational purposes, can be a place where students can attend to and practice English

PAGE 54

40 determiners. In order to answer this general research question, the following two research questions were formulated: Research Question 1 How will treatment type and meta-awareness affect students’ performance, as reflected in the use of determiners in the pretest and posttest? This research question was motivated by a desire to investigate how differential treatment types (i.e., input enhancement only, chat only, and the alternation of input enhancement and chat) and students’ self-rated differential meta-awareness of the targeted L2 forms during treatment result in the choice of English determiners in the posttest. This will be compared with the pretest for a measurement of students’ performance. By analyzing students’ errors in the preand posttests, it is possible to learn in what condition(s) students notice the target structures better. Research Question 2 How will treatment types, meta-awareness and task type affect students’ performance, as reflected in the use of determiners during chat? This research question aimed to determine how to help students notice L2 forms better in chat mode. Despite the potential of chat as a tool for L2 learning, chat also has unique characteristics such that participants in chat conversations may often omit less meaningful words like English articles unless omission interrupts communication between chatters. For this reason, we needed to determine if it was possible to encourage students to focus on and use determiners, so that students could notice the targeted forms during a chat activity, which we believe would result in the acquisition of these forms. Originally, my study did not look at task type as a variable and, therefore, three different types of task: information gap (e.g., spot-the-difference), jigsaw (e.g., picture-story-telling) and discussion task, were employed with no particular

PAGE 55

41 methodological consideration. However, given that task type was considered a major factor in many studies (e.g., Loschky & Blev-Vroman, 1993; Pellettiere, 2000; Skehan, 1996, 1998a, 2003), it is worthwhile looking at how students’ use of English determiners varies during different types of chat activities.

PAGE 56

CHAPTER 3 METHODOLOGY The previous chapter presented a review of the research on the determiner system and noticing in SLA, and generated research questions in the basis of this review. This chapter presents the methodology of the study designed to test these research questions. Participants Participants were 64 Korean students from seven English classes at Inha University in South Korea (Originally, 129 students participated at the beginning of the experiment, but only 64 students successfully finished all the assignments required for the experiment). The seven English classes consisted of four classes of English for College Students 3 (ECS3) and three classes of Practical English Grammar (PEG). ECS3 was the third in a four-course sequence (ECS1~ECS4) necessary for all students at Inha University as a requirement for graduation. Each ECS course consisted of two weekly conversation classes with a native speaker of English and one weekly reading class with a Korean teacher. The PEG course was open to all the students who have taken ECS2 or beyond. Since PEG was not a course that is required for graduation, it is possible that the students in PEG might have had a higher level of motivation and proficiency than those in ECS3, although the present study does not look at this potential difference in motivation as a variable. All the participants had taken at least six credits of English prior to the semester of their participation in the project (i.e., ECS1 and ECS2), and therefore they were assumed to be at the intermediate (or higher) level at the time of testing, with minimum conversation skills required for my study. 42

PAGE 57

43 There was one control group of 16 students and three experimental groups of 16 students according to their treatments: the enhanced input group (Group E), the chat only group (Group C), and the group of input enhancement plus chat (Group EC). The control group consisted of 16 students from ECS3 (4) and PEG (12), and the experimental groups were made up of 48 students from three ECS3 classes (21) and two PEG classes (27). The students in the experimental groups were randomly assigned to one of the three groups, and each student in Group C and Group EC was randomly paired with another student for the chat part of the study. All the participants were native speakers of Korean. It was not necessary to control for potential dialect differences, for these are primarily phonological and all dialects of Korean exhibit a uniform determiner system that lacks articles. Intact classes were used, and no particular effort was made to control for the numbers of males and females. Table 3-1. The participants’ demographic information Group E Group C Group EC Control Total Number of Participants 16 16 16 16 64 Mean 22.9 23.6 21.5 23.8 Age Range 19-25 20-28 19-25 20-26 Male 8 11 6 12 37 Gender Female 8 5 10 4 27 ECS3 3 8 10 4 25 Class PEG 13 8 6 12 39 Target Structure To maximize the effects of learning articles, the target structure was limited to the core-determiner, which includes articles, but not preand post-determiners. As a result, the target structures of my study include articles (e.g., , a(n), the), possessives (e.g., my, your, his, her, its, their, -’s), demonstratives (e.g., this, that, these, those), and some quantifiers (e.g., some, any, no, each, every, either, neither, enough). The quantifiers in

PAGE 58

44 either preor post-determiners and cardinal/ordinal numbers are excluded in my study (e.g., all, both, half, many, much, (a) few, (a) little, several, more, less, most, least, two, three, first, second, third). For exceptions, “one” was considered an indefinite article rather than a cardinal number in my study, because it can often be replaced by “a(n)” as in “I have one/a friend.” Also, “another” was considered to have an indefinite article since, in some cases, students were confused with “the other.” Table 3-2. Target structures of study Non-target structures Target structures Non-target structures Quantifiers: all, both, half Multipliers: twice, three times, etc. Articles: , a(n), the Possessives: my, your, his, her, its, their, our, -’s Demonstratives: this, that, these, those Quantifiers: some, any, no, each, every, either, neither, enough, one, another Cardinal numbers: two, three, etc. Ordinal numbers: first, second, third, etc General ordinals: next, last, etc. Quantifiers: many, much, (a) few, (a) little, several, more, less, most, least Phrasal quantifiers: a great deal of, a lot of, a good number of, etc. Instruments and Materials The pretest and posttest were each a set of multiple choice cloze tests in a story of several paragraphs, in which the participants were asked to select the appropriate core-determiner from four possible choices. Multiple-choice cloze tests were employed rather than open ended cloze passages to avoid any potential situations where more than one possible answer could lead to confusion. Each question of the multiple-choice tests included four examples, which consisted of , “a(n),” “the,” and another core-determiner (Appendix C).

PAGE 59

45 Each text for the input enhancement session consisted of passages of a fairy tale, which were written at a relatively low proficiency level so that various levels of readers could enjoy the story. For the input enhancement session, all core-determiners and the following head nouns in the passages were highlighted, and each of the core-determiners was additionally underlined. Nouns with no salient core-determiner were considered to follow the zero article (), which was also highlighted and underlined. The materials for pretest and posttest and the input enhancement session were adopted from Peters (2001) (Appendix D). The materials for chat tasks included information gap tasks (e.g., spot-the-difference), jigsaw tasks (e.g., picture-story-telling), and discussion tasks. An information gap task is a task wherein one person has certain information that must be shared with others in order to solve a problem, gather information or make decisions. Jigsaw tasks are tasks wherein subgroups of learners in a class are asked to read or listen to different information concerning a particular topic or situation. The full picture is then pieced together, like a jigsaw, when the groups combine in discussion to complete the task (Appendix E). The pictures for story-telling activities were adopted from Yule (1997) and The College Board, and the pictures for spot-the-difference were from Cromer Lifeboat Station and RockyRamber.co.uk. MSN Messenger was used by the participants for the chat sessions of my study. The MSN Messenger program enables people to send instant text-based messages in real time to other members that are logged on. MSN Messenger is freely downloadable from the MSN Messenger website ( http://messenger.msn.com/ ).

PAGE 60

46 Treatment and Procedures The procedures of the experiment are shown in Table 3-3 and further explained below. Table 3-3. The procedures of the experiment Notes: C=Chat task. E=input enhancement task. Group C= chat only group. Group E= Group Time Group E Group C Group EC Control Week 1 Pre-treatment Session Fill out background questionnaire. Complete the pretest. Week 2 E+E C+C C+E Week 3 E+E C+C C+E Week 4 E+E C+C C+E Week 5 Treatment Sessions E+E C+C C+E N/A Fill out follow-up questionnaire. N/A Week 6 Post-treatment Session Complete the posttest. enhanced input group. Group EC= input enhancement plus chat Data was collected during the Spring 2004 semester of study at Inha University. In Week 1, all the participants took the pretest. At the same time, background information was elicited through the background questionnaire (Appendix B) asking for information regarding students’ contact information, the length of formal English education, their major field of study, their age, gender, and experience studying other foreign languages, their learning preferences, chat frequency, English typing skills, and so on. The questionnaire was written in Korean to ensure complete comprehension. Between Week 2 and Week 5, the students in the experimental groups were given the directions and materials for their homework activities, which were posted online and updated weekly with new assignments. The students were reminded of these assignments by e-mail and through an announcement on an online space (i.e., Cyberspace), used by Inha University to grant students access to announcement, homework, etc. Each of the eight treatment sessions took approximately one hour, and the participants were required

PAGE 61

47 to complete two one-hour treatments each week, as shown in Table 3-3. The treatment sessions were conducted on their own out of class, and students received extra credit for doing each homework activity. For each task, students in Group E were required to read an enhanced passage (Appendix D) and write a one-paragraph summary of it. The students were asked to write the summary in Korean, so as to ensure that they could not attend to the use of determiners while writing an English summary, which could be seen as a confounding factor in addition to enhanced input. Students in Group C met with their partners in their chat rooms every week, and each pair was assigned a certain task, which was a two-hour session (Appendix E). Students in group EC completed both an input enhancement and a chat session in a week, each of which took one hour (Appendix F). The first chat meeting for both Group EC and Group C included casual conversation to help familiarize themselves with MSN Messenger and to get to know each other. When each treatment was completed, the students submitted their chat transcripts on Cyberclass. All chat conversations were conducted in English. In Week 6, all participants took the posttest, and the students in the experimental groups answered the follow-up questionnaire about whether or not they felt they focused on the correct use of grammar, more specifically English determiners, during online chat; and how much they felt the enhanced parts (i.e., determiners and the following head nouns) drew their attention during the enhanced reading. These questions were made as Yes-No questions. These questions were motivated to investigate how students’ (self-reported) meta-awareness on targeted L2 forms during enhanced reading and chat affected the use of the targeted forms that were reflected in the posttest and chat

PAGE 62

48 transcripts. At the same time, students in the control group also took the posttest, although they did not fill out any additional questionnaires. Data Collection and Analyses Once each chat session was completed, each pair of students was required to save their chat dialogue as a text file and then submit it online to their teacher as homework. The chat transcripts submitted were then downloaded into the author’s data to be further analyzed. Two native speakers of English who are majoring in linguistics at the University of Florida were hired to identify all obligatory occasions for use of core-determiners in the chat transcripts. They were asked to note all errors with core-determiners, including errors of omission as well as commission, in the transcripts and to give their correct forms. Because chat data is in a written form but the register is informal and more open to variation, it was not easy to set the criteria for the two proofreaders. As a result, while their viewpoints on the errors were mostly identical, some differences were found in their proofreading. The judges consulted with the author about these discrepancies to come to an agreement about their differing opinions on the usage of the target items. The primary explanation for these differences relates to the participants’ lack of proficiency in English. The judges reported some difficulty in understanding the contexts when two chatters used English that was comprehensible for linguistic or cultural reasons only between Koreans. Also, the participants’ English created many errors involving structures and forms, mostly with prepositions and nouns, which led to further confusion among the judges. With the proofreading of transcripts completed, the frequency of head nouns was counted and each of the obligatory core-determiners was categorized as a zero article (),

PAGE 63

49 an indefinite article (a(n)), a definite article (the), and one of the other core-determiners (i.e., demonstratives, possessives, and quantifiers). Finally, all types of errors in the pretest and posttest as well as the chat transcripts were classified in obligatory occasions as follows: Table 3-4. Types of errors with English core-determiners Symbol (Correct formError) Example of errors a(n) I am * university student. the He is * one I mentioned. a(n)/the He is * student who attends in Inha Univ. D He is * teacher. (my) a(n) I need *a water. the I like *the English. D I need the *some book you showed me. a(n)the He is *the teacher. a(n)D He is *that student. thea(n) He is *a person I mentioned. theD I hate *his way he speaks. Da(n) This is not *a style considering the way it was done. (his) Dthe I don’t understand *the idea. (your) DD I need *few water. (some) Note: D = the symbol of the core-determiners other than articles (i.e., demonstratives, possessives, and quantifiers) Among deletion errors in the chat transcripts, if the zero article () can be replaced by either “a(n)” or “the,” this case was categorized a(n)/the. Phonologically related errors with indefinite articles were excluded as errors (e.g., a exam), because the main focus of the study is to examine the different usage of the core-determiners at discourse level, not phonological nuances. Each analysis was based on research questions posed in chapter 2, and included descriptive statistics and statistical analyses for the dependent variables. Statistical analyses were performed in order to determine whether any group(s) outperformed the other(s) on the posttest and whether the difference in the dependent variables is

PAGE 64

50 statistically significant. For statistical analyses, nonparametric tests (i.e., Mann-Whitney U tests) were employed, because my study had a relatively small sample size. As compared with parametric tests, which rely on the estimation of parameters, such as the mean or SD, nonparametric tests do not rely on such estimation, describing the distribution of the variable of interest in the population. The statistical significance was set at a p-value of 0.05. SPSS (Statistical Package for the Social Sciences) 11.5 for Windows was employed to analyze the results. For the Research Question 1, the independent variables were treatment type (i.e., Group C, Group E, Group EC, and the control group) and meta-awareness (i.e., attention group and no attention group during enhanced reading; and attention group and no attention group during chat). The dependent variables were the scores of tests, and various types of errors shown in the pretest and posttest, such as the ratio of deletions to total errors (Del/Err); the ratio of retentions to total errors (Ret/Err); the ratio of overuse of “the” to total errors (The/Err); and the ratio of overuse of “a(n)” to total errors (A/Err). Therefore, the scores in the pretest and posttest were presented and compared according to the two independent variables, treatment type and meta-awareness. The ratio of deletions / retentions to total errors (i.e., Del/Err and Ret/Err) and the ratio of overuse of “the” / “a(n)” to total errors (i.e., The/Err and A/Err) were then presented and compared in order to investigate how the difference in the scores is reflected in the misuse of determiners in the pretest and posttest. For the Research Question 2, task type (i.e., a convergent task like information gap / jigsaw and a divergent task like discussion) was added as an independent variable in addition to treatment type and meta-awareness. The dependent variables were the ratio of

PAGE 65

51 errors to total head nouns (Err/N) during chat and the ratio of deletions to total errors (Del/Err); the ratio of retentions to total errors (Ret/Err); the ratio of overuse of “the” (The/Err); and the ratio of overuse of “a(n)” (A/Err). Therefore, these dependent variables are presented and compared according to treatment type and meta-awareness. Each analysis was conducted in the two different types of tasks (i.e., information gap / jigsaw and discussion tasks).

PAGE 66

CHAPTER 4 RESULTS In this chapter, the results of the data analysis were presented in order to answer the two research questions. Each analysis includes the descriptive statistics and the statistical analyses of the dependent variables. The descriptive statistics show the number of participants (N), and the mean, maximum, minimum and standard deviation (SD) of the dependent variables. For the statistical analyses, as mentioned in the previous chapter, nonparametric tests (i.e., Mann-Whitney U tests) were employed for the main comparison, and the statistical significance was set at a p-value of 0.05. The analyses of the quantitative data are followed by a discussion of the qualitative data obtained in the questionnaire and in students’ chat transcripts. Research Question 1 How will treatment type and meta-awareness affect students’ performance, as reflected in the use of determiners in the pretest and posttest? To present the results of the experiment for this research question, the analyses of the scores of tests were offered, followed by the analyses of error types of the pretest and posttest. The error types for the analyses are the ratio of deletions to total errors (Del/Err), the ratio of retentions to total errors (Ret/Err), the ratio of overuse of “the” to total errors (The/Err), and the ratio of overuse of “a(n)” to total errors (A/Err). The analyses of treatment type are first offered, followed by the analyses of meta-awareness during input enhancement and chat, respectively. 52

PAGE 67

53 Therefore, for the analyses of the effects of treatment type on the tests, the following four groups were considered: Group C (the chat group) Group E (the input enhancement group) Group EC (the group which alternated chat and input enhancement activities) Control (No treatment) Then, for the analyses of the effects of meta-awareness during input enhancement and on chat on the tests, the following three distinctions were made: Attention during input enhancement or chat (Attention–E or C) No attention during input enhancement or chat (No attention–E) Control (No treatment) Analyses of Test Scores This section reports on the descriptive statistics and the statistical analyses carried out on the scores of the pretest and posttest. Tables 4-1 to 4-4 and Figure 4-1 present descriptive statistics and graphs for the scores of the pretest and posttest, according to treatment type. Table 4-1. Group C scores for pretest and posttest (N=16) Pretest Posttest Mean 56.14 59.37 Maximum 82.46 88.10 Minimum 28.07 28.57 SD 18.22 19.85 Table 4-2. Group E scores for pretest and posttest (N=16) Pretest Posttest Mean 57.34 60.56 Maximum 82.46 80.95 Minimum 36.84 40.48 SD 13.43 13.95 Table 4-3. Group EC scores for pretest and posttest (N=16) Pretest Posttest Mean 59.86 64.58 Maximum 80.95 80.95

PAGE 68

54 Table 4-3 Continued Pretest Posttest Minimum 35.09 45.24 SD 13.81 8.60 Table 4-4. Control group scores for pretest and posttest (N=16) Pretest Posttest Mean 57.89 58.78 Maximum 75.44 76.19 Minimum 40.35 38.10 SD 10.21 11.65 Note: The maximum scores for each test were originally 57 (pretest) and 42 (posttest), which were reevaluated as percentage for easy comparison. Therefore, the numbers indicate percentage and their maximum value is 100(%). 020406080100PretestPosttestScore Group C Group E Group EC Control Figure 4-1. Scores for pretest and posttest according to treatment type Independent samples tests were performed on the scores of tests, and it was determined that there was no significant difference in the scores of pretest and posttest between groups. Tables 4-5 to 4-7 and Figure 4-2 present descriptive statistics and graphs for the scores of the pretest and posttest, according to meta-awareness during input enhancement. Table 4-5. Attention-E scores for pretest and posttest (N=7) Pretest Posttest Mean 47.62 62.58 Maximum 57.89 73.81 Minimum 35.09 52.38 SD 8.96 8.55

PAGE 69

55 Table 4-6. No attention-E scores for pretest and posttest (N=25) Pretest Posttest Mean 61.68 62.57 Maximum 82.46 80.95 Minimum 36.84 40.48 SD 12.99 12.43 Table 4-7. Control group scores for pretest and posttest (N=16) Pretest Posttest Mean 57.89 58.78 Maximum 75.44 76.19 Minimum 40.35 38.10 SD 10.21 11.65 0102030405060708090100PretestPosttestScore Attention-E No attention-E Control Figure 4-2. Scores for pretest and posttest according to meta-awareness during input enhancement To determine whether or not there is any significant difference, independent samples tests were performed on the scores of the pretest and posttest, and the results are presented in Table 4-8: Table 4-8. Statistical analysis on the test scores Mann-Whitney U Scores GROUP1 GROUP2 p-value 0.000* Attention-E No attention-E 0.000* Attention-E Control 0.000* No attention-E Control 0.885 Note: * indicates statistical significance (p < 0.05).

PAGE 70

56 Mann-Whitney U tests show that there are significant differences in the gains in the posttest between attention-E and no attention-E and between attention-E and Control. According to the results, attention-E gained significantly more in the posttest than no attention-E and Control (p = 0.000). However, when an independent samples test was performed on the pretest scores between the two groups, the pretest scores of the attention-E group are significantly lower than the other groups (p = 0.007). It is possible that the students at a lower proficiency level might have had a better chance to gain more in the posttest. In other words, given the pretest effect, the variable “pretest” might have been involved as an intervening variable. To determine if this was the case, a Regression analysis was run with the variable “pretest” as an independent variable, and with the variable “gains” as a dependent variable. Table 4-9. Regression of “pretest” r 2 Beta T F Sig. Regression 0.086 5.844 0.003* Pretest -0.293 -2.417 0.019* Predictors: Pretest Dependent Variable: Gains in the posttest Table 4-9 indicates that “pretest” is a significant variable for the gains as suggested in the Correlations (p = 0.019). However, given the small r 2 value (0.086), it is hard to say that this variable alone is responsible for much of the difference in the gains. For this reason, a Regression analysis was run again, this time with the variable “treatment type” and “attention-E” as independent variables, “pretest” as a control variable, and with the variable “gains” as a dependent variable.

PAGE 71

57 Table 4-10. Regression of “pretest,” “treatment type,” and “attention-E” r 2 Beta T F Sig. Regression 0.591 13.507 0.000* Pretest -0.232 -1.685 0.103 Treatment type -0.029 -0.228 0.822 Attention-E -0.642 -4.558 0.000* Predictors: Pretest, Attention-E, Treatment type Dependent Variable: Gains in the posttest As shown in Table 4-10, when the three variables were run together, “attention-E” is the only significant factor for the gains (p > 0.01), while neither the “pretest” nor the “treatment type” is. Additionally, the r 2 value increases (0.591) with “attention-E” included, which suggests that although “pretest” is related to the gains, it is the level of attention during enhancement that more adequately explains the difference in the improvement, and not the variable “treatment type.” In other words, among the students in Group E and Group EC, those who attended to the enhanced forms during reading gained more scores in the posttest, regardless of their treatment type. Tables 4-11 to 13 and Figure 4-3 present descriptive statistics and graphs for the scores of the pretest and posttest, according to meta-awareness during chat. Table 4-11. Attention-C scores for pretest and posttest (N=15) Pretest Posttest Mean 58.11 62.79 Maximum 82.46 88.19 Minimum 28.07 28.57 SD 15.36 16.72 Table 4-12. No attention-C scores for pretest and posttest (N=17) Pretest Posttest Mean 57.89 61.16 Maximum 82.46 88.10 Minimum 29.82 30.95 SD 17.15 14.19

PAGE 72

58 Table 4-13. Control group scores for pretest and posttest (N=16) Pretest Posttest Mean 57.89 58.78 Maximum 75.44 76.19 Minimum 40.35 38.10 SD 10.21 11.65 0102030405060708090100PretestPosttestScore Attention-C No attention-C Control Figure 4-3. Scores for pretest and posttest according to meta-awareness during chat Independent samples tests were performed on the scores of tests, and it was determined that there was no significant difference in the scores of pretest and posttest between groups. This result suggests that the act of chat did not help students to learn the targeted forms significantly. Analyses of Error Types in the Test This section reports on the descriptive statistics and the statistical analyses carried out on the error types in the pretest and posttest, such as the ratio of deletions to total errors (Del/Err), the ratio of retentions to total errors (Ret/Err), the ratio of overuse of “the” to total errors (The/Err) and the ratio of overuse of “a(n)” to total errors (A/Err). Because the results of the previous analyses on the scores suggested that meta-awareness during input enhancement was responsible for the difference in the gain between groups, analyses of this variable are offered. On the other hand, the analyses of treatment type

PAGE 73

59 and meta-awareness during chat illustrated that there was no significant difference in the error types in the pretest and posttest, as in the gain of scores. Therefore, analyses of this variable are not offered here. Tables 4-14 to 16 and Figure 4-4 to Figure 4-7 present descriptive statistics and graphs for error types in the pretest and posttest, and Table 4-17 presents their statistical analyses, according to meta-awareness during input enhancement. Table 4-14. Attention-E error types in pretest and posttest (N=7) Del/Err Ret/Err The/Err A/Err Mean 40.27 14.39 18.81 12.33 Pretest Maximum 72.00 21.88 28.13 43.75 Minimum 15.63 0.00 6.06 2.78 SD 19.44 7.58 8.98 14.86 Deletions 84 32 42 21 Total errors 212 212 212 212 Mean 27.69 16.18 26.97 22.95 Posttest Maximum 63.64 25.00 40.00 60.00 Minimum 10.00 9.09 15.00 0.00 SD 22.17 6.18 9.92 19.31 Deletions 31 20 34 31 Total errors 125 125 125 125 Table 4-15. No attention-E error types in pretest and posttest (N=25) Del/Err Ret/Err The/Err A/Err Mean 22.38 19.72 38.38 13.64 Pretest Maximum 50.00 50.00 72.73 33.33 Minimum 0.00 3.70 16.67 0.00 SD 14.89 9.22 16.22 9.63 Deletions 124 105 200 74 Total errors 543 543 543 543 Mean 28.40 19.03 28.90 20.57 Posttest Maximum 65.22 44.44 70.00 42.86 Minimum 0.00 0.00 4.35 0.00 SD 17.03 11.80 17.16 11.41 Deletions 109 65 102 86 Total errors 381 381 381 381

PAGE 74

60 Table 4-16. Control group error types in pretest and posttest (N=16) Del/Err Ret/Err The/Err A/Err Mean 17.23 25.86 43.01 15.77 Pretest Maximum 41.94 43.75 86.96 28.57 Minimum 0.00 11.11 7.41 3.70 SD 14.28 9.73 24.03 7.96 Deletions 71 94 152 60 Total errors 385 385 385 385 Mean 19.04 19.85 38.52 27.78 Posttest Maximum 47.37 33.33 86.96 50.00 Minimum 0.00 3.85 7.69 0.00 SD 14.23 9.63 21.54 13.43 Deletions 57 51 98 78 Total errors 276 276 276 276 Note: The numbers for mean, maximum, minimum and SD indicate percentage (i.e., ratio), and, therefore, the maximum value of each ratio is 100 (%). On the other hand, the numbers of deletions and total errors indicate raw frequency that the students in this group (N=7) made. 020406080100PretestPosttestRatio Attention-E No attention-E Control Figure 4-4. Del/Err in pretest and posttest according to meta-awareness during input enhancement 020406080100PretestPosttestRatio Attention-E No attention-E Control Figure 4-5. Ret/Err in pretest and posttest according to meta-awareness during input enhancement

PAGE 75

61 020406080100PretestPosttestRatio Attention-E No attention-E Control Figure 4-6. The/Err in pretest and posttest according to meta-awareness during input enhancement 020406080100PretestPosttestRatio Attention-E No attention-E Control Figure 4-7. A/Err in pretest and posttest according to meta-awareness during input enhancement Table 4-17. Statistical analyses on error types in the pretest and posttest Mann-Whitney U scores ERROR TYPE Del/Err Ret/Err The/Err A/Err GROUP1 GROUP2 p-value Mann-Whitney U Attention-E No attention-E 0.034* 0.452 0.011* 0.732 41.000 71.000 32.000 80.000 Attention-E Control 0.019* 0.067 0.033* 0.738 21.000 26.000 24.000 51.000 No attention-E Control 0.741 0.179 0.347 0.188 187.500 149.000 164.000 150.000 The results of the independent samples tests determined that meta-awareness during input enhancement was responsible for the significant differences in the ratio of deletions to total errors (Del/Err) and the ratio of overuse of “the” to total errors (The/Err)

PAGE 76

62 of pretest and posttest between attention-E and no attention-E and between attention-E and Control (p < 0.05). In other words, those who had been attending to the enhanced forms during reading decreased their ratio of deletions to total errors (Del/Err) and increased their ratio of overuse of “the” to total errors (The/Err) in the posttest. The differences in the ratio of deletions to total errors (Del/Err) and the ratio of overuse of “the” to total errors (The/Err) of pretest and posttest may be related to the significant difference in the gain of scores. This is because those who gained more scores in the posttest often decreased their ratio of deletions to total errors (Del/Err) and increased their ratio of overuse of “the” to total errors (The/Err) to a greater extent than those who gained less in the posttest. This decrease in the deletion rate might reflect their improvement, since research suggests the deletion of determiners by the lowest proficiency learners as one pattern for the sequence of acquisition (e.g., Simons, 2001). To summarize the findings of the first research question, the results show that input enhancement with meta-awareness was responsible for the significant difference in the gain of scores, the ratio of deletions to total errors (Del/Err), and the ratio of overuse of “the” to total errors (The/Err). In other words, those who reported that their attention had been drawn to the enhanced forms during input enhancement gained more scored in the posttest, while decreasing their ratio of deletions to total errors (Del/Err) and increasing their ratio of overuse of “the” to total errors (The/Err) in the posttest. On the other hand, there were no significant effects of chat on the gains and error types of the tests, regardless of students’ meta-awareness. In answering the first research question, treatment type (i.e., chat, input enhancement, or the alternation of chat and input enhancement) was not a significant factor for students’ performance, as reflected in the

PAGE 77

63 use of determiners in the pretest and posttest. However, meta-awareness did seem to be a necessary condition for noticing, given that the input enhancement groups (i.e., Group E and Group EC) benefited in learning the targeted forms when accompanied by meta-awareness. Research Question 2 How will treatment types, meta-awareness and task type affect students’ performance, as reflected in the use of determiners during chat? Unlike the first research question, this research question included task type as another independent variable. Therefore, to present the results of the experiment for this research question, each analysis was performed separately for the two task types, the convergent tasks (i.e., information gap and jigsaw) and the divergent tasks (i.e., discussion). The analyses of the ratio of errors to total head nouns (Err/N) shown in the students’ chat transcripts were first presented, followed by the analyses of the ratio of deletions to total errors (Del/Err). In this case, to get an overall trend in the direction of these ratios, the first week (Week 1) and the last week (Week 4) are compared for the ratio of these dependent variables. The ratio of retentions to total errors (Ret/Err) and the ratio of overuse of “the” (The/Err) and “a(n)” (A/Err) to total errors were not included for the analyses, because, during chat, these types of errors were found to be minor, and their small number of frequency caused difficulty in the statistical analyses. Therefore, for the analyses of the effects of meta-awareness during input enhancement on the chats, the following two groups are distinguished: Attention during input enhancement (Attention–E) No attention during input enhancement (No attention–E) Then, for the analyses of the effects of meta-awareness during chat on the chats, the following two groups are distinguished:

PAGE 78

64 Attention during chat (Attention–C) No attention during chat (No attention–C) To help understand how task type affected performance, the participants’ use of determiners is compared and presented under the two task types in the following section before presenting the main analyses of this research question. The Use of Determiners in Convergent and Divergent Tasks In order to compare the effect of the two task types on the use of determiners, this section presents the differences in the ratio of total head nouns to total words (N/W), the ratio of errors to total head nouns (Err/N) and the ratio of deletions to total errors (Del/Err) shown in students” chat transcripts. This is then followed by correlations of the total scores of pretest and posttest and the ratio of errors to total head nouns (Err/N) in chat. Tables 4-18 to 4-23 and Figure 4-8 present descriptive statistics and graphs for these differences, and Table 4-24 presents their statistical analyses. Table 4-18. Total head nouns to total words (N/W) in chat convergent and divergent tasks Convergent task Divergent tasks Mean 21.31 12.37 Maximum 27.83 20.58 Minimum 13.86 7.52 SD 3.51 2.80 Total head nouns 2,577 5,051 Total words 12,001 40,841 Table 4-19. Ratio of errors to total head nouns (Err/N) in chat convergent and divergent tasks Convergent task Divergent tasks Mean 27.16 22.29 Maximum 51.81 44.13 Minimum 5.56 4.69 SD 12.33 7.12 Total errors 700 1,126 Total head nouns 2,577 5,051

PAGE 79

65 Table 4-20. Ratio of deletions to total errors (Del/Err) in chat convergent and divergent tasks Convergent task Divergent tasks Mean 87.57 87.21 Maximum 100.00 100.00 Minimum 42.86 56.00 SD 16.64 9.27 Total deletions 613 982 Total errors 700 1,126 Table 4-21. Ratio of retentions to total errors (Ret/Err) in chat convergent and divergent tasks Convergent task Divergent tasks Mean 3.43 9.23 Maximum 33.00 36.00 Minimum 0.00 0.00 SD 7.32 8.16 Total retentions 24 103 Total errors 700 1,116 Table 4-22. Ratio of overuse of “the” to total errors (The/Err) in chat convergent and divergent tasks Convergent task Divergent tasks Mean 2.29 6.36 Maximum 9.09 11.11 Minimum 0.00 0.00 SD 2.42 2.53 Total overuse of “the” 16 71 Total errors 700 1,116 Table 4-23. Ratio of overuse of “a(n)” to total errors (A/Err) in chat convergent and divergent tasks Convergent task Divergent tasks Mean 10.14 6.27 Maximum 57.14 20.00 Minimum 0.00 0.00 SD 15.48 5.89 Total overuse of “a(n)” 71 70 Total errors 700 1,116

PAGE 80

66 020406080100N/WErr/NDel/ErrRet/Err'the''a(n)'Ratio Information gap & Jigsaw Discussion Figure 4-8. N/W, Err/N, Del/Err, Ret/Err, The/Err and A/Err shown in chat convergent and divergent tasks Table 4-24. Statistical analyses on N/W, Err/N, Del/Err, Ret/Err, The/Err and A/Err in chat convergent and divergent tasks ITEM p-value Mann-Whitney U N/W 0.000* 38.000 Err/N 0.054 343.500 Del/Err 0.637 446.500 Ret/Err 0.001* 242.500 The/Err 0.146 404.500 A/Err 0.318 409.500 The results of the independent samples tests show that there are significant differences in the ratio of head nouns to total words (N/W) (p = 0.000) and the ratio of retentions to total errors (Ret/Err) (p = 0.001) between the convergent and divergent tasks. Though not statistically significant, there is a noticeable difference in the ratio of errors to total head nouns (Err/N) between tasks (p = 0.054). Based on these results, an convergent task has higher ratio of head nouns to total words than does a divergent task. In the divergent tasks, students incorrectly retain more determiners than in the convergent tasks. There is no significant difference in the ratio of deletions to total errors (Del/Err) between the two task types, showing that much higher ratios of deletions are shown in both tasks (87.57% and 87.21%) than in the pretest and posttest (21.84%). This is

PAGE 81

67 probably because a multiple-choice cloze test is a form-focused activity, whereas chat is a meaning-focused one. Tables 4-25 and 4-26 present correlations of the test scores and the ratio of errors to total head nouns (Err/N). Table 4-25. Correlations of test scores and Err/N for convergent tasks Total scores of tests Err/N Total scores of tests Pearson Correlation Sig. (2-tailed) N 1 . 64 -0.423 0.020* 30 Err/N Pearson Correlation Sig. (2-tailed) N -0.423 0.020* 30 1 . 30 Table 4-26. Correlations of test scores and Err/N for divergent tasks Total scores of tests Err/N Total scores of tests Pearson Correlation Sig. (2-tailed) N 1 . 64 -0.241 0.184 32 Err/N Pearson Correlation Sig. (2-tailed) N -0.241 0.184 32 1 . 32 The analyses of correlations show that there is a significant correlation between the sum of preand posttest scores and the ratio of errors to total head nouns (Err/N) in the convergent tasks, but no significant correlation in the divergent task. We can hypothesize that the convergent tasks may reflect students’ grammatical competence as shown in the multiple-choice cloze test more precisely than does the divergent task. This may be because in a divergent task like a discussion students can avoid the expressions they have little confidence in, whereas it is difficult to do so in a convergent task designed to elicit particular structures.

PAGE 82

68 Analyses of Errors to Total Head Nouns This section reports on the descriptive statistics and the statistical analyses carried out on the ratio of errors to total head nouns (Err/N) during chat. Tables 4-27 and 4-28 and Figure 4-9 present descriptive statistics and graphs for the ratio of errors to total head nouns (Err/N) in chat convergent tasks, according to meta-awareness during input enhancement. Table 4-27. Attention-E Err/N in chat convergent tasks (N=5) Week1 WEEK2 WEEK3 WEEK4 Mean 44.81 30.04 38.62 30.52 Maximum 66.67 37.93 66.67 58.82 Minimum 22.73 25.00 16.67 21.35 SD 18.71 5.13 21.01 15.93 Errors 54 37 35 22 Total head nouns 119 120 82 57 Table 4-28. No attention-E in chat convergent tasks (N=27) Week1 WEEK2 WEEK3 WEEK4 Mean 24.61 18.57 28.74 28.57 Maximum 53.33 55.56 75.00 65.52 Minimum 4.35 0.00 0.00 0.00 SD 13.23 13.77 18.74 16.56 Errors 127 99 132 194 Total head nouns 542 547 487 623 Independent samples tests were performed on the ratio of errors to total head nouns (Err/N) from week 1 to week 4 between the two groups, and there was no significant difference between groups in the convergent tasks. However, Tables 4-27 and 4-28 and Figure 4-9 show that the ratio of errors to total head nouns (Err/N) appears to have decreased more noticeably in attention-E than no attention-E. It is important that my study employed nonparametric tests for the statistical analyses, while the figures in my study were based on parameters (i.e., mean value of the dependent variables), so these figures may not necessarily reflect the results of statistical analysis. On the other hand,

PAGE 83

69 parametric t-tests show that while there was no significant difference between attention-E and no attention-E, the difference did suggest a trend in the direction mentioned (i.e., more improvement for attention-E) (p < 0.07). 0102030405060708090100W1W2W3W4Ratio Attention-E No attention-E Figure 4-9. Err/N in chat convergent tasks according to meta-awareness during input enhancement Tables 4-29 and 4-30 and Figure 4-10 present descriptive statistics and graphs for the ratio of errors to total head nouns (Err/N) in chat divergent tasks, according to meta-awareness during input enhancement. Independent samples tests determined that there was no significant difference in the ratio of errors to total head nouns (Err/N) from week 1 to week 4 between the two groups in the divergent tasks. These results suggest that the act of paying attention during the input enhancement did not significantly affect the students’ accuracy. Table 4-29. Attention-E Err/N in chat divergent tasks (N=5) Week1 WEEK2 WEEK3 WEEK4 Mean 28.23 22.84 15.16 20.29 Maximum 38.24 34.62 23.08 28.80 Minimum 20.00 11.43 0.00 12.50 SD 6.99 9.04 9.19 7.49 Errors 57 49 26 52 Total head nouns 210 207 157 244

PAGE 84

70 Table 4-30. No attention-E Err/N in chat divergent tasks (N=27) Week1 WEEK2 WEEK3 WEEK4 Mean 23.69 25.20 22.55 22.98 Maximum 85.71 47.62 47.37 48.00 Minimum 2.86 7.50 2.15 1.67 SD 17.44 11.32 10.50 10.05 Errors 187 278 233 244 Total head nouns 906 1,175 1,079 1,073 0102030405060708090100W1W2W3W4Ratio Attention-E No attention-E Figure 4-10. Err/N in chat divergent tasks according to meta-awareness during input enhancement Tables 4-31 and 4-32 and Figure 4-11 present descriptive statistics and graphs for the ratio of errors to total head nouns (Err/N) in chat convergent tasks, according to meta-awareness during chat. Table 4-31. Attention-C Err/N in chat convergent tasks (N=15) Week1 WEEK2 WEEK3 WEEK4 Mean 28.82 17.98 31.67 26.00 Maximum 66.67 38.89 75.00 65.52 Minimum 4.35 0.00 0.00 0.00 SD 18.52 12.92 22.71 16.44 Errors 90 59 79 80 Total head nouns 324 284 265 290

PAGE 85

71 Table 4-32. No attention-C Err/N in chat convergent tasks (N=17) Week1 WEEK2 WEEK3 WEEK4 Mean 27.70 22.99 29.11 31.79 Maximum 61.90 55.56 54.55 58.93 Minimum 10.00 0.00 0.00 4.26 SD 14.15 13.83 15.42 15.98 Errors 91 77 88 136 Total head nouns 337 383 304 390 0102030405060708090100W1W2W3W4Ratio Attention-C No attention-C Figure 4-11. Err/N in chat convergent tasks according to meta-awareness during chat Independent samples tests were performed on the ratio of errors to total head nouns (Err/N) from week 1 to week 4 between the two groups in the chat transcripts, and they determined that there was no significant difference between the groups. Tables 4-33 and 4-34 and Figure 4-12 present descriptive statistics and graphs for the ratio of errors to total head nouns (Err/N) in chat divergent tasks, according to meta-awareness during chat. Independent samples tests were performed on the ratio of errors to total head nouns (Err/N) from week 1 to week 4 between the two groups, and they confirm that there was no significant difference between groups.

PAGE 86

72 Table 4-33. Attention-C Err/N in chat divergent tasks (N=15) Week1 WEEK2 WEEK3 WEEK4 Mean 26.31 26.70 19.01 22.14 Maximum 85.71 47.62 35.29 48.00 Minimum 6.48 7.50 2.15 1.67 SD 20.59 12.39 8.87 11.23 Errors 103 176 103 138 Total head nouns 492 652 579 630 Table 4-34. No attention-C Err/N in chat divergent tasks (N=17) Week1 WEEK2 WEEK3 WEEK4 Mean 22.72 23.19 23.50 22.93 Maximum 50.00 40.00 47.37 39.02 Minimum 2.86 8.51 0.00 12.50 SD 11.59 9.48 11.65 8.34 Errors 141 151 156 158 Total head nouns 624 730 657 687 0102030405060708090100W1W2W3W4Ratio Attention-C No attention-C Figure 4-12. Err/N in chat divergent tasks according to meta-awareness during chat Analyses of Deletions to Total Errors This section reports on the descriptive statistics and the statistical analyses carried out on the ratio of deletions to total errors (Del/Err) during chat.

PAGE 87

73 Tables 4-35 and 4-36 and Figure 4-13 present descriptive statistics and graphs for the ratio of deletions to total errors (Del/Err) during chat convergent tasks, according to meta-awareness during input enhancement. Table 4-35. Attention-E Del/Err in chat convergent tasks (N=5) Week1 WEEK2 WEEK3 WEEK4 Mean 85.87 72.00 73.25 70.45 Maximum 100.00 100.00 100.00 90.00 Minimum 71.43 0.00 25.00 0.00 SD 14.09 41.47 32.50 45.09 Deletions 49 19 17 10 Total errors 57 26 24 16 Table 4-36. No attention-E Del/Err in chat divergent tasks (N=27) Week1 WEEK2 WEEK3 WEEK4 Mean 77.96 78.73 88.91 79.78 Maximum 100.00 100.00 100.00 100.00 Minimum 0.00 0.00 0.00 25.00 SD 29.92 35.61 25.94 19.01 Deletions 98 89 125 165 Total errors 124 110 143 200 0102030405060708090100W1W2W3W4Ratio Attention-E No attention-E Figure 4-13. Del/Err in chat convergent tasks according to meta-awareness during input enhancement Independent samples tests determined that there was no significant difference between groups. This result indicates that meta-awareness during input enhancement

PAGE 88

74 does not necessarily result in a significant difference in the ratio of deletions to total errors (Del/Err) during convergent chat tasks. Tables 4-37 and 4-38 and Figure 4-14 present descriptive statistics and graphs for the ratio of deletions to total errors (Del/Err) during chat divergent tasks, according to meta-awareness during input enhancement. Table 4-37. Attention-E Del/Err in chat divergent tasks (N=5) Week1 WEEK2 WEEK3 WEEK4 Mean 78.59 88.16 77.67 84.76 Maximum 100.00 94.44 91.67 100.00 Minimum 33.33 80.00 50.00 71.43 SD 27.45 5.25 21.11 8.34 Deletions 30 45 15 42 Total errors 39 50 19 49 Table 4-38. No attention-E in chat divergent tasks (N=27) Week1 WEEK2 WEEK3 WEEK4 Mean 82.20 85.06 82.42 81.85 Maximum 100.00 100.00 100.00 100.00 Minimum 0.00 50.00 16.67 0.00 SD 26.14 18.45 21.35 19.99 Deletions 167 150 199 206 Total errors 205 277 240 247 0102030405060708090100W1W2W3W4Ratio Attention-E No attention-E Figure 4-14. Del/Err in chat divergent tasks according to meta-awareness during input enhancement

PAGE 89

75 Independent samples tests were performed on the ratio of deletions to total errors (Del/Err) from week 1 to week 4 between the three groups and they confirm that there was no significant difference between groups. This result indicates that meta-awareness during input enhancement does not necessarily result in a significant difference in the ratio of deletions to total errors (Del/Err) during divergent chat tasks. Tables 4-39 and 4-40 and Figure 4-15 present descriptive statistics and graphs for the ratio of deletions to total errors (Del/Err) during chat convergent tasks, according to meta-awareness during chat, and Table 4-41 presents their statistical analyses. Table 4-39. Attention-C Del/Err in chat convergent tasks (N=15) Week1 WEEK2 WEEK3 WEEK4 Mean 84.49 69.16 72.90 72.22 Maximum 100.00 100.00 100.00 100.00 Minimum 33.33 0.00 0.00 0.00 SD 19.79 43.04 36.30 33.03 Deletions 74 50 64 65 Total errors 90 59 79 80 Table 4-40. No attention-C Del/Err in chat divergent tasks (N=17) Week1 WEEK2 WEEK3 WEEK4 Mean 74.93 84.52 97.04 89.45 Maximum 100.00 100.00 100.00 100.00 Minimum 0.00 0.00 83.33 70.00 SD 33.18 28.45 5.65 11.73 Deletions 79 72 84 125 Total errors 91 77 88 136 The result of statistical analysis shows that there is a significant difference in the ratio of deletions to total errors (Del/Err) from week 1 to week 4 between attention-C and no attention-C (p = 0.013). This result indicates that meta-awareness during chat may be responsible for the difference in the ratio of deletions to total errors (Del/Err) from week 1 to week 4 during the convergent chat tasks.

PAGE 90

76 0102030405060708090100W1W2W3W4Ratio Attention-C No attention-C Figure 4-15. Del/Err in chat convergent tasks according to meta-awareness during chat Table 4-41. Statistical analyses on Del/Err during chat convergent tasks Mann-Whitney U scores GROUP1 GROUP2 p-value Mann-Whitney U Attention-C No attention-C 0.013* 26.000 Tables 4-42 and 4-43 and Figure 4-16 present descriptive statistics and graphs for the ratio of deletions to total errors (Del/Err) during chat divergent tasks, according to meta-awareness during chat, and Table 4-44 presents their statistical analyses. Table 4-42. Attention-C Del/Err in chat divergent tasks (N=15) Week1 WEEK2 WEEK3 WEEK4 Mean 84.84 84.77 77.09 74.21 Maximum 100.00 100.00 100.00 100.00 Minimum 33.33 50.00 16.67 0.00 SD 20.29 17.03 24.09 22.85 Deletions 90 154 86 112 Total errors 103 176 103 138 Table 4-43. No attention-C Del/Err in chat divergent tasks (N=17) Week1 WEEK2 WEEK3 WEEK4 Mean 78.81 86.23 86.29 89.45 Maximum 100.00 100.00 100.00 100.00 Minimum 0.00 54.55 33.33 75.00 SD 30.39 17.55 17.31 9.81 Deletions 118 132 138 143 Total errors 141 151 156 158

PAGE 91

77 0102030405060708090100W1W2W3W4Ratio Attention No attention Figure 4-16. Del/Err in chat divergent tasks according to meta-awareness during chat Table 4-44 . Statistical analyses on Del/Err during chat divergent tasks Mann-Whitney U scores GROUP1 GROUP2 p-value Mann-Whitney U Attention-C No attention-C 0.033* 71.000 The result of statistical analysis shows that there is a significant difference in the ratio of deletions to total errors (Del/Err) from week 1 to week 4 between attention-C and no attention-C (p = 0.033). This result indicates that meta-awareness during chat may be responsible for the difference in the ratio of deletions to total errors (Del/Err) during the divergent chat tasks. To summarize our findings relating to the second research question, the results of the analyses of the ratio of errors to total nouns (Err/N) and the ratio of deletions to total errors (Del/Err) in chat showed that there was no significant difference in the ratio of errors to total nouns (Err/N) from week 1 to week 4 in the convergent tasks, but a noticeable difference between attention-E and no attention-E. Additionally, the results showed that meta-awareness during chat resulted in a significant decrease in the ratio of deletions to total errors (Del/Err) in week 4 in both task types. In answering the second research question, as in the first research question, treatment type (i.e., chat versus chat

PAGE 92

78 along with input enhancement) was not a significant factor for students’ performance, as reflected in the use of determiners in the chat transcripts. Instead, meta-awareness should be considered as a variable, given that the groups of chat (i.e., Group C and Group EC) showed decreased ratios of deletions to total errors (Del/Err) in week 4, as compared with week 1, when accompanied by meta-awareness. Task type also needs to be considered as a variable, since the convergent tasks reflect students’ grammatical competence better than the divergent tasks, as shown in the pretest and posttest in my study. However, there was no significant difference in the ratio of deletions to total errors (Del/Err) between the two task types. In conclusion, the results of the quantitative analyses on the first research question determined that input enhancement with meta-awareness was responsible for the significant difference in the gain of scores, the ratio of deletions to total errors (Del/Err), and the ratio of overuse of “the” to total errors (The/Err). On the other hand, there were no significant effects of chat on the gains and error types of the tests, regardless of students’ meta-awareness. In addition, the results of the quantitative analyses on the second research question determined that deletion of determiners is one major type of error in Korean students’ chat conversation, and students’ meta-awareness during chat is responsible for the difference in the ratio of deletions to total errors (Del/Err). Nevertheless, both attention-C and no attention-C show a much higher ratio of deletions to total errors (Del/Err) in the chats than in the pretest and posttest, showing that chat may not be a good place for students to practice and notice articles. Qualitative Data This section presents qualitative data regarding students’ (self-rated) meta-awareness during chat activities, based on students’ answers to the questionnaire

PAGE 93

79 and on their chat transcripts. The results of the quantitative analyses on the second research question determined that deletion of determiners is one major phenomenon in Korean students’ chat conversation (if it is not an error), and students’ meta-awareness during chat is responsible for the difference in the ratio of deletions to total errors (Del/Err), although both attention-C and no attention-C show a much higher ratio of deletions to total errors (Del/Err) in the chats than in the pretest and posttest. In this section, we examine students’ responses when asked if they attended to grammar during their chat conversation, and why not, if they answered no, etc. Then, the comparison of a pair of students’ use of determiners during chat is presented to show how meta-awareness affected their performance during chat. These students got similar scores on the tests but showed extremely different patterns in terms of the ratio of deletions to total errors (Del/Err) during chat. First, we will look at students’ answers to the questionnaire regarding meta-awareness and their analyses. To the question if they consciously focused on the correct use of grammar during the chat activities, 15 out of 32 students reported that they had attended to grammar as much as they could, while 17 students reported that they had mostly ignored it. To the question if they chose “Yes,” how much the act of consciously focusing on grammar affected their comprehension and production during the chat activities, 11 out of the 15 students chose “a little,” and the other four students chose “a lot.” No one reported that it did not affect them in their chat conversation at all. To the question if they chose “No,” why they did not focus on it, the following are general answers from them: “As conversation got faster, it was hard for me to focus on grammar. The nature of chat makes us prefer brevity, I think.”

PAGE 94

80 “I focused on communication for the sake of convenience.” “It took too much time to write grammatically correct expressions. At first, I didn’t submit what I typed without re-checking my English, but eventually I realized grammar was not very important for communication.” “I tried to attend to grammar at first, but eventually neglected grammar because it was too difficult to keep doing.” “I preferred being comfortable during chat (by not attending to grammar).” “Because I felt I was talking (instead of writing) during chat, I was just rushed to communicate, and this made me neglect grammar (e.g., tense, spelling, etc).” “(I ignored grammar,) because I did not know grammar very well.” “Communication was possible without attending to grammar.” As indicated by their responses, a major reason students did not attend to grammar is that they thought that the main purpose of chat conversation was communication itself. Based on this finding, students seem to think that grammar and communication are mutually exclusive. As in reading, where comprehension of content and attention to form are difficult to occur simultaneously, during the chat conversations, it seemed to be difficult for students to attend to grammar while attending to communication at the same time. As discussed in Chapter 2, the limited attention capacity of human being can explain this difficulty, and the written nature of chat did not seem to be an advantage in performing mental work. Grammatical competence also seems to be a reason for which students omit determiners, given that some students reported that grammar was difficult and there were positive correlations between the ratio of errors to total head nouns (Err/N) and the ratio of deletions to total errors (Del/Err) in both chat task types (p = 0.045 for the divergent task and p = 0.051 for the convergent tasks). The students in my study, who were at the intermediate (or lower) level of English, might have difficulty in focusing on the less

PAGE 95

81 urgent grammatical features (i.e., articles), while there were more urgent grammatical features to be attended to. Maybe for this reason, deletion of articles was unavoidable throughout chat conversations. Nevertheless, this tendency may also have to do with individual factors, such as motivation and attitude. In order to get a more detailed picture of these individual factors as well as particular performance on the chats, we examine two particular students in detail. Table 4-45 presents descriptive statistics for their error types in the pretest and posttest and during chat task. Table 4-45. Comparison of the two students’ determiner use in the test and chat Student 1 Student 2 Group Group C Group C Meta-awareness during chat Attention No attention Pretest scores 82.46 82.46 Posttest scores 85.71 88.10 Test Del/Err 12.50 (2/16) 20.00 (3/15) Ret/Err 25.00 (4/16) 26.67 (4/15) The/Err 68.75 (11/16) 53.33 (8/15) A/Err 6.25 (1/16) 20.00 (3/15) Convergent tasks Err/N 7.50 (6/80) 15.97 (19/119) Del/Err 16.67 (1/6) 100.00 (19/19) Chat Divergent tasks Err/N 4.69 (16/341) 23.75 (138/581) Del/Err 81.25 (13/16) 95.65 (132/138) Table 4-45 presents the use of determiners by a pair of students who showed two extreme cases in terms of the ratio of deletions to total errors (Del/Err) during chat. Although their accuracy and their choice of determiners are not very different in the pretest and posttest, the ways they used determiners during chat conversation are quite different from each other. Particularly, their ratio of deletions to total errors (Del/Err) is remarkably different during chat. Student 2 has 100% and 95.65% in the ratio of

PAGE 96

82 deletions to total errors (Del/Err) in convergent tasks and divergent tasks respectively, whereas Student 1 has 16.67% and 81.25%. Likewise, the ratio of errors to total head nouns (Err/N) is quite different between the two students; this was possibly caused by their difference in the ratio of deletions to total errors (Del/Err). Although they were in the same group (Group C), their self-rated meta-awareness was different, and this appears to have caused their difference in the ratio of deletions to total errors (Del/Err) during chat. The conversations between these two students show that Student 2 was aware of his omission of determiners during chat. The following is part of a conversation between the two students: 1) Conversations from the chat transcripts between Student 1 and Student 2 Student 1: cuz, as you and I are doing, we are omitting almost "the, a, an, some" things. Student 2: huh?? Student 1: His objective is Student 1: to observe the students' using the "determiner", Student 2: . Student 1: but we are not trying hard to use it. In fact we are all ignoring it, right? Student 2: poh In the chat conversations, Student 2 explicitly mentioned that he intentionally omitted articles in the chat mode. This indicates that function words like articles could be intentionally omitted at any time during chat conversations, because they are not crucial for communication. In this sense, it may be hard to consider deletions real errors. For more sample chat transcripts between these two students, see Appendix G.

PAGE 97

CHAPTER 5 DISCUSSION AND CONCLUSION This chapter further discusses the findings as a whole and provides conclusions. Then, it addresses pedagogical implications regarding the use of chat and textual enhancement for developing L2 function words. Finally, the limitations of my study and suggestions for future research are presented. Discussion We return to the research questions that motivated the study in order to now provide answers. Then, discussion of the general research question follows. Research Question 1 Research Question 1 asked how treatment type and meta-awareness would affect students’ performance, as reflected in the use of English determiners in the pretest and posttest. The answer to this question showed that treatment type (i.e., chat, input enhancement, or the combination of chat and input enhancement) was not a significant factor for students’ performance, as reflected in their use of determiners in the pretest and posttest. However, meta-awareness was a necessary variable to be considered, given that the groups with input enhancement (i.e., Group E and Group EC) benefited in learning the targeted forms only when they attended to them, although the chat group did not benefit from attention. How, then, does meta-awareness during input enhancement explain the difference in the gains and the ratio of deletions to total errors (Del/Err) and the ratio of overuse of “the” to total errors (The/Err)? For this, it is necessary to consider the nature of reading. 83

PAGE 98

84 Considering that the primary purpose of reading is comprehension of content, and the limited capacity in working memory makes comprehension effortful in execute all of the process for reading (Just & Carpenter, 1992), it may be difficult for readers to attend to form and content at the same time. According to VanPatten (1996), “although meaning and form are not necessarily mutually exclusive, they may compete for processing resources, and meaning generally wins out” (p. 18). Textual enhancement was employed here for this reason, to help students attend to the targeted forms while comprehending the content, which otherwise would be difficult. Nevertheless, we must note that although 7 out of 32 students reported that their attention had been drawn to the enhanced parts during reading, more students (25 out of 32) reported that their attention had not been drawn to the enhanced parts while comprehending the material. Therefore, there is no guarantee that students will attend to or notice targeted forms while reading, even with textual enhancement, not to mention during casual reading for comprehension. On the other hand, those who reported attending to the enhanced parts gained more scores in the posttest, while decreasing their ratio of deletions to total errors (Del/Err) and increasing their ratio of overuse of “the” to total errors (The/Err) in the posttest. Given these results, input enhancement seems to be beneficial only when accompanied by meta-awareness. Specifically, it seems to be that those who attended to the enhanced forms during input enhancement noticed the use of definite articles, and started to overuse “the” while decreasing their ratio of deletions (Del/Err). It is not clear from these results whether the improvement is a result of attending to the enhanced parts, because my study did not consider the semantic matrix of the definite article “the.”

PAGE 99

85 Previous studies often employed the semantic matrix for the analyses of the sequence of acquisition (e.g., Huebner, 1983; Master, 1988; Thomas, 1989). For example, Thomas (1989) suggests that both L1 and L2 learners overgeneralize “the” in first-mention contexts ([+SR, -HK]) but not in [-SR, -HK] contexts. Nevertheless, my study assumes that the decrease in the deletion rate reflects improvement, since research suggests that the deletion of determiners among the lowest proficiency learners is one pattern for the sequence of acquisition (e.g., Simons, 2001). Students’ comprehension of the reading passages was not taken into account in my study, since it was assumed that the material was at an appropriate level. The questionnaire did ask students whether or not the material was difficult for them (See Appendix B). Seven out of 32 students in Group E and Group EC reported that the reading material was difficult to comprehend, and three out of these seven reported that their attention had been drawn to the enhanced parts. It is not clear whether or not meta-awareness was related to a student’s self-rated reading proficiency, although having students’ attention drawn to form could have affected their comprehension. Unlike the act of attention to grammar during chat, which students could control consciously, it is not clear whether or not attention was consciously paid to the enhanced parts during textual enhancement. The questionnaire asked how much the bold and highlighted words drew students’ attention rather than asking whether they paid attention to the enhanced parts consciously. Students’ attention to the targeted forms during input enhancement might be related to individual factors (e.g., reading style) rather than to students’ effort to attend to them. Therefore, it is not clear why some of the students

PAGE 100

86 attended to the enhanced forms while others did not, and, more importantly, it is not known what helped the students notice the targeted forms. Further, as was seen above, there are some conflicting positions on consciousness and noticing in SLA. Recall that Schmidt (1993) claims that learners must consciously pay attention to or notice input in order for L2 data to become intake, because subliminal language learning is impossible. On the other hand, Tomlin & Villa (1994) claim that conscious awareness may not be necessary in the attention process, postulating that awareness can be separated from attention. Because the questionnaire does not clarify this, the results of my study do not clearly support either position. We can, at least, suggest that merely reading for comprehension without meta-awareness is unlikely to help students notice the targeted forms, even in textually manipulated material. Although input enhancement was found to be effective for noticing the targeted forms, when accompanied by meta-awareness, there was no significant effect of chat, regardless of students’ meta-awareness. Seven out of 16 students overall in Group C reported that they had not been very concerned with attending to grammar, because communication was the main goal of chat, and it was too demanding to attend to grammar while communicating, given their current grammatical knowledge. They reported that, in many cases, they had simply sent their message despite finding errors. On the other hand, the other nine students in Group C reported that they had attended to grammar as much as they could throughout the activities. Nevertheless, these efforts did not result in a significant difference in performance or in error types of pretest and posttest.

PAGE 101

87 Although some believe that the act of production can promote L2 learners’ meta-awareness (or noticing function) for targeted forms (e.g., Leow, 2003; Swain, 1993, 1995, 1998), the production of English core-determiners, particularly articles, seems difficult to induce, most likely because the misuse or deletion of articles does not usually cause trouble for comprehension between interlocutors. As addressed in Chapter 2, because the main purpose of chat is often no more than communication itself, chatters have a tendency to omit less crucial words for communication in order to save time and effort. In my study, there was not much need for students to use articles during the chat sessions, and this was not much different when students tried to consciously attend to grammar. Consequently, students did not have many chances to notice the targeted forms through chat. Originally, the input enhancement session was employed in my study for the students of Group EC, rather than Group E, in the hope that this additional session could promote their meta-awareness during chat and provide the students of Group EC with models of the targeted forms. However, among the students in Group EC, 9 out of the 16 students overall reported that they had not been very concerned with attending to grammar during chat, for the same reasons as students in Group C. Given that there were no correlations between treatment type (i.e., chat only versus chat along with input enhancement) and meta-awareness during chat, the input enhancement session did not appear to promote meta-awareness during chat for the students in Group EC. On the other hand, the results showed that the input enhancement session was beneficial only when accompanied by meta-awareness, and the chat session was not in any case. Therefore, it is hard to say that the input enhancement session provided the

PAGE 102

88 students with models of the targeted forms, unless meta-awareness during input enhancement is considered as a variable. This is why there was no significant improvement for Group EC over Group C and Group E. Research Question 2 Research Question 2 asked how treatment type, students’ and task type would affect meta-awareness performance, as reflected in their use of English determiners during chat. The answer to this question showed again that, as in Research Question 1, treatment type (i.e., chat versus chat along with input enhancement) was not a significant factor for students’ performance, as reflected in the use of determiners in the chat transcripts. Instead, meta-awareness should be considered as a variable, given that the groups that participated in chat (i.e., Group C and Group EC) showed a more decreased ratio of deletions to total errors (Del/Err) in week 4 when accompanied by self-rated meta-awareness. Task type also needs to be considered as a variable, since the convergent tasks seem to represent students’ grammatical competence best, as was seen in the pretest and posttest in my study. However, the results showed that there was no significant difference in the ratio of deletions to total errors (Del/Err) between the two task types. The answer to the second research question showed that the input enhancement session, even when accompanied by meta-awareness, had no significant effect on the use of English determiners in either task type, as reflected in the ratio of errors to total head nouns (Err/N) and the ratio of deletions to total errors (Del/Err). However, recall that Figure 4-8 show that, in the information gap and jigsaw tasks, the difference of a trend in the direction of the ratios suggests more improvement for those who attended to enhanced parts during input enhancement.

PAGE 103

89 We need to remember that the analyses of Correlations between the test scores and the ratio of errors to total head nouns (Err/N) in chat showed that there was a significant correlation between the two variables in the information gap and jigsaw tasks but no significant correlation in the discussion task. As mentioned earlier, the information gap and jigsaw tasks may reflect students’ grammatical competence shown in the multiple-choice cloze test more precisely than does the discussion task. For this reason, students’ improved performance on the posttest might have been reflected as the lowered ratio of errors to total head nouns (Err/N) in the information gap and jigsaw chat tasks, while there is no such noticeable reflection in the discussion chat tasks. The results showed that meta-awareness during chat resulted in a significant decrease in the ratio of deletions to total errors (Del/Err) in both task types. From the qualitative data based on the questionnaire, we learned that many of the no attention-C group reported that they had tried to attend to grammar until they eventually gave up, which could possibly be the factor for the difference. Specifically, these students seemed to attend to and use determiners at first, but their efforts were decreased over the course of the chat. As discussed in the qualitative data, students thought that the main purpose of chat conversation was communication itself, and they might have thought that grammar and communication were mutually exclusive. Retrospective interview sessions would have helped us to understand the difference in the ratio of deletions to total errors (Del/Err) between the attention group and the no attention group during chat. Even though students’ meta-awareness resulted in a significant difference in the ratio of deletions to total errors (Del/Err), the overall ratio of deletions to total errors (Del/Err) remained very high throughout the chat conversation in both task types. This

PAGE 104

90 may be partly because the telegraphic nature of chat seems to be difficult to overcome even for educational purposes. In addition, it was too difficult for students to focus on the less-serious grammatical features (i.e., articles), while there were more-urgent grammatical features the participants needed to pay attention to, such as prepositions, pronouns, and verb tense. On the other hand, although there was a significant difference in the ratio of deletions to total errors (Del/Err) of week 1 and week 4 between attention-C and no attention-C, there was no significant difference in the ratio of errors to total head nouns (Err/N) of week 1 and week 4. Combining these results with our answers to the first research question, we can conclude that the act of chat itself is not likely to be effective for learning English articles. This is apparently because the need to focus on articles was not there and, accordingly, students did not have many chances to notice the targeted forms. Previous studies on the acquisition of articles often used L2 learners’ production data (either written or spoken language) for analysis, but, in production data, learners often avoided using articles when they faced occasions they were not sure of. For example, Simons (2001) reported that bare nouns were prevalent among students at all levels. Likewise, the chat data used in my study show a high ratio of deletions to total errors, and, consequently, a comparatively low ratio of other types of errors. In contrast, the students were forced to choose at least one core-determiner as an answer in the multiple-choice tests, in which each cloze passage included four options, such as zero article, “a(n),” “the,” or another core-determiner (The “zero article” here is not basically different from deletions of articles, but, it implies that, in the multiple-choice tests,

PAGE 105

91 students did not omit articles by/through negligence as they did in chat; instead, they actively chose zero articles). In this case, a zero article was always an option in the multiple-choice cloze tests. The comparison of the ratio of deletions to total errors (Del/Err) in the pretest and posttest and chat shows that deletions were the prominent type of error in chat (87.39%), whereas there was no such one prominent error in the multiple-choice cloze tests. Additionally, the ratio of errors to total head nouns (Err/N) on the tests was 40.68%, whereas the ratio in chats was 24.73% (p = 0.000). There may be at least two explanations for the large differences in this ratio. First, chat conversations (as well as oral production) may consist of comparatively less difficult lexical items, because students can avoid any expressions that they are unsure of. Therefore, it might be less difficult for students to use target determiners correctly in chat conversation than in the multiple-choice cloze tests, because they are in control of the context and their output. Second, during chat, the students can use alternative determiners, such as demonstratives, possessives and quantifiers, when they are not sure of the correct usage of articles. To illustrate, in the first sentence in the following example, if they do not know whether “rice cake” is a count or a non-count noun, students can use the quantifier “some” over the article , “a” or “the.” Likewise, in the second sentence, if “earth” is ambiguous for students, the alternative determiner “our” can be used instead of one of the three articles, which was not possible in the multiple-choice cloze tests. 1) Examples a) I would bring some rice cake. b) We need to find something to save our earth.

PAGE 106

92 For further research on English determiners, therefore, type of data collection should be controlled for a more precise picture of determiner use. General Research Question The general research question asked if chat is an appropriate forum to ask learners to focus on and use determiners and other function words. The answer to the first research question showed that there was no significant effect of the act of chat on the scores and error types of pretest and posttest. Additionally, the answer to the second research question showed that there was no significant effect of the act of chat on the ratio of errors to total head nouns (Err/N) in either task type. On the other hand, the results showed that conscious attention to grammar during chat resulted in a significant decrease in the ratio of deletions to total errors (Del/Err) in both task types. It seems that it is possible to ask students to focus on determiners and other function words during certain tasks with the hope of helping them increase their accuracy in the L2. Nevertheless, the overall ratio of deletions in chat was too high for us to conclude that chat could be a place where students can practice determiners, regardless of meta-awareness. Given this result, the nature of chat (i.e., the omission of function words) and students’ proficiency need to be considered, as discussed earlier. Chat may be used as a place where students develop other language skills, such as vocabulary, interactional skills, and grammatical competence other than function words, as suggested in earlier research (e.g., Blake, 2000; Salaberry, 2000; Pellettieri, 2000; Kitade, 2000; Lee, 2002). However, it seems difficult for teachers to use chat as a forum to ask students at the intermediate (or lower) level of English, as in my study, to focus on and use English articles for educational purposes.

PAGE 107

93 Implications There are several pedagogical implications to be taken from these findings. First, it was shown here that intermediate L2 readers have a tendency not to attend to form while reading for comprehension even in textually manipulated readings, and merely reading for the content did not promote further cognitive process of the target forms. Therefore, as Leow (2001) suggests, teachers need to design activities and tasks that promote learners’ noticing of targeted L2 forms or structures while interacting with the L2, because the higher level of awareness should ultimately lead to more learning. If such a task environment is not available, teachers need to encourage students to consciously attend to enhanced parts of a reading in addition to attending to the content of material for comprehension. In my study, accidental learning seems to be an unrealistic goal, at least, for intermediate learners as far as grammatical features like English articles are concerned. Another implication regards the use of chat in the classroom. It seems difficult for ESL teachers to use chat for the purpose of developing English articles, at least for the learners in my study. This is because although meta-awareness during chat is somewhat responsible for students’ deletion of determiners, given the telegraphic nature of chat and students’ proficiency (i.e., intermediate or lower), deletion of articles seems to be unavoidable throughout chat conversations. From a research perspective, we can also suggest that using chat transcripts to investigate theoretical aspects of acquisition of English articles, such as the sequence of acquisition and acquisition order, may not provide adequate data. This is because the learners’ developmental state may not be accurately reflected in the chat transcript due to

PAGE 108

94 frequent deletion of articles, while research on the acquisition order of articles often deals with the deletion/retention of articles and the overuse of “the” and “a(n)” for the analysis. Limitations and Suggestions for Future Research My study has methodological limitations that need to be considered for future research: Participants. A relatively small number of participants (four groups of 16 participants for a total of 64) were utilized here. A larger number of participants in each group would improve the robustness of the findings for future research, and allow for greater and more varied comparisons between groups. In addition, the proficiency level of the students in my study may not be reliable because no formal test was conducted to measure the students’ level of proficiency. Their proficiency levels were interpreted as intermediate or higher, according to Inha University criteria. However, two native proofreaders reported that many of the students had not reached this level, judging from the students’ chat conversations. Because my study did not control students’ proficiency level, that might have been an intervening variable for the outcome. For future research, a formal test is needed to confirm the students’ proficiency level. Treatment. The length of treatment for the three experimental groups was only four weeks. However, English articles are difficult for Korean ESL students to learn, and it could take longer periods of treatment to fully investigate the development of this feature. Although certain groups showed some changes in the use of determiners within this period, longer periods of treatment would provide a more accurate picture of the acquisition of determiners. Additionally, the different lengths of chat treatments for Group C (2 hours) and Group EC (1 hour) could have been a problem. Because two one-hour chat treatments were modified into one two-hour treatment for Group C, the students of Group C might have experienced more fatigue than those of Group EC during

PAGE 109

95 chat sessions. More fatigue might have affected Group C for attending to grammar during the chat sessions. For future research, the same conditions should be established for each group. Method of collecting information. Previous research on this issue often employed think-aloud protocols to ensure that focal attention was indeed being directed toward the target item (e.g., Leow, 2001, 2003; Rosa & O’Neil, 1999). However, my study utilized a questionnaire for students to self-report on their own awareness, because the primary focus of study was on chat and treatment type rather than textual enhancement and meta-awareness. Since the questionnaire may be too subjective and easy to misinterpret according to the students’ own criteria, more objective methods like think-aloud protocols are needed for future research. Method of data analysis. My study adapted the data analysis method from Huebner (1983, 1985), Master (1988), Thomas (1989) and Trademan (2002), and examined the ratio of deletion/retention of determiners and the overuse of “the” and “a(n)” for the difference. Although my study employed the methods, deletion and retention, in the same way, the semantic/discourse matrix used for overuse of “the” and “a(n)” was not considered in my study. A detailed discourse analysis that considers the semantic features of hearer knowledge ( + HK) and specific reference ( + SR) could have provided a more accurate picture for the overuse of “the” and “a(n)” for thorough analyses. Topic familiarity. While previous research claims that topic familiarity affects learning of L2 forms (e.g., Overstreet, 1998, 2003), my study did not address this variable, and simple assumed that the eight fairy tales were unknown to the participants.

PAGE 110

96 However, it is likely that some students were familiar with some of these stories, which might have impacted the results. For example, it was shown that it can decrease the demand on the processing resources of humans. Specifically, when a familiar L2 passage is given to the learner, the learner processes the L2 forms more easily than when an unfamiliar one is provided (e.g., Lesser, 2003; Pulido, 2003, 2004). For this reason, future research should control carefully for topic familiarity. Measurement of noticing. My study assumes that a gain in scores is a result of noticing, which is a result of awareness. Statistical analyses (e.g., independent samples tests and Regressions) determined that meta-awareness during input enhancement did help students to gain more in scores and to lower the ratio of deletion to total errors simultaneously. Given the research which suggests that the deletion of determiners is mostly found in lowest proficiency learners’ performance as one pattern for the sequence of acquisition (e.g., Simons, 2001), my study assumes that the decrease of deletion rate reflects their improvement. However, the claim that gain of scores is a result of noticing could be shaky as long as there is no research conducted to back up the links. Conclusions The findings of my study lead to the following conclusions: First, we can conclude that chat is unlikely to be an appropriate forum to ask learners to focus on and use English determiners, particularly articles. In an effort to optimize the advantages of chat despite its primary disadvantage of encouraging abbreviated discourse, my study employed input enhancement sessions in the hope that they would induce students to focus on and use the determiners during the chat sessions. However, the results showed that input enhancement did not promote students’ usage of articles during chat. Rather, the students’ meta-awareness seemed to be more related to

PAGE 111

97 individual factors. Even though students’ meta-awareness resulted in a significant difference in the ratio of deletions to total errors (Del/Err), the overall ratio of deletions to total errors (Del/Err) were very high throughout chat conversations in both task types. Chat may not be a good forum to teach, practice, or test certain grammatical features. In addition, students’ L2 proficiency seems to be an important factor to be considered, because students at the intermediate (or lower) level of English might have difficulty in focusing on the less-urgent grammatical features (i.e., articles), as their focus is on meaning. Our second major conclusion is that textual enhancement alone may not guarantee learners’ noticing of determiners. This is because during reading students often attend to the content of passages for comprehension rather than also attending to form, and this tendency was also consistent during input enhancement. On the other hand, given that input enhancement did prove effective for noticing the target forms when accompanied by meta-awareness, we suggest that meta-awareness is necessary for noticing targeted forms during input sessions.

PAGE 112

APPENDIX A INFORMED CONSENT Informed Consent Protocol Title: Developing English determiners through Internet chat: An experiment with EFL students Please read this consent document carefully before you decide to participate in this study. Purpose of the research study: The purpose of this study is to investigate if text-based online chat helps EFL learners aware of and learn the usage of English determiners. What you will be asked to do in the study: You will be asked to perform a series of tasks that deal with English grammar throughout eight weeks. There will be four groups of 16. In the first week, all of you will take a pretest. Then, two groups out of three will conduct eight one-hour treatments once per week as homework. Each student in the first group will be required to read a set of passages, in which some grammatical features are highlighted and some others are additionally underlined. And then you will also be required to write a one-paragraph summary. Each student of the second group will meet with his or her partner in the assigned chat room every week, when each pair will be assigned some tasks. The third group will alternate passage reading and chat session each week. When all treatments are completed, all of you will take the posttest. Time required: 30 min (pretest) + one hour x 8 (treatments) + 30 min (posttest) = total 9 hours. Risks and Benefits: This project involves no risk but potential benefits in that the participants will get chances to be exposed to and practice English throughout the treatments. Compensation: No compensation Confidentiality: Your identity will be kept confidential to the extent provided by law. Your information will be assigned a code number. The list connecting your name to this number will be kept in a locked file. Your name will not be used in any report. Voluntary participation: Your participation will be conducted as homework activities and, accordingly, will be part of your final grade. Right to withdraw from the study: Your participation will be conducted as a homework assignment, but you have the option as to whether or not you want your information to be used as part of a research study. Whom to contact if you have question about the study: Jongbum Ha , Ph.D. Candidate, Program in Linguistics, UF. 391 Maguire VLG #5, Gainesville, FL 32603, 846-5422, e-mail: jbha@ufl.edu 98

PAGE 113

99 Dr. Roger Thompson, Department of English, UF, P.O. Box 117310, 392-6650 ext. 263 e-mail: rthompso@english.ufl.edu Whom to contact about your rights as a research participant in the study: UFIRB Office, Box 112250, University of Florida, Gainesville, FL 32611-2250; ph 392-0433. Agreement: I have read the procedure described above. I voluntarily agree to participate in the procedure and have received a copy of this description. Participant: ________________________________________ Date: ________________ Principal Investigator: _______________________________ Date: ________________

PAGE 114

APPENDIX B QUESTIONNAIRE Background Questionnaire, Korean Translation What group are you in? 1) Reading 2) Chat 3) Chat & Reading 4) Control What is your name? What is your email address? In what year were you born? What is your gender? ( M / F ) What is your major field of study? At what age did you begin to learn English? For how many years have you been studying English? Have you been in a foreign country where English is spoken? If so, where and how long? Have you ever studied a foreign language(s) other than English? If so, which language and for how long? What condition do you prefer when you study / learn? 1) Interactive 2) Non-interactive 3) No preference 1) Oral mode 2) Written mode 3) No preference 1) Collaborative 2) Individual 3) No preference Only for Group C and Group EC How often do you participate in the text-based online chat in Korean? 1) Almost everyday 2) Between two to five times per week 3) Once a week 4) Less than once per week 5) Never For what purposes do you usually chat in Korean? How often do you participate in the text-based online chat in English? 1) Almost everyday 2) Between two to five times per week 3) Once a week 4) Less than once per week 5) Never For what purposes do you usually chat in English? Are you comfortable with typing in English? 1) Comfortable 2) Not comfortable 100

PAGE 115

101 1

PAGE 116

102 Follow-up Questionnaire, Korean Translation What group are you in? 1) Reading 2) Chat 3) Chat & Reading What is your name? What is your email address? Only for Group E and Group EC When you read the passages, how much did the bold and highlighted words draw your attention? 1) A lot 2) A little 3) Not at all Do you think the materials were difficult to understand? 1) Yes 2) No Only for Group C and Group EC Did you enjoy your chatting throughout the experiment? 1) Yes 2) No If not, what made you feel so (e.g., My partner was not sincere at all; The activities were too difficult (or too easy); I had little motivation)? Did you consciously focus on the use of English determiners during the chat activities? 1) Yes 2) No If you chose ‘Yes’, how much did the act of consciously focusing on grammar affect your comprehension and production during the chat activities? 1) A lot 2) A little 3) Not at all Did you consciously focus on the correct use of grammar during the chat activities? 1) Yes 2) No If you chose ‘Yes’, how much did the act of consciously focusing on grammar affect your comprehension and production during the chat activities? 1) A lot 2) A little 3) Not at all If you chose ‘No’, why did you not focus on it?

PAGE 117

103 2

PAGE 118

APPENDIX C PRETEST AND POSTTEST Pretest Circle the right answer. The Snow Queen Once upon a time (a / the / / any) magician made (a / the / / some) magic mirror. It was (a / the / / that) mirror of opposites. If (a / the / / some) kind face looked into (a / the / / its) mirror, (a / that / / some) wicked face looked back. If (a / the / / its) loving look was cast at (a / the / / its) mirror, (a / the / / its) look of hate was reflected. One day, (a / the / / its) mirror broke. If (a / the / / its) sliver of glass from (a / the / / its) mirror entered someone’s eye, that person became evil; if another sliver pierced (a / the / / its) heart, (a / its / / that) heart grew hard and cruel. (Those / The / / Some) two children, Karl and Gerda, were very (those / the / / their) close friends. One evening, Karl was watching the snow fall when he noticed (a / the / / any) white flake slowly turn into a beautiful ice maiden. Karl was startled to hear the ice maiden speak (a / the / / his) name. He didn’t know he was looking at the Snow Queen. Spring came and one afternoon, as Karl and Gerda looked at (a / the / / some) book, the little boy told her, “I feel a pain in (a / that / / my) heart! And something’s pricking my eye!” “Don’t worry,” said Gerda comfortingly. “I don’t see anything!” But, unfortunately, splinters from the shattered mirror had pierced the little boy. Now he was under (an / the / / any) evil spell. Because of this, he snapped at his best friend, “You’re so ugly!” Ripping (any / the / / these) two roses from her rosebush, he ran off. From that day on, Karl turned into (a / the / / that) very nasty boy, and nobody knew what had happened to cause this. Only Gerda still loved him, though all she got in return were insults and angry words. Winter came again, bringing far more (a / the / / some) snow than anyone could remember. One day, just after going outdoors to play in (a / the / / some) snow, Karl saw (a / the / / each) beautiful maiden he had seen before. She was coming toward him wrapped in (a / the / / some) luxurious white fur coat. She stood in front of him and told him to tie his sled to (a / the / / her) own, which was drawn by (a / the / / some) white horse, and they sped away. Suddenly, (a / the / / its) great sled soared into (a / the / / some) sky and through (each / the / / any) clouds. Stretched out on his own little sled, Karl didn’t 104

PAGE 119

105 dare move (a / the / / some) muscle for (a / the / / that) fear of falling into space. At last, they came to a halt on a huge white plain, dotted with lots of sparkling frozen lakes. “Come into (some / the / / my) arms,” said the Snow Queen, opening her soft fur coat. “Come and keep warm!” Karl allowed himself to be hugged by (a / the / / each) unknown maiden, and (a / the / / some) chill ran up his spine as (a / the / / some) two icy lips touched his forehead. The Snow Queen kissed him again, and in an instant, the little boy forgot all about Gerda and his past life and fell into (a / the / / his) deep sleep. In the meantime, Gerda was anxiously searching for Karl, but (some / the / / no) one had seen him. Finally, she went down to (a / the / / some) river. “Great River:’ she said, “please tell me if you’ve seen Karl or if you’ve carried him away! I’ll give you these, if you do!” And she threw her shoes into the river. But the river’s swift current paid no attention to her and just swept (a / the / / some) shoes back to (a / the / / each) bank. Not far away stood (an / the / / each) old boat and Gerda climbed into it. As she drifted with the current, she pleaded, “Great River, take me to Karl.” As night fell, she stopped by (a / the / / some) riverbank carpeted with all kinds of (those / the / / any) colorful flowers. After resting, she went into (a / the / / her) forest. Although she did not know how she would ever find her friend, (a / the / / each) mysterious voice inside her told her to be brave. After wandering for hours, Gerda stopped, tired and hungry. A crow flapped out from a hollow tree. “Caw! Caw! If you’re looking for Karl,” it said, “I know where he is! I saw him with the Snow Queen on her sled in the sky!” “And where is her kingdom?” Gerda asked the crow. “In Lapland, where all is icy cold. (A / That / / Some) reindeer over there might take you!” Gerda ran over to the big reindeer, threw her arms around his neck, and, laying her cheek against his soft muzzle, said, “Please help me to find my friend!” The reindeer’s kindly eyes told her that he would, and she climbed onto his back. They traveled ‘till they came to the frozen tundra, lit by the fiery glow of the Northern Lights. “Karl! Karl! Where are you?” shouted Gerda. When, at last, she found the little boy, Karl was still in (a / the / / some) deep sleep of (a / the / / some) wicked spell. Gerda threw her arm around him, and (any / the / / each) teardrops dripped onto his chest and heart. (Any / The / / some) tears washed away the slivers of glass, and (a / the / / each) evil spell was broken. Karl woke from (each / some / / his) long sleep, and when he set (a / the / / some) eyes on Gerda, he too began to cry. They had found (an / the / / each) other again at last, thanks to Gerda’s love, and the reindeer carried them home. From then on, they remained close friends, happily ever after.

PAGE 120

106 Posttest Circle the right answer Tom Thumb Once upon a time, in the days of King Arthur, there lived (a / the / / some) very wise wizard named Merlin. He knew all the fairies, and even (a / the / / this) fairy queen was a friend of his. One day Merlin knocked at (a / the / / that) door of (a / the / / that) small cottage and asked for (a / the / any / some) food. He looked so hungry that the farmer and (a / this / / his) wife took pity on him. They not only gave him a bowl of (a / the / / that) milk with (a / the / its / some) tasty bread, but they said he could spend the night in their home. Merlin saw that the farmer and his wife were very sad. “Why are you so sad?” asked Merlin. “Oh!” said the woman, “we are unhappy because we have (a / the / / no) children. I would be the happiest woman in the world if I had (a / the / / some) son. Why, even if he were no bigger than my husband’s thumb, we would love him dearly?’ “That would be a very unique kind of child:’ said Merlin, “but I hope (a / the / / your) wish comes true?’ Then Merlin went on his way to visit the queen of (a / the / / those) fairies. When he came to her castle, he told the fairy (a / the / / some) wish of the farmer’s wife. The queen of (a / the / / some) fairies said, “The good woman shall have her wish. I will give her a son (a / the / / that) size of her husband’s thumb.” Soon after this the farmer’s wife had a son—exactly (a / the / / that) size of his father’s thumb. People came from far and wide to see (a / the / / each) famous tiny boy. One day the fairy queen and some other fairies came to see him. The queen kissed the little boy and named him Tom Thumb. Tom never grew any larger than (a / the / / any) man’s thumb, but he got into quite a bit of mischief. One day his mother was mixing a cake. Tom leaned over (a / the / / that) edge of (a / the / / that) bowl to see and fell in, headfirst. His mother did not see him fall, and she kept stirring. Tom kicked and kicked inside the batter, and it moved and tossed about. His mother was afraid. “There must be gremlins in it’ she said. She went to (a / the / / that) window to throw the batter out. Just then (a / the / / some) poor beggar was passing by. “Here is (a / the / / some) batter you may have, if you like:’ said Tom’s mother. The beggar thanked her and took it. He had not gone very far, when Tom got his head out of the batter and shouted, “Take me out! Take me out!” (A / The / / Each) poor beggar was so frightened that he dropped the batter and ran off. Tom crawled out of the batter and ran home where his mother scrubbed him thoroughly and put him to bed. Another time, Tom’s mother took him with her when she went to milk the cow. So she wouldn’t lose him, she tied him to (a / the / / some) piece of hay. When Tom’s

PAGE 121

107 mother was not looking, the cow took the wisp of (a / the / / its) hay into her mouth. She began to chew and chew. Tom began to jump and shout. He frightened the cow, so she opened her great mouth and Tom jumped out. Then Tom’s mother took him in her apron and ran with him to (a / the / / that) house, but, fortunately, he was not hurt. One day Tom was in the field helping his father. “Let me drive the horse home:’ said Tom. “You drive the horse?” said the father. “How could you hold (a / the / / some) reins?” “I could stand in the horse’s ear and tell him which way to go:’ said Tom. So his father put him in the horse’s ear, and he got them home safely. “Mother! Mother!” cried Tom. But when Tom’s mother came out, she could see (a / the / / no) one. She began to be afraid. ‘Where are you, Tom?” she cried. “Here I am in (a / the / / his) horse’s ear. Please take me down:’ said Tom. His mother lifted him gently down, kissed him, and gave him a plump blueberry for supper. Tom’s father made him a whip out of (a / the / / that) straw. Tom tried to drive (a / the / / those) cows, but he fell into a deep ditch. There a great bird saw him and thought he was a mouse. The bird seized Tom in (a / the / / her) claws and carried him toward (a / the / / her) nest. As they were passing over (a / the / / that) sea, Tom got away and fell into (a / the / / that) water, where a great fish swallowed him. Soon after this, the fish was caught, and it was such a big one that it was sent at once to King Arthur. When the cook cut open the fish, out jumped Tom Thumb. Tom was brought before the king and he told (a / that / / his) story. The king grew very fond of Tom and took Tom with him wherever he went. If it began to rain, Tom would creep into the king’s pocket. In the hot sun, he also found shade in the king’s pockets. The king had a new suit made for Tom and gave him a needle for (a / the / / each) sword. A mouse was trained for Tom to ride. The king and (a / the / / its) queen never tired of seeing him ride (a / its / / his) little mouse-horse and bravely wave his sword. One day, as they were going hunting, a cat jumped out and caught Tom’s mouse. Tom drew his needle-sword and tried to drive the cat away. The king ran to help poor Tom, and the cat ran away. Tom was scratched and bitten badly, but he did not die. Soon he was well again, and fought (a / the / / those) many brave battles and did (a / the / / those) many brave deeds to please the king. And, several times a year, the king took Tom to see his parents, for he always loved his dear mother and father.

PAGE 122

APPENDIX D MATERIALS FOR GROUP E Week 1 The Frog Prince Breaking enchantments is not an easy business. In “ The Frog Prince,” a beautiful princess must be persuaded to kiss a wet, slimy frog. Would you be able to kiss a frog or fall in love with the beast? Or would the Frog Prince be destined to live his life hopping from lily pad to lily pad? Once upon a time, there was a princess who loved to play with her golden ball. She couldn’t wait to throw the ball high into the air and then catch it. Every day she would play for hours in the garden with her ball. She thought of little other than her golden ball. One day, she decided to see how high she could throw the ball. She threw it up high and caught it easily. “That’s too easy;’ she thought, so she threw it even higher and still it was too easy to catch. Finally, she gave the ball a great heave and threw it so high that she lost sight of it. When it came down, she didn’t catch it, and the ball rolled away and landed, ker-splashhh, in a nearby pond. She stood at the edge of the pond, but could not see her ball anywhere. “Excuse me, Princess;’ said a voice, “I can get your ball for you.” The princess looked around to see who was speaking. All she could see was a very slimy and warty frog. “Did you say something to me?” the princess asked the frog, for these were magical times and frogs and other animals were known to speak occasionally. “Yes;’ croaked the frog. “I said that I could get your ball for you?’ “I’m not used to talking with warty frogs,” said the princess arrogantly. “Well, then,” said the frog indignantly, “you shall never see your ball again.” And he turned to hop away. “Wait, wait!” cried the princess. “I’m sorry. Will you please get my ball for me?” “On one condition;’ said the frog. “When I retrieve the ball, you will give me one small kiss?’ The princess felt that she had little choice since she desperately wanted her golden ball back. So she grudgingly agreed to this condition. The frog paddled down into the pond and in the wink of an eye came back with the ball. “Here it is!” he said, dropping the ball at the princess ’s feet. “Thank you, good Frog,” she said, and she began tossing the ball into the air. “Wait,” said the frog. “Aren’t you forgetting your promise?” 108

PAGE 123

109 “Oh;’ said the princess. “You can’t really expect me to kiss you, can you? After all, I am royalty and you are warty!” With that, she ran away, returning to the palace. That evening at dinner, a servant came into the dining room. “There is a frog here to see you, Your Majesty?’ “ A what?” said the king. “ A what?” said the princess. “ A frog;’ said the servant. “He says he has business with the princess. Something about a broken promise?’ “Send him away immediately!” said the princess. “Bring him in;’ said the king. A few minutes later, in hopped the frog. “Do you know this frog?” said the king. “I rescued your daughter ’s golden ball;’ said the frog. “But she has broken her promise to me?’ “Is that true?” asked the king. “He wanted me to kiss him,” said the princess. “I cannot stand the thought of touching my lips to his green mouth!” “ No daughter of mine will break her promise;’ said the king. “Do as you must.” “No, sire,” said the frog. “She must do it voluntarily.” “Never!” shouted the princess. This frog was a persistent fellow, though, so for many weeks, he hopped to the palace every morning and sat with the princess. They ate and played together. At the end of six months, as they were going up to bed, the princess turned to the frog and without really thinking said, “Good night, Frog;’ and gave him a little kiss on the top of his head. In an instant, the frog vanished, and in his place was a handsome man. “Who are you?” said the princess, stepping back. “What have you done with my frog?” For by now, the princess loved the little frog, warts and all. “I am your frog,” said the man. “I am also a prince. I was enchanted by a wicked fairy and turned into a frog. Only the kiss of a beautiful princess could break the enchantment.” “Why didn’t you tell me before?” said the princess. “Because;’ said the princess, “you had to do it of your own free will.” The two were soon married, and the princess grew to love the princess just as much as she had loved the frog. The princess and princess lived happily together and had many children who loved to play with the golden ball and hear the story of how their parents met.

PAGE 124

110 The Little Pear Girl Once upon a time, there lived a peasant who worked hard to make a living by farming the land. Every year his pear tree produced four bountiful baskets of fruit, which had to be given to the king. He was a greedy ruler who grew rich by taking from his subjects. One year, part of the pear harvest went bad, and the peasant was able to pick only three and a half baskets of fruit. The poor man was beside himself with fear, for the king refused to take less than four baskets. The peasant knew he would be punished. All he could do was put his youngest daughter into one of the baskets and cover her with a layer of pears, so that the basket looked full. The king ’s servants took away the four baskets without noticing the trick. Soon the little girl found herself all alone in the pantry, under the pears. One day, the cook went into the pantry and discovered her. Nobody could understand where she had come from. Not knowing what to do with her, Cook decided she should become a maid in the castle. She was given the name Violetta because her eyes were the color of violets. Violetta was a pretty girl, sweet and generous. One day, as she was watering the flowers in the royal gardens, she met the king ’s son, a boy of her own age, and the two became friends. The other maids, jealous of Violetta ’s beauty, did everything they could to get her into trouble. They started spreading nasty rumors about her. One day, the king sent for her and said severely, “I’m told you boast of being able to steal the witche s’ treasure trove. Is that true?” ‘Violetta said no, but the king refused to believe her and drove her out of his kingdom. “You may return only when you have laid hands on the treasure,” he said. All of Violetta ’s close friends, including the prince, were sorry to hear of the king ’s decision, but they could do nothing to stop her going. The girl wandered through the forest and, when she finally came to a pear tree, she climbed into its branches and fell asleep. She was awakened at dawn by an old woman calling her, “What are you doing up there, all by yourself?” Violetta told the old woman her tale, and the old woman offered to help her. She gave Violetta a broom, some round loaves of good bread, a little oil, and some good advice. Again, the girl set off on her journey. Soon she reached a clearing with a large wood stove where she saw three women tearing out their hair and using it to sweep the ashes from the stove. Violetta offered them her broom and, in return, the women pointed out the way to the witche s’ palace. As she headed toward the palace, two hungry mastiffs blocked her path. Violetta threw them the loaves. The dogs ate the bread and let her pass. Then she came to the bank of a flooding river. Remembering the old woman ’s advice, Violetta sang a lullaby to the river to calm it. The minute her song wafted into the air, the water stopped flowing. She crossed the river and at last reached the witche s’ palace. The door was unlocked, but Violetta could not push it open for the hinges were rusted. So she rubbed a little oil on them and the door swung open. The little girl walked through the empty halls until she came to a splendid room in which sat a magnificent coffer full of jewels. Holding the coffer under her arm, Violetta made

PAGE 125

111 for the door. But the coffer was enchanted and it cried out, “Door! Don’t let her out! You don’t know what she’s about!” However, the door opened, for Violetta had oiled its hinges. Down at the river, the coffer cried out again. This time it said, “Water! Drown her! Please, don’t deter.” But the river did not stop the little girl from crossing. The two mastiffs did not attack, and the three strange women did not burn her in their stove, for each was repaying the girl ’s courtesy. Back at the king ’s palace again, the prince ran happily to meet Violetta, telling her, “When my father asks you what you want as a reward, ask him for the basket of pears in the pantry!” And this Violetta did. Pleased at paying such a modest price, the king instantly ordered the humble basket to be brought forth. But nobody ever imagined for a minute that underneath the pears lay the handsome prince. The young man came out of his hiding place, professed his love for Violetta, and said he wanted to marry her. The king gave his consent. They all agreed that when Violetta became old enough, she and the Prince would marry. Happy with this arrangement, Violetta brought her family to the palace, and they all began a new and happy life together.

PAGE 126

112 Week 2 Rapunzel Once upon a time, a man and a woman lived in a snug home near a beautiful walled garden. They were quite happy, except for one thing: They had no children. Finally, after years of waiting, the wife became pregnant. Oddly enough, the woman found that she craved something called rampion, a green for making salads. Rampion is also called Rapunzel. Her husband spied some inside the walls of the garden next door. Although he’d never met the person who tended the garden, he thought he’d try his luck at getting some for his hungry wife. That night, by the light of the moon, he climbed over the stone wall into the beautiful garden and picked a small basketful of the rampion. He rushed back over the wall and gave the leaves to his wife who ate them happily. The next night she craved more, so he was forced to make another trip over the wall into the neighbor ’s garden. This time though, he was caught in the act—by an old, warty fairy. “How dare you steal into my garden and take my rampion?” she hissed. “Oh, I’m sorry,” he said. “But I felt I had to. My wife is pregnant and craving your delicious rampion. You’re quite a gardener!” “I think she’ll have a daughter:’ said the fairy. “I was planning to kill you, but I will spare you on one condition?’ “You name it,” said the man. “ Your wife shall have as much rampion as she likes, and you shall live:’ said the fairy, “but when your daughter is twelve years old, you will give her to me. Then, you are never to see her again.” The man agreed, because he was afraid the fairy would kill him otherwise. When he told his wife, she cried, but she agreed that her husband really had no choice. When the child was born, the hideous fairy appeared and gave her the name Rapunzel. Then she vanished, only to reappear, as promised, on the girl ’s twelfth birthday. “I have come for you, Rapunzel’ said the fairy. The girl agreed to go because she did not want her father to die. The fairy shut Rapunzel up in a stone tower in the middle of a forest. There was no door to this tower, and no stairs. There was only one window high at the top, far too high for any ladder to reach. When the fairy wanted to go up, she stood at the bottom and cried in her hoarse voice, “Rapunzel, Rapunzel, let down your golden hair.” Rapunzel, who had spectacularly long hair, would drop her luxurious locks down to the fairy standing far below. Then the woman would climb up the hair. After a few years, a prince riding through the forest happened to hear a beautiful voice singing. He hurried toward the sound and saw the tower. In the window far above, he saw a beautiful girl, singing sweetly to herself The prince wanted to climb the tower, but he saw right away that it wasn’t possible. There was no door or stairs. He rode home, but every day he came back to the

PAGE 127

113 tower and listened to the girl ’s sweet song and waited for a glimpse of her face in the window. One day while he was waiting, he saw the fairy arrive and croak, “Rapunzel, Rapunzel, let down your golden hair?’ As he watched the wicked woman climb the hair, he came up with a plan to reach the beautiful maiden. The next day when it grew dark, he went to the tower and croaked like the old fairy, “Rapunzel, Rapunzel, let down your golden hair?’ Down came the beautiful golden hair, and up he climbed. As you can imagine, the prince ’s appearance was a bit of a shock for Rapunzel. The prince reassured her, though, and told her that he had heard her singing for many months and that he had fallen in love with her through her splendid songs. Rapunzel fell in love with him as well. The prince asked her to marry him, and she happily said yes. “But how will you escape from this prison?” the prince asked. “Bring a length of silk thread with you every time you visit;’ Rapunzel said. “I will weave a thread ladder, and when it is ready I will come with you?’ So, every evening, the prince returned with a fine silk thread, and as they talked she wove the ladder of silk. Finally, the ladder was finished. It had not been easy hiding it from the old fairy, for she was quite suspicious and questioned Rapunzel harshly about strange human smells in the tower. Rapunzel, however, was able to convince her that nothing was amiss. One night, by the glowing light of the moon, the prince came to rescue Rapunzel. She hung the delicate ladder from the windowsill and climbed down carefully into the arms of her waiting prince charming! They were married shortly after at the prince ’s royal residence and lived there very happily. As for the ugly fairy, to this day she can’t understand how Rapunzel escaped. And she’s quite vexed about it!

PAGE 128

114 The Magic Tinderbox In the hundreds of years since this fairy tale was written, the meaning of the word tinderbox has changed. In the time of the soldier in “ The Magic Tinderbox,” the word tinderbox referred to a metal box that was used to hold flint. Flint is a material that produces sparks when struck by a piece of metal and was used for starting fires. Today, the word tinderbox refers to any highly flammable object or building. Maybe you’ve heard someone say something like, “ That deserted building is a tinderbox.” This means that it would burn easily. Once upon a time a brave soldier was returning from war. His only possession was his sword. As he walked through a forest on his way home, he ran into a witch, who said to him: “Hello, good soldier, would you like to earn a bag of money?” “Money? Yes, I’d do anything for money.” “Great!” said the witch. “It won’t be hard! All you have to do is go down that hollow tree till you reach a cave. There, you’ll find three doors. When you open the first door, you’ll see a big dog with eyes like plates, guarding a large chest of copper coins. Behind the second door is a treasure of silver coins, guarded by a dog with eyes the size of millstones. When you open the third door, you’ll come upon another dog, with eyes the size of a castle tower, beside a treasure of gold. Now, if you lay this old apron of mine before these dogs, they’ll crouch on it and won’t hurt you. You’ll be able to carry away all the coins you want. What do you think of that?” The soldier asked, “What do you want in return?” “Just bring me back an old tinderbox my grandfather left down there, long ago?’ So the young soldier tied a rope around his waist and he lowered himself into the hollow tree. To his surprise, he found the three doorways and the three dogs, just as the witch had said. Soon he was back, his pockets bulging with coins, but before he handed the tinderbox to the old witch, he asked her, “What do you want it for?” The witch threw herself at the soldier, screaming, “Give it to me!” When the witch attacked him, the soldier exclaimed, “So this is the thanks I get! I’ll show you!” He undid the rope from around his waist and tied up the old woman. Then off he went. When he reached the town, he said to himself, “Now I can eat at as many restaurants as I like!” After years of getting by on little pay, with his sudden wealth the soldier felt almost like a prince. He bought a new pair of boots and he went to the best tailor in town, ordering up lots of fancy new clothes. The soldier was quick to spend his newfound wealth, and the money soon ran out. When the innkeeper discovered that the soldier could no longer pay his bill, he kicked him out. So the poor soldier ended up in an attic of a deserted building. One evening, he realized he had never used the old witch ’s tinderbox. So he rubbed it, and as it sparked, the dog with the eyes like saucers suddenly appeared. “Tell me your wish:’ it said. “Bring me loads of money!” said the soldier. A second later, the dog was back with a bag of coins. Every time he rubbed the tinderbox, the dog brought him more money. Then when he rubbed it quickly twice in a row, the dog with eyes like

PAGE 129

115 millstones stood before him, carrying silver . And when the soldier rubbed the tinderbox three times in a row, the third dog came carrying gold. Rich once again, the soldier chose the best hotel in town and went back to living the high life. One day, the soldier was told that the king would not allow anyone to meet his gorgeous daughter, for he believed in a saying that the Princess ’s fate was to marry a simple soldier. That evening, the soldier rubbed the tinderbox. “Bring me the princess:’ was his new wish. Immediately the dog returned with the beautiful princess, fast asleep. The soldier kissed her. Next morning, the girl told her parents that she had had a dream that she was kissed by a soldier. The queen became suspicious, and ordered one of the ladies-in-waiting to guard her daughter every minute. The dog was seen when it came the next evening and the alarm sounded. The king ’s guards followed the dog, and the soldier was arrested and put in the dungeon. When the king asked the prisoner to appear before him and present his case, the soldier brought the tinderbox. He rubbed it and the three dogs appeared. Amazed at this, the king agreed that the soldier and his daughter should marry. They lived happily every after, occasionally using the tinderbox to charm the people of the kingdom.

PAGE 130

116 Week 3 The Tale of the First Sheik And so, the first sheik began: Once upon a time, oh Genie, this gazelle was my wife. She also happened to be a magician. We had been married for nearly thirty years, but we had no children. As was allowed then, I married another woman and soon had a baby boy. My first wife was very jealous of the young boy and his mother. When the boy was fifteen years old, I had to go on a trip. While I was gone, my wicked wife used her magic to change my son into a calf and his mother into a cow. When I returned, she told me, “ Your second wife is dead, and your son ran away because he doesn’t like you. I don’t know where he went?’ For a year I grieved and ate practically nothing. Then, as an important feast approached, I ordered my herdsman to bring me a fat cow to butcher. He brought me the one that had been my second wife. I rolled up my sleeves and took the knife to kill it, but the cow began to weep. I had never seen a cow cry, and I said to my herdsman, “I can’t kill this cow. Please, bring me another.” He said to me, “This is the best there is. I will kill her for you.” So he killed the poor cow. But instead of finding meat inside her skin, there was nothing but bones. “Bring me a calf then,” I ordered. The herdsman brought to me the calf that was my son. The calf also began to cry. “Do I need to kill this one, too?” said the herdsman. “No’ I cried, “I will do it myself?’ I took the knife in my hand and held it to the throat of the calf. Just then Scheherazade noticed the light of dawn creeping in through the window, and stopped talking. “What a wonderful story!” Dunyazad said. “Can you tell me the end?” “If the king wills me to live another day;’ the brave girl replied, “I will finish the tale.” The king, who was held spellbound by the tale, agreed. “You may live until I have heard the rest of your story?’ That morning, the vizier went to the bedroom, prepared to kill his beloved daughter. He was astonished when the king did not give him the order but went about the business of the day. When the second night finally fell, Dunyazad said to her sister, Scheherazade, “Please, finish the story of the merchant and the genie.” “If the king will let me;’ said Scheherazade. “Begin;’ said the king. So, she began: Just as the merchant was about to kill the calf, he saw it weeping and he couldn’t bring himself to use the knife. “Keep this calf among my cattle,” he told his herdsman. All of this, the sheik told the genie, who listened in wonder. “Oh, Genie;’ the sheik said, “While all this occurred, my wicked wife watched and laughed at her revenge. She turned to me and ordered me to kill the calf, but I would not. Instead I sent it away with my herdsman?’

PAGE 131

117 The next day the herdsman returned. “Oh, my master;’ he said. “I have good news for you. My daughter learned magic in her childhood from an old woman. Yesterday, when I brought that calf home, she took one look at it and said, ‘ This calf is the son of our master, who has been bewitched by his wicked stepmother. The first fat cow that was killed was the boy ’s poor mother.” “Is this true? Bring her to me now! Bring the calf as well?’ The herdsman brought his daughter and the calf, and all stood before me. “Oh, please,” I begged, “will you release my son? I will give you every thing I own?’ The good girl smiled and answered, “I don’t want your riches, but I have two conditions. First, I would like to marry your son. Second, I would like to bewitch the witch who bewitched him.” The sheik immediately agreed. The herdsman ’s daughter took a cup and filled it with a magic potion she kept in a vial. Then she chanted words the sheik could not understand. Instantly, the calf shivered and became a man! “Then, Genie;’ said the sheik, “I married the girl to my son. She ordered my wicked wife brought before us. Then she transformed my wife into this gazelle, who travels with me even unto this day. You see her standing here.” “And that is the end of my story?’ The genie turned to the first sheik and said, “I like your tale. I gladly give you half of this merchant ’s blood!” The genie stood and held his sword high. “Wait, oh Genie;’ said the second sheik. “If I tell you an even better tale, about these two dogs, and about what happened to me because of my brothers, will you grant me the other half of this man ’s blood?” “If your story is even better;’ replied the genie.

PAGE 132

118 The Country Mouse and the City Mouse Once upon a time there was a city mouse that, having tired of the hustle and bustle of his neighborhood, took a trip to the country. Once there, he met a country mouse. They spent the day together and became friends. The country mouse took his new friend into the lush meadows and vegetable gardens. The city mouse sampled all of the vegetables and found them to be very good. Never having seen the beauties of the country, the city mouse was thrilled. But the country mouse ’s plain food wasn’t nearly as fine as his own usual meals of fine cheeses and hams found in the home where he lived. To thank his friend for the lovely outing, he invited the country mouse to visit him in town. And when the country mouse saw the pantry at his friend ’s house—full of hams, cheese, oil, flour, honey, jam, and stacks of other delicious goodies—he stood speechless. After he got over his surprise, he said, “I’ve never seen anything like it! Are all those wonderful things for eating?” “Of course!” came the reply. “You’re my guest. Dig in!” So, the pair began to feast. The country mouse tried not to stuff himself, taking small samples of everything, as he wanted to taste each item before finding his tummy full. “You’re the luckiest mouse I’ve ever met!” said the country mouse. The city mouse was listening with delight to his friend ’s praise, when suddenly the sound of heavy footsteps interrupted their feast. “Run for it!” whispered the town mouse to his friend. They were just in time; for within an inch of them was the ungainly foot of the lady of the house. Luckily, the lady went away, and the two mice returned to enjoy their feast. “It’s all right! Come on!” reassured the town mouse. “Don’t worry, she’s gone. Now let’s have some honey! It’s delicious! Have you ever tasted it?” “Yes, once, a long time ago:’ the country mouse lied, trying to sound casual. But when he tasted it, he couldn’t contain his enthusiasm: “Wow! It’s scrumptious! By the king of mice, I’ve never eaten anything so remarkable in my life!” Suddenly there came the sound of footsteps, this time thumping heavily. The two mice fled. The man of the house had come to get some bottles; and when he saw the spilt honey, he groaned, “ Those rotten mice again! I thought I’d gotten rid of them. I’ll send the cat!” And trembling with fear, the mice hid. This time it was not only the sudden visit that had given them a fright; it was the man ’s terrifying words. The mice were so scared that they held their breath, making absolutely no sound. Then, since all remained quiet, they began to feel braver and gathered enough courage to venture out. “We can come out now! There’s nobody here!” the town mouse whispered. Suddenly, the pantry door creaked, and the two luckless mice froze in fear. Out of the dim light glowed a pair of horrid, ghostly yellow eyes. But this was no ghost. It

PAGE 133

119 was a large cat. The country mouse and the town mouse tiptoed silently back to their hiding place. They wished their pounding hearts would stop beating, for fear the cat would hear the noise they made. But, as luck would have it, the cat discovered a juicy sausage. Forgetting why his master had sent him into the pantry, he stopped to eat it. No longer hungry after that, the cat decided that he might as well leave mouse hunting for another day. Off he padded to have a nap. Now, as soon as the country mouse realized that all danger was past, he did not lose a second. He hastily shook hands with his friend, saying, “Thanks so much for everything! But I must rush off now! I can’t stand all these shocks! I’d far rather sit down to a meal of a few crumbs in peace, than face a fancy feast, surrounded by danger.”

PAGE 134

120 Week 4 The Tale of the Second Sheik And so the second sheik began: These two dogs are my brothers. When our father died, he left each of us a small fortune in gold. I opened a shop with my money and sold fine embroidered silks, brocades, and linens. My two brothers decided to use their money to travel around the world and buy and sell spices. After my brothers had been gone about a year, a beggar appeared in my shop. It took me some time to recognize that this bedraggled fellow was my brother. I took him back to my home, where he bathed. I gave him some of my fine clothes and a portion of the gold I had earned from my shop. He took this money and opened a shop of his own near mine. Not long after that, another beggar appeared at my shop. And, as had happened before, it took me some time to recognize this filthy creature as my second brother. When I realized that indeed it was my brother, I took him to my home, bathed him, and gave him some decent clothing and gold I had earned from my shop. He, too, took the gold and opened a shop of his own near mine. All went well for some time, but my brothers became consumed by wanderlust and asked me to join them on a journey around the globe. For some time, I resisted their entreaties, but finally I agreed. First, though, I asked that we all share what we had earned from our shops to finance our travels. My brothers had not saved a single gold piece! They had spent their money on parties and fun. I, on the other hand, had saved six thousand pieces. It was agreed that we would take three thousand of the gold pieces on our travels and bury the other three thousand in our town. Then, when we returned from our travels, we would have some gold with which to reopen our shops. So, we set sail and traveled around the globe. We bought and sold and traded, earning back ten gold pieces for every one that we invested. Then, one day, we came upon a maiden in tattered clothes standing on the beach. When she saw me, she kissed my hand and asked me if I could help her. “Yes, of course,” I answered and offered her some gold pieces. She responded by saying, “Marry me, then. I will be a wonderful wife?’ My heart warmed when I heard these words, and we agreed to be married. We found room for her on the ship and once again set sail. My brothers, though, became jealous of the time I was spending with my new wife and the gold I was accumulating through my savvy business dealings and trades. So, they plotted to throw my new wife and me overboard. After we were dead, they would take all of our gold. One night, the brothers crept into our cabin and threw us off the ship. At that instant, my wife turned into an Ifritah, a wonderfully powerful spirit. She picked me up from the sea, rescuing me, and carried me to an island. “Now you know that I am a genie:’ she said, “but you did not know that when we were married. You were kind to me, so I have saved your life. However, your wicked brothers have made me angry and now I must go and kill them?’

PAGE 135

121 I thanked her for saving my life but begged her not to kill my brothers. She would not relent. First, though, she picked me up and flew me back to the roof of my own house. I dug up the gold and reopened my shop. One evening, I came home and found these two dogs tied up at the door to my house. They nuzzled me and whined and acted as if they wanted to speak to me. Before I knew what had happened, my wife told me, “ These two dogs are your brothers. I would have killed them, but my sister said I should be more forgiving. So instead, my sister has turned them into two dogs. They will remain dogs for ten years.” “That is why,” concluded the sheik, “I am here, searching for my wife ’s sister, so that she may remove the curse that has turned my two brothers into dogs. That is my tale?’ Then, the genie turned to the second sheik and said, “ Your tale is even stranger than the first. I happily give you the other half of this merchant ’s blood?’ At that, the merchant tearfully embraced the two sheiks and thanked them for saving his life. So, Scheherazade continued to spin tales for one thousand and one nights. As she was telling her fanciful stories, the years passed and the king and Scheherazade had three sons together. By then, the king realized what a wonderful woman Scheherazade was and he no longer wanted to kill her. The two ruled wisely and happily. And the vizier was very relieved to know his beloved daughter was safe and content

PAGE 136

122 The Monkey-Crab War Once upon a time a kind crab was walking along sideways when she found a delicious rice ball. She picked it up with her large pincer to take home to her family. A monkey perched high on top of a persimmon tree spotted the crab. He wanted her rice ball, even though he was full from the meal of persimmons he’d just finished. “Hello!” the monkey called, “Do you want to trade your rice ball for a persimmon?” “Okay;’ the crab replied. Well, this is almost too easy, thought the monkey. So, he said, “I’m almost out of persimmons. Will you take a persimmon seed instead?” “Sure,” the crab replied. The monkey slid down the tree and picked off a persimmon seed that was stuck in his fur. He dropped it in front of the crab, stuffed her rice ball greedily in his mouth, and then ran to sit on a rock to sun himself. “Thank you, Sam-don,” the crab called to him. She took the seed back to her hole and planted it. Every day the crab tended the seed and before long a sprout popped up, and it quickly grew. Some time later, when the sprout had become a tree and was covered with blossoms, a bee pollinated them. When the blossoms fell, they left behind the start of fruit that grew into shiny green balls, and by late fall many of them had become delicious orange persimmons. The crab realized she had a problem— the fruit was far beyond the reach of her pincers. She was too short! Just then, the monkey came by. “Sam-don, Sn-don:’ the crab called. “Do you remember the seed you traded for a rice ball? It’s grown now, and the fruit is ripe. Will you help me pick some?” The monkey scampered up the tree, and stuffed a juicy orange persimmon in his mouth. “Almost ready:’ he called down, and stuffed in two more. “These aren’t too bad:’ he said as he moved up to the next branch, “delicious, in fact.” As the monkey continued to eat one persimmon after another, the crab called up to him, “Sam-don, please save me one!” “Don’t nag so much:’ the monkey growled. He plucked a hard, green persimmon and threw it down so hard that it cracked the poor crab ’s shell. The bee found the crab at the foot of the persimmon tree. He helped her back to her home and then flew off to find the rice-flour mortar. “Usu-don, Usu-don! The monkey has injured the crab!” the bee said. The mortar, cut from a tree stump years before, rolled out of the kitchen. They hurried back to the crab ’s home, but on the way they met the chestnut. “Kuri-don, Kuri-don!” the mortar said. “ The monkey has injured the crab!” When the three friends arrived, the crab told them the whole story. The mortar advised her not to worry, and the friends devised a plan. “ This monkey is a real threat, Usu-don,” the chestnut said. “Someone has to do something.” “There is no one to do it but us, Kuri-don,” the mortar replied.

PAGE 137

123 The bee flew off a little before dawn, leaving the others to watch over the crab. He returned late in the morning and reported that the monkey was out taking a walk and swinging through trees. The three left the crab in the care of her three sons and hurried to the monkey ’s home. “I’ll wait here in the back of the fire pit,” the chestnut said. “Perhaps you could wait in the water barrel, Hachi-don.” The bee flew into the water barrel, and the mortar went up into the eaves. Finally, late in the afternoon, the monkey returned. He puffed on the coals in the fire pit until they began to glow. The chestnut found himself growing hotter and hotter as he thought about the monkey ’s rude manner. Finally he burst with rage and flew out of the fire pit, striking the monkey in the eye with tremendous force and great heat. In agony, the monkey leaped to the water barrel. Once he removed the lid, the bee stung his nose. The monkey rushed to the door to escape. Just then, the heavy mortar dropped from the eaves and pinned the monkey to the dirt floor. They remained there while the chestnut explained how angry everyone was about the monkey ’s wild deeds. In the end, the monkey went with the others back to the crab ’s home. They stopped by the persimmon tree, and the monkey, accompanied by the bee, climbed up and selected four shiny, ripe, orange persimmons. Once inside the crab ’s home, the monkey pushed the fruit forward and apologized. “Kani-san, I’m sorry for being so rude and nasty to you. I won’t do it again.” The monkey was true to his word. After that, the monkey visited the crab often. He helped to pick persimmons, and he helped protect the crab and her family from other rude characters, such as he himself had once been!

PAGE 138

APPENDIX E MATERIALS FOR GROUP C For Student A Week 1 Time: two hours 1. Find out personal information of your chat partner. (e.g., Name, major, interests and hobbies, future plans after graduation) 2. Share with your partner any experiences, episodes, opinions and thoughts regarding English learning. (e.g., Why do you study English?; Do you think English is difficult to learn for Koreans? If so, why?) 3. In the given website ( http://plaza.ufl.edu/jbha/week3language.html ), you will find four pictures out of eight. Your chat partner has the other four pictures. Those pictures are in order according to their number. Cooperate with your partner and figure out the whole story. 4. Talk about your achievement, for example, getting your driver’s license, learning a new skill, getting a part-time job. Ask each other some of these questions: What had you been doing before your achievement? How did you prepare for it? Had you considered giving up before you succeeded? 5. Talk about a goal that you are working toward. What steps will you take to achieve your goal? When will you have completed each step? 6. Choose any topics you wish to discuss (if time allows). 124

PAGE 139

125 Pictures at http://plaza.ufl.edu/jbha/week3language.html Picture 1 Picture 3 Picture 5 Picture 7

PAGE 140

126 Week 2 Time: two hours 1. You are now in online chat conversation. Discuss the differences between face-to-face conversation and online chat conversation. 2. Talk about Inha University. First, talk about what might be done to improve it, and then about what shouldn’t be changed. 3. In the given website ( http://plaza.ufl.edu/jbha/week5inha.html ), there are eight differences between the pictures you and your partner have. Spot as many differences as you can. 4. Talk about a relationship that is important to you. How did you meet? What were you doing when you met? Describe some events in the relationship. 5. Talk about your life ten years from now. What will you be doing for a living? What kind of family life will you have? What hobbies will you be enjoying? What will you do to achieve these things? 6. Choose any topics you wish to discuss (if time allows). Pictures at http://plaza.ufl.edu/jbha/week5inha.html

PAGE 141

127 Week 3 Time: two hours 1. Imagine that you are planning to participate in an international food festival. Which foods would you like to bring and why? 2. Heavy traffic is a big problem in both cities and suburbs. What can we do about it? 3. In the given website ( http://plaza.ufl.edu/jbha/week11kamsa.html ), you will find three pictures out of six. Your chat partner has the other three pictures. Those pictures are in order according to their number. Cooperate with your partner and figure out the whole story. 4. Talk about how you feel in your home, office, or classroom. What makes you feel good? What makes you feel bad? What would you like to change? 5. Talk about a place you remember from your childhood. Where is it? When is the last time you went there? Did you like there? Why or why not? 6. Choose any topics you wish to discuss (if time allows). Pictures at http://plaza.ufl.edu/jbha/week11kamsa.html Picture 1

PAGE 142

128 Picture 3 Picture 5

PAGE 143

129 Week 4 Time: two hours 1. Talk about a situation in your life that you have regrets about. Describe the situation and talk about what you wish had happened and why. 2. Imagine two of you are taking a trip together to another country. You’ll be gone for several weeks. Decide where you’re going. Then make a list of things you have to do and arrange before the trip. 3. In the given website ( http://plaza.ufl.edu/jbha/week9advanced.html ), there are ten differences between the pictures you and your partner have. Spot as many differences as you can. 4. If you hadn’t been born, what would have been different for your family, friends, teammates, or community? Choose two areas of your life to discuss and talk about all the things that would be different. 5. Do you think tipping is a good system? Why or why not? Were you ever in a situation where you didn’t know what to do about a tip? 6. Choose any topics you wish to discuss (if time allows). Pictures at http://plaza.ufl.edu/jbha/week9advanced.html

PAGE 144

130 For Student B Week 1 Time: two hours 1. Find out personal information of your chat partner. (e.g., Name, major, interests and hobbies, future plans after graduation) 2. Share with your partner any experiences, episodes, opinions and thoughts regarding English learning. (e.g., Why do you study English?; Do you think English is difficult to learn for Koreans? If so, why?) 3. In the given website ( http://plaza.ufl.edu/jbha/week3linguistics.html ), you will find four pictures out of eight. Your chat partner has the other four pictures. Those pictures are in order according to their number. Cooperate with your partner and figure out the whole story. 4. Talk about your achievement, for example, getting your driver’s license, learning a new skill, getting a part-time job. Ask each other some of these questions: What had you been doing before your achievement? How did you prepare for it? Had you considered giving up before you succeeded? 5. Talk about a goal that you are working toward. What steps will you take to achieve your goal? When will you have completed each step? 6. Choose any topics you wish to discuss (if time allows).

PAGE 145

131 Pictures at http://plaza.ufl.edu/jbha/week3linguistics.html Picture 2 Picture 4 Picture 6 Picture 8

PAGE 146

132 Week 2 Time: two hours 1. You are now in online chat conversation. Discuss the differences between face-to-face conversation and online chat conversation. 2. Talk about Inha University. First, talk about what might be done to improve it, and then about what shouldn’t be changed. 3. In the given website ( http://plaza.ufl.edu/jbha/week5yonghyun.html ), there are eight differences between the pictures you and your partner have. Spot as many differences as you can. 4. Talk about a relationship that is important to you. How did you meet? What were you doing when you met? Describe some events in the relationship. 5. Talk about your life ten years from now. What will you be doing for a living? What kind of family life will you have? What hobbies will you be enjoying? What will you do to achieve these things? 6. Choose any topics you wish to discuss (if time allows). Pictures at http://plaza.ufl.edu/jbha/week5yonghyun.html

PAGE 147

133 Week 3 Time: two hours 1. Imagine that you are planning to participate in an international food festival. Which foods would you like to bring and why? 2. Heavy traffic is a big problem in both cities and suburbs. What can we do about it? 3. In the given website ( http://plaza.ufl.edu/jbha/week11hamnida.html ), you will find three pictures out of six. Your chat partner has the other three pictures. Those pictures are in order according to their number. Cooperate with your partner and figure out the whole story. 4. Talk about how you feel in your home, office, or classroom. What makes you feel good? What makes you feel bad? What would you like to change? 5. Talk about a place you remember from your childhood. Where is it? When is the last time you went there? Did you like there? Why or why not? 6. Choose any topics you wish to discuss (if time allows). Pictures at http://plaza.ufl.edu/jbha/week11hamnida.html Picture 2

PAGE 148

134 Picture 4 Picture 6

PAGE 149

135 Week 4 Time: two hours 1. Talk about a situation in your life that you have regrets about. Describe the situation and talk about what you wish had happened and why. 2. Imagine two of you are taking a trip together to another country. You’ll be gone for several weeks. Decide where you’re going. Then make a list of things you have to do and arrange before the trip. 3. In the given website ( http://plaza.ufl.edu/jbha/week9conversation.html ), there are ten differences between the pictures you and your partner have. Spot as many differences as you can. 4. If you hadn’t been born, what would have been different for your family, friends, teammates, or community? Choose two areas of your life to discuss and talk about all the things that would be different. 5. Do you think tipping is a good system? Why or why not? Were you ever in a situation where you didn’t know what to do about a tip? 6. Choose any topics you wish to discuss (if time allows). Pictures at http://plaza.ufl.edu/jbha/week9conversation.html

PAGE 150

APPENDIX F MATERIALS FOR GROUP EC Chat activities For Student A Week 1 Time: one hour 1. Find out personal information of your chat partner. (e.g., Name, major, interests and hobbies, future plans after graduation) 2. Share with your partner any experiences, episodes, opinions and thoughts regarding English learning. (e.g., Why do you study English?; Do you think English is difficult to learn for Koreans? If so, why?) 3. In the given website ( http://plaza.ufl.edu/jbha/week3language.html ), you will find four pictures out of eight. Your chat partner has the other four pictures. Those pictures are in order according to their number. Cooperate with your partner and figure out the whole story. 4. Choose any topics you wish to discuss (if time allows). Note: Pictures in the given websites are presented in Appendix D. Week 2 Time: one hour 1. You are now in online chat conversation. Discuss the differences between face-to-face conversation and online chat conversation. 2. Talk about Inha University. First, talk about what might be done to improve it, and then about what shouldn’t be changed. 3. In the given website ( http://plaza.ufl.edu/jbha/week5inha.html ), there are eight differences between the pictures you and your partner have. Spot as many differences as you can. 4. Choose any topics you wish to discuss (if time allows). 136

PAGE 151

137 Week 3 Time: one hour 1. Imagine that you are planning to participate in an international food festival. Which foods would you like to bring and why? 2. Heavy traffic is a big problem in both cities and suburbs. What can we do about it? 3. In the given website ( http://plaza.ufl.edu/jbha/week11kamsa.html ), you will find three pictures out of six. Your chat partner has the other three pictures. Those pictures are in order according to their number. Cooperate with your partner and figure out the whole story. 4. Choose any topics you wish to discuss (if time allows). Week 4 Time: one hour 1. Talk about a situation in your life that you have regrets about. Describe the situation and talk about what you wish had happened and why. 2. Imagine two of you are taking a trip together to another country. You’ll be gone for several weeks. Decide where you’re going. Then make a list of things you have to do and arrange before the trip. 3. In the given website ( http://plaza.ufl.edu/jbha/week9advanced.html ), there are ten differences between the pictures you and your partner have. Spot as many differences as you can. 4. Choose any topics you wish to discuss (if time allows).

PAGE 152

138 For Student B Week 1 Time: one hour 1. Find out personal information of your chat partner. (e.g., Name, major, interests and hobbies, future plans after graduation) 2. Share with your partner any experiences, episodes, opinions and thoughts regarding English learning. (e.g., Why do you study English?; Do you think English is difficult to learn for Koreans? If so, why?) 3. In the given website ( http://plaza.ufl.edu/jbha/week3linguistics.html ), you will find four pictures out of eight. Your chat partner has the other four pictures. Those pictures are in order according to their number. Cooperate with your partner and figure out the whole story. 4. Choose any topics you wish to discuss (if time allows). Week 2 Time: one hour 1. You are now in online chat conversation. Discuss the differences between face-to-face conversation and online chat conversation. 2. Talk about Inha University. First, talk about what might be done to improve it, and then about what shouldn’t be changed. 3. In the given website ( http://plaza.ufl.edu/jbha/week5yonghyun.html ), there are eight differences between the pictures you and your partner have. Spot as many differences as you can. 4. Choose any topics you wish to discuss (if time allows).

PAGE 153

139 Week 3 Time: one hour 1. Imagine that you are planning to participate in an international food festival. Which foods would you like to bring and why? 2. Heavy traffic is a big problem in both cities and suburbs. What can we do about it? 3. In the given website ( http://plaza.ufl.edu/jbha/week11hamnida.html ), you will find three pictures out of six. Your chat partner has the other three pictures. Those pictures are in order according to their number. Cooperate with your partner and figure out the whole story. 4. Choose any topics you wish to discuss (if time allows). Week 4 Time: one hour 1. Talk about a situation in your life that you have regrets about. Describe the situation and talk about what you wish had happened and why. 2. Imagine two of you are taking a trip together to another country. You’ll be gone for several weeks. Decide where you’re going. Then make a list of things you have to do and arrange before the trip. 3. In the given website ( http://plaza.ufl.edu/jbha/week9conversation.html ), there are ten differences between the pictures you and your partner have. Spot as many differences as you can. 4. Choose any topics you wish to discuss (if time allows).

PAGE 154

140 Input enhancement activities Week 1 The Frog Prince Week 2 Rapunzel Week 3 The Tale of the First Sheik Week 4 The Tale of the Second Sheik Note: The contents of input enhancement activities are presented in Appendix D.

PAGE 155

APPENDIX G SAMPLE CHAT TRANSCRIPTS Following are their sample chat transcripts for information gap and discussion tasks. To retain the authenticity of chat conversation, no revisions have been made to their transcripts other than their ID. Information gap and Jigsaw tasks Student 1: ok. let's make whole story. Student 1: My picture1 is.. Student 1: A lady is in a grocery store. Student 1: she just entered. Student 1: she is carring a cart. Student 1: not sure if anything in it. Student 1: ur picture2 is? Student 2: um Student 2: that lady met someone Student 2: carrying a kid Student 2: in the cart Student 2: . Student 1: wait. Student 1: u seeing picture2? Student 2: yeah Student 1: ok. Student 1: cuz my pic3 is same as you just said. Student 1: ok keep going. Student 2: don't know what it;s about Student 2: -.Student 1: picture5 is Student 2: i have picture 4 Student 1: ok Student 1: tell me Student 2: they start talking Student 2: and the kid picks upa bottol Student 2: that's it Student 1: from the shelve? 141

PAGE 156

142 Student 2: yes Student 1: ok Student 1: now the baby put the bottle in the first lady's back. Student 1: two ladies are still talking without knowing what the baby is doing. Student 2: in the bag you mean? Student 1: oh. yes. Student 1: bag. Student 1: keke (laughing) Student 2: ok Student 2: and Student 2: in pic 6 Student 2: the cashier is doinig the thing Student 1: the thing is ? Student 2: caculating? Student 1: oh. Student 1: ok Student 2: watchacall it? Student 1: dunno. keke Student 1: But we know it. Student 2: ok Student 1: in pic7, Student 1: a man is talking to her. Student 1: he is grabbing her in the shoulder. Student 1: seems like arresting her for theft. Student 1: And she seems confused. Student 2: in 8 Student 2: policeman are checking her Student 1: the same guy in pic 7? or a policeman in uniform? Student 2: both Student 1: so there are two men in 8? Student 2: with the woman sitting on a chair Student 1: oh, maybe my man is the cashier. Student 1: did the cashier wear glass? Student 1: glasses* Student 2: no the cashier was a woman Student 1: ok. Student 2: the man is the person who was maiting in line behind her Student 1: maiting? Student 2: waiting Student 1: ok. Student 2: k Student 1: hmm. Student 1: who is the man... Student 1: is he just customer? Student 2: yes Student 2: I think he's the one who saw the bottol in her bag

PAGE 157

143 Student 1: what is he doing in my pic 7, then. Student 1: oh. Student 2: do you have to answer questions? Student 1: wait I will check. Student 1: well Mr. Ha wanted us to make whole story. Student 2: so the story is about the baby makin ghis mom a thief afterall Student 1: mom's friend? Student 2: by mistake Student 1: yeh. Student 2: oh yes her friend Student 1: that sounds right./ Student 2: good Student 1: and the man snatching her. Student 2: yeah Student 1: good. Discussion tasks Student 1: you got 3 classes of Midterm left? Student 2: yes Student 2: two of them on tuesday Student 2: so should we go on to #5? Student 1: ok. Student 1: I dont understand no.5. Student 1: Do you? Student 2: a goal you are working toward Student 1: like what I'm aiming at for career? Student 2: you must have a goal in your life Student 2: and what are doing to achive it step by step Student 1: what' yours? Student 2: like for me Student 2: i want to live a happy life with the people i live Student 2: in a luxurious setting Student 2: so Student 2: in oder to make that possible kk Student 1: luxurious setting will be like big apartment? Student 2: I must get a nice career Student 1: and nice car, Student 1: ? Student 2: and to have a good job Student 2: i'm trying to be good school Student 2: i guess that's all i can do for now Student 1: be good school? Student 2: good in school Student 1: you mean you're trying to get good grade

PAGE 158

144 Student 1: ok. Student 2: and i try to see many things Student 2: other people can't see Student 1: like what? Student 1: on earth? Student 2: try to lot of experiences and stuff Student 1: are there anytihin left? Student 2: that's about it for me Student 1: ok. Student 2: your turn keke Student 1: I will leave a long life, keke. I will survive all the people I know. Student 1: heathily. Student 2: how long do you want to live Student 1: about 80? Student 1: dont know. Student 2: isn't that like the average these days? Student 1: yeh. Student 2: what do you mean you want to survive all the people you know Student 1: live longer that them. Student 1: than Student 2: you wanna live longer than them? Student 1: yes. Student 2: oh Student 1: now, I have to live longer than you, keke. Student 2: kk Student 2: but what for? Student 1: to learn things. Student 1: like about universe Student 2: oh kk Student 1: or the one principle on which this world is made Student 1: keke. Student 2: Student 2: all right Student 1: It's funny. Student 2: what is? Student 1: to type 'all righ' in Korean key makes Student 1: ' Student 1: keke Student 2: kk Student 2: so next one? Student 1: is any topic. Student 1: I guess this is good enough for 1week. Student 1: plus the last one. Student 2: yeah Student 1: wow, still it tood 52 min. Student 1: I doubt this experiment will work..

PAGE 159

145 Student 2: Student 2: why Student 1: cuz, as you and I are doing, we are omminting alomst "the, a, an, some" things. Student 2: huh?? Student 1: His objective is Student 1: to observe the students' using the "", Student 2: Student 1: but we are not trying hard to use it. In fact we are all ignoring it, right? Student 2: poh Student 2: oh Student 1: so. Student 2: ? Student 2: are we done for today? Student 1: yes. Student 1: I know you got a lot to do. Student 2: time flies Student 1: wish you good luck with exams. Student 2: flys Student 2: thanks Student 1: and I will message you Student 2: you save it and send ot to him ok? Student 1: when is you last exam? Student 2: it's in may Student 1: within this week? Student 2: but it's all right after tuesday Student 1: ok. Student 1: then I will message you about Thursday. Student 1: See you, then. Student 2: all right Student 2: bye^^ Student 2: send to him okay? Student 1: ok. Student 1: Bye. Student 2: thank you

PAGE 160

LIST OF REFERENCES Bickerton, D. (1981). Roots of language. Ann Arbor, Karoms. Blake, R. (2000). Computer mediated communication: A window on L2 Spanish interlanguage. Language Learning and Technology, 4, 1, 120-136. Bump, J. (1990). Radical changes in class discussion using networked computers. Computers and the Humanities, 24, 1-2, 49-65. Celce-Murcia, M., & Larsen-Freeman, D. (1999). The grammar book: An ESL/EFL Teacher’s Guide. Boston, Heinle & Heinle. Chapelle, C. (2001). Computer Applications in Second Language Acquisition: Foundations for Teaching, Testing, and Research. New York, Cambridge University Press Danks, J. (1980). Comprehension in listening and reading: Same or different? Reading and Understanding. J. Danks, & Pezdez, K. Newark, DE, International Reading Association, 1-39. Doughty, C., & Williams, J. (1998). Pedagogical choices in focus on form. Focus on Form in Classroom Second Language Acquisition. C. Doughty, & Williams, J. New York, Cambridge University Press, 197-261. Erlam, R. (2003). Evaluating the relative effectiveness of structured-input and output-based instruction in foreign language learning: Results from an experimental study. Studies in Second Language Acquisition, 25, 4, 559-582. Greenslade, T., Bouden, L., & Sanz, C. (1999). Attention to form and content in processing L2 reading texts. Spanish Applied Linguistics, 3,65-90. Hakuta, K. (1976). A case study of a Japanese child learning English as a second language. Language Learning, 26, 321-351. Horibe, S. (2004). The Output Hypothesis and Cognitive Processing: An Examination via Acquisition of Japanese Temporal Subordinate Conjunctions, Unpublished doctoral dissertation, Purdue University. Huebner, T. (1983). A Longitudinal Analysis of the Acquisition of English. Ann Arbor, MI., Karoma. 146

PAGE 161

147 Huebner, T. (1985). System and variability in interlanguage syntax. Language Learning, 35, 141-163. Izumi, S. (2002). Output, input enhancement, and the noticing hypothesis: an experimental study on ESL relativization. Studies in Second Language Acquisition, 24, 547-577. Jespersen, O. (1933). Essentials of English Grammar. London, Allen and Unwin. Johnson, J. S. (1992). Critical period effects in second language acquisition: The effect of written versus auditory materials on the assessment of grammatical competence. Language Learning, 42, 217-248. Johnson, J. S., & Newport, E. L. (1989). Critical period effects in second language learning: The influence of maturational state on the acquisition of English as a second language. Cognitive Psychology, 21, 60-99. Kahneman, D. (1973). Attention and effort. Englewood Cliffs, NJ, Prentice Hall. Kern, R. (1995). Restructuring classroom interaction with networked computers: Effects on quantity and characteristics of language production. Modern Language Journal, 79, 457-476. Kitade, K. (2000). L2 learners’ discourse and SLA theories in CMC: Collaborative interaction in Internet chat. Computer Assisted Language Learning, 13, 2, 143-166. Koyanagi, K. (1998). The Effects of Focus-on-Form Tasks on the Acquisition of a Japanese Conditional “to”: Input, Output and “Task-essentialness”. Unpublished doctoral dissertation, Georgetown University. Krashen, S. D. (1994). The input hypothesis and its rivals. Implicit and Explicit Learning of Languages. N. C. Ellis. London, Academic Press, 45-77. Lee, L. (2001). Online interaction: Negotiation of meaning and strategies used among learners of Spanish. ReCALL, 13, 2, 232-244. Leow, R. P. (1993). To simplify or not to simplify: A look at intake. Studies in Second Language Acquisition, 15, 333-355. Leow, R. P. (1995). Modality and intake in second language acquisition. Studies in Second Language Acquisition, 17, 79-89. Leow, R. P. (2001). Do learners notice enhanced forms while interacting with the L2? An online and offline study of the role of written input enhancement in L2 reading. Hispania, 84, 3, 496-509.

PAGE 162

148 Leow, R. P., Egi, T., Nuevo, A. M., & Tsa, Y. (2003). The roles of textual enhancement and type of linguistic item in adult L2 learners’ comprehension and intake. Applied Language Learning, 13, 2, 93-108. Lund, R. (1991). A comparison of second language listening and reading comprehension. Modern Language Journal, 75, 2, 196-204. Matthews, P. M. (1997). The Concise Oxford Dictionary of Linguistics. New York, Oxford University Press. Murphy, V. (1997). The effects of modality on a grammaticality judgment task. Second Language Research, 13, 1, 34-65. Negretti, R. (1999). Web-based activities and SLA: A conversation analysis research approach. Language Learning and Technology, 3, 1, 75-87. Oliva, M., & Pollastrini, Y. (1995). Internet resources and second language acquisition: An evaluation of virtual immersion. Foreign Language Annals, 28, 4, 551-563. Overstreet, M. H. (1998). Text enhancement and content familiarity: The focus of learner attention. Spanish Applied Linguistics, 2, 229-258. Overstreet, M. H. (2003). The Effect of Textual Enhancement on Second Language Learner Reading Comprehension and Form Recognition, Unpublished doctoral dissertation, University of Illinois, Urbana-Champaign. Pellettieri, J. (2000). Negotiation in cyberspace: The role of chatting in the development of grammatical competence. Network-based Language Teaching: Concepts and Practice. M. Warschauer, & Kern, R. Cambridge, UK., Cambridge University Press, 59-86. Peters, A. (2001). The Everything Fairy Tales Book: A Magical Collection of All-Time Favorites to Delight the Whole Family. Adams Media Corporation. Avon, Massachusetts. Pica, T. (1983). The article in American English: what the textbooks don’t tell us. Sociolinguistics and Language Acquisition. N. Wolfson, & Judd, E., Newberry House Publishers, Inc. Pica, T., Holliday, L., Lewis, N., & Morgenthaler, L. (1989). Comprehensible output as an outcome of linguistic demands on the learner. Studies in Second Language Acquisition, 11, 1, 63-90. Pienemann, M. (1984). Psychological constraints on the teachability of languages. Studies in Second Language Acquisition, 6, 186-214. Pienemann, M. (1998). Language Processing and Second Language Development: Processability theory. Amsterdam, Benjamins.

PAGE 163

149 Radford, A. (1997). Syntactic Theory and the Structure of English: A Minimalist Approach. Cambridge, U.K., Cambridge University Press. Robinson, P. (1995). Attention, memory, and the “noticing” hypothesis. Language Learning, 45, 283-331. Rosa, E., & O’Neill, M. (1999). Explicitness, intake and the issue of awareness: Another piece to the puzzle. Studies in Second Language Acquisition, 21, 511-556. Salaberry, M. R. (2000). L2 morphosyntactic development in text-based computer-mediated communication. Computer Assisted Language Learning, 13, 1, 5-27. Schmidt, R. W. (1990). The role of consciousness in second language learning. Applied Linguistics, 11, 129-158. Schmidt, R. W. (1993). Awareness and second language acquisition. Annual Review of Applied Linguistics, 13, 206-226. Schmidt, R. W. (1994). Deconstructing consciousness in search of useful definitions for applied linguistics. AILA Review, 11, 11-26. Schmidt, R. W. (1995). Consciousness and foreign language learning: A tutorial on the role of attention and awareness in learning. Attention and Awareness in Foreign Language Learning. R. W. Schmidt. Honolulu, University of Hawaii: 1-63. Schmidt, R. W. (2001). Attention. Cognition and Second Language Instruction. P. Robinson. Cambridge, Cambridge University, 3-32. Schmidt, R. W., & Frota, S. (1986). Developing basic conversational ability in a second language: A case study of an adult learner of Portuguese. Talking to Learn. R. W. Schmidt. Rowley, MA, Newbury House. Sharwood-Smith, M. (1981). Consciousness-raising and the second language learner. Applied Linguistics, 11, 2, 159-168. Sharwood-Smith, M. (1991). Speaking to many minds: On the relevance of different types of language information for the L2 learner. Second Language Research, 7, 2, 118-132. Sharwood-Smith, M. (1993). Input enhancement in instructed SLA. Studies in Second Language Acquisition, 15, 2, 165-179. Sharwood-Smith, M. (2000). Attentional mechanisms and the language acquisition device: Reflections on communication and developmental processes. Language Processing and Simultaneous Interpreting: Interdisciplinary Perspectives. B. E. Dimitrova, & Hyltenstam, K. Amsterdam, John Benjamins: 25-44.

PAGE 164

150 Simard, D., & Wong, W. (2001). Alertness, orientation, and detection: The conceptualization of attentional functions in SLA. Studies in Second Language Acquisition, 23, 102-124. Simons, G. V. (2001). The Acquisition of the English Determiner System: Sequence, Order, and Transfer, Unpublished doctoral dissertation, Temple University. Skehan, P. (1996). A framework for the implementation of task-based instruction. Applied Linguistics, 17, 1, 38-62. Skehan, P. (1998a). Task-based instruction. Annual review of applied linguistics, 18, 268-286. Skehan, P. (1998b). A Cognitive Approach to Language Learning. Oxford, Oxford University Press. Skehan, P. (2003). Focus on form, tasks, and technology. Computer Assisted Language Learning, 16, 5, 391-411. Smith, F. (1994). Understanding Reading: A Psycholinguistic Analysis of Reading and Learning to Read. Hillsdale, N. J., L. Erlbaum. Swain, M. (1985). Communicative competence: Some roles of comprehensible input and comprehensible output in its development. Input in Second Language Acquisition. S. Gass, & Madden, C. Cambridge, MA., Newbury House, 235-253. Swain, M. (1993). The output hypothesis: Just speaking and writing aren’t enough. The Canadian Modern Language Review, 50, 158-164. Swain, M. (1995). Three functions of output in output in second language learning. Principle and Practice in Applied Linguistics: Studies in Honor of H. G. Widdowson. G. Cook, & Seidlhofer, B. Oxford, Oxford University Press, 125-144. Swain, M. (1998). Focus on form through conscious reflection. Focus on Form in Classroom Second Language Acquisition. C. Doughty, & Williams, J. New York, Cambridge University Press, 64-81. Swain, M., & Lapkin, S. (1995). Problems in output and the cognitive processes they generate: A step towards second language learning. Applied Linguistics, 16, 371-391. Thomas, M. (1989). The acquisition of English articles by firstand second-language learners. Applied Psycholinguistics, 10, 335-355. Tomlin, R., & Villa, V. (1994). Attention in cognitive science and second language acquisition. Studies in Second Language Acquisition, 16, 183-204.

PAGE 165

151 Trademan, J. E. (2002). The Acquisition of the English Article System by Native Speakers of Spanish and Japanese: A Cross-Linguistic Comparison, Unpublished doctoral dissertation, University of New Mexico. VanPatten, B. (1990). Attending to form and the content in the input: An experiment in consciousness. Studies in Second Language Acquisition, 12, 287-301. VanPatten, B. (1996). Input Processing and grammar instruction: Theory and research. Norwood, NJ, Ablex. Warschauer, M. (1998). Interaction, negotiation, and computer-mediated learning. Practical Application of Educational Technology in Language Learning. M. Clay. Lyon, France, National Institute of Applied Sciences. Retrieved January 15, 2001: from http://www.insa0lyon.fr/Departments/CDRL/interaction.html . Werry, C. C. (1996). Linguistic and interactional features of Internet relay chat. Computer-Mediated Communication: Linguistic, Social and Cross-Cultural Perspectives. S. C. Herring. Amsterdam, John Benjamins, 47-63. Wickens, C. (1980). The structure of attentional resources. Attention and Performance, VIII. R. S. Nicherson. Hillsdale, NJ, Erbaum, 239-257. Wickens, C. (1984). Processing resources in attention. Varieties of Attention. R. Parasuraman, & Davies, D. Orlando, FL, Academic Press, 63-98. Wickens, C. (1989). Attention and skilled performance. Human Skills. D. Holding. New York, John Wiley, 71-105. Wong, W. (2001). Modality and attention to meaning and form in the input. Studies in Second Language Acquisition, 23, 3, 345-368. Yule, G. (1997). Referential Communication Tasks. Mahwah, N.J.: Erlbaum.

PAGE 166

BIOGRAPHICAL SKETCH Jongbum Ha was born September 30, 1967 in Seoul, South Korea. After graduating from Dongsan High School in 1986, he began his undergraduate studies at Inha University in the department of English and Literature that same year. With his military service fulfilled from 1988 to 1989, he graduated from Inha University with a Bachelor of Arts in 1992, when he began his graduate studies at Inha University majoring in English linguistics. He completed his Master of Arts degree in 1994, when he began to teach English conversation in the Foreign Language Educational Center at Inha University until 1995. After 4 years of nonacademic work experience at Aekyung PetroChemical Corporaton. Ltd., he resumed his study at University of Florida in 2000 to begin his doctoral work in linguistics with a specialization in Second Language Acquisition. During his study at University of Florida, he earned a certificate in Teaching English as a Second Language (TESL) in 2001, and taught English reading and grammar at Inha University in 2004 while collecting data for his dissertation. 152