1 ADOLESCENT S USE OF ACADEMIC LANGUA G E IN HISTORICAL WRITING By TING SHEN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2011
2 2011 Ting Shen
3 To my parents, Emma, and sunshine State
4 ACKNOWLEDGMENTS I am grateful to Dr. Zhihui Fang, my committee chair, who advised me with his professional knowledge and life experience. He illuminated me in the dissertation research and guided me through the whole process. This dissertation could not be completed without his instruction. I am grateful to my committee members, Dr. Ho lly Lane, Dr. Ruth Lowery, and Dr. Sevan Terzian. All these professors taught me with professionalism and warmest support. Their wisdom in teaching and research and their love and care to students are the lifetime treasure that I found at UF. I must also e xpress my gratitude to Dr. Heribeto Godina who is both my friend and mentor. His professional advices were extremely meaningful in the writing of this dissertation. I must express my gratitude to the world history teachers at Buchholz High School and the Vice Principal, Mr. Shelnutt for their generous help. They opened their classrooms for me and offered all the assistance that I needed. This research could not be conducted without their support. I thank University of Florida for offering me this opportuni ty to study and supporting me with fellowship and assistantships. I will never forget the five years I spent in the beautiful campus at UF. I want to say thank you to my life long friend Jiahang. He listened to my complaining, he tolerated my temper, and he comforted my frustration with all the patience. He supported me from the beginning of my doctoral program through the defense and final revision, which will be remembered forever. I am grateful to all my friends and especially PC. They made my life in Florida exceptional and unforgettable. Those days and nights working on my dissertation were never difficult. They accompanied me every moment in this long journey. And your love
5 and friendship help me to grow. I am grateful to my friend who has changed e very dismal moments into beautiful memories. I also thank my daughter, Emma, for being the best child in the world. She had to play by herself because her mom was busy writing the dissertation. But she seldom complained. She made me the happiest mom by bei ng a healthy, beautiful, and happy girl. I am working hard to grow up and be a mom she will be proud of. Last but not least, I thank my parents, Houmei Yang and Zhengxin Shen, for diligently and patiently raising me to be a healthy and inte lligent human b eing They have revealed the power of reading and writing and been teaching me the meaning of education. They have spent all their time and energy in helping me to grow in all aspects of my life. They are always there whenever I need them. I dedicate my di ssertation to those who have loved me with all their heart You love is all the reason why I am finally here.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 9 GLOSSARY ................................ ................................ ................................ .................. 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 14 Background ................................ ................................ ................................ ............. 14 Conceptualization of Academic Language ................................ .............................. 16 Statement of the Problem ................................ ................................ ....................... 19 Research Questions ................................ ................................ ............................... 25 Significance of the Study ................................ ................................ ........................ 25 Li mitations and Delimitations ................................ ................................ .................. 28 Overview of the Study ................................ ................................ ............................. 29 2 LITERATURE REVIEW ................................ ................................ .......................... 31 Conceptualizing Academic Language ................................ ................................ ..... 31 Cognitive Perspective on Academic Language ................................ ................ 31 Sociocultural Perspective on Academic Language ................................ ........... 36 Functional Linguistics Perspective on Academic Language ............................. 44 Academic language: register of the school context ................................ .... 46 Experiential meaning ................................ ................................ ................. 47 Interpersonal meaning ................................ ................................ ............... 48 Textual meaning ................................ ................................ ........................ 48 Grammatical Features of Academic Language ................................ ................ 49 Academic vocabulary ................................ ................................ ................. 50 Expanded noun phrase ................................ ................................ .............. 51 Nominalization ................................ ................................ ........................... 52 Embedded clause ................................ ................................ ...................... 54 Lexical dens ity ................................ ................................ ........................... 56 Academic language: different content areas and different genres ............. 58 Developing Academic Language Across and Within Content Areas ....................... 61 Overview of the Academic Language Instruction ................................ ............. 61 Academic Language Features of Content Areas ................................ .............. 62 Writing in Academic Language ................................ ................................ ......... 67 Instructional Practices to Develop Academic Language ................................ ... 69 Academic vocabulary ................................ ................................ ................. 70 Other li nguistic features ................................ ................................ ............. 72 Training Teachers to Teach Academic Language Features ............................. 77
7 Teachers' knowledge and practice of academic language ......................... 77 Professional development ................................ ................................ .......... 81 Deve loping Academic Language in History Classrooms ................................ .. 84 Language demands in the discourse of history ................................ .......... 84 Developing academic language in history ................................ .................. 87 Gaps in the Literature ................................ ................................ ............................. 90 3 METH ODOLOGY ................................ ................................ ................................ ... 99 R esearch S ite ................................ ................................ ................................ ......... 99 Data Collection ................................ ................................ ................................ ..... 100 Background ................................ ................................ ................................ .... 100 Procedures ................................ ................................ ................................ ..... 104 Data Analysis : Variables ................................ ................................ ....................... 107 Reading Abilities ................................ ................................ ............................. 107 Writing Quality ................................ ................................ ................................ 109 Raters ................................ ................................ ................................ ...... 109 Rubric ................................ ................................ ................................ ...... 110 Training and reliability ................................ ................................ .............. 112 Wr iting groups ................................ ................................ .......................... 11 3 Academic Language Features ................................ ................................ ........ 114 Inter rater Reliability ................................ ................................ ....................... 117 Statistical Analysis ................................ ................................ ................................ 117 4 FIND ING ................................ ................................ ................................ ............... 121 Research Question I: Descriptive Statistics ................................ .......................... 121 Academic Vocabulary ................................ ................................ ..................... 121 Embedded Clause ................................ ................................ .......................... 122 Expanded Noun Phrase ................................ ................................ ................. 125 Nomina lization ................................ ................................ ................................ 127 Lexical Density ................................ ................................ ............................... 131 Academic Language Features in the Text ................................ ...................... 131 Academic vocabulary ................................ ................................ ............... 131 Expanded noun phrase ................................ ................................ ............ 135 Embedded clause ................................ ................................ .................... 137 Nominalization ................................ ................................ ......................... 138 Lexical dens ity ................................ ................................ ......................... 138 Research Question II: MANOVA Results ................................ .............................. 139 Correlation ................................ ................................ ................................ ...... 140 MANOVA ................................ ................................ ................................ ........ 140 Research Question III: Regression ................................ ................................ ....... 144 ANOVA ................................ ................................ ................................ ........... 144 Multiple Regression ................................ ................................ ........................ 145 5 DISCUSSION AND CONCLUSION ................................ ................................ ...... 158
8 Discussion of Findings ................................ ................................ .......................... 158 Adol escents' Use of Academic Language ................................ ...................... 158 Academic vocabulary ................................ ................................ ............... 158 Expanded noun phrase ................................ ................................ ............ 159 Embedded clause ................................ ................................ .................... 160 Nominalizati on ................................ ................................ ......................... 162 Lexical density ................................ ................................ ......................... 164 Academic Language Features and Reading Abilities ................................ ..... 166 Academic Language Features and Writing Qualiti es ................................ ...... 169 Issues of Interest ................................ ................................ ............................ 176 Limitations and Implications ................................ ................................ .................. 176 Limitations ................................ ................................ ................................ ...... 176 Implications for Practice ................................ ................................ ................. 178 Bu ilding awareness ................................ ................................ .................. 178 Features that must be emphasized ................................ .......................... 179 Suggestions for preparing teachers to teach academic language ............ 181 Implications for Research ................................ ................................ ............... 182 APPENDIX A BENCHMARK ESSAYS ................................ ................................ ........................ 188 B ACADEMIC VOCABULARY LIST (Coxhead, 2000) ................................ ............. 195 LIST OF REFERENCES ................................ ................................ ............................. 217 BIOGRAPH ICAL SKETCH ................................ ................................ .......................... 233
9 LIST OF TABLES Table page 2 1 Academic t ext and w ritten t ext ................................ ................................ ............ 96 2 2 Academic t exts and i nteractional t ext in clause l evel ................................ .......... 96 2 3 Clause types ................................ ................................ ................................ ....... 97 2 4 Register f eatures of a cademic l anguage ................................ ............................ 97 2 5 Genres of s chooling and t heir l anguage f eatures ................................ ............... 98 3 1 Features of students ................................ ................................ ......................... 120 3 2 Writing groups ................................ ................................ ................................ .. 120 4 1 Descriptive statistics ................................ ................................ ......................... 149 4 2 Frequent word and their occurrence frequency ................................ ................ 149 4 3 Expanded noun phrase analysis ................................ ................................ ....... 150 4 4 Embedded clause analysis ................................ ................................ ............... 151 4 5 Nominalization analysis ................................ ................................ .................... 152 4 6 Correlation analysis ................................ ................................ .......................... 152 4 7 Box's t est of e quality of c ovariance ................................ ................................ ... 153 4 8 Multivariate t ests ................................ ................................ ............................... 153 4 9 Tests of b etween s ubjects e ffects ................................ ................................ .... 154 4 10 Levene's t est of e quality of e rror v ariances ................................ ...................... 154 4 11 Multiple c omparisons ( Scheffe ) ................................ ................................ ........ 155 4 12 Compare means of five academic language features ................................ ....... 155 4 13 ANOVA ................................ ................................ ................................ ............. 156 4 14 Variables entered/removed ................................ ................................ .............. 156 4 15 Model summary ................................ ................................ ................................ 157
10 GLOSSARY Academic l anguage Simply speaking, academic language is the language used in academic settings and for academic purposes. It is used interchangeably with academic literacy and academic English in this study. In this study, adopting Systemic Functional Linguistics, academic language is defined as a constellation of language resources that realize more formal and abstract styles and construe academic register in different disciplinary communities. Academic language tends to be more technical, abstract, information laden, and h ierarchically organized than everyday language. Academic l iteracy Academic literacy refers to the knowledge and skills (e.g., speaking, listening, reading, and writing) students need in order to be successful in an academic setting. Clause A clause is a unit of grammatical analysis on a rank scale that may include other elements ranking below it, such as a noun group, a verb group and a prepositional phrase (Eggins, 2004). Four types of clauses are discussed in this study: main clause, hypotactic clause paratactic clause, and embedded clause. Embedded c lause Embedded clause functions as a nominal element or part of a nominal element within another clause, sometimes as a postmodifier in a noun phrase. It includes restrictive relative clauses, comparative clauses, and nominalized clauses that function as subject or complement. English language learner (ELL) An English language learner is a person whose first, home, or dominant language is a language other than English and who is in the proces s of learning English. ELL can also be used as an adjective to describe a student, e.g. ELL student. Genre Genre describes particular types of texts that serve specific social purposes and characterize particular social contexts (Schleppegrell, 2004). Each genre can be identified by its communicative purpose and its linguistic features. Grammatical m etaphor Grammatical metaphor is a pervasive feature of the academic language that expresses meanings in incongruent ways F or example, when causality is not construed through conjunctions (e.g., because) but through verbs (e.g., lead to ) or nouns (e.g., cause, rea son), these verbs and nouns are considered grammatical metaphors.
11 Lexical Density Lexical Density is defined as a measure of the number of content words per non embedded clause in a text. It reveals how tightly the lexical items have been packed in a tex t and how difficult a text is to read. Nominalization Nominalization is the process by which verbs, adjectives, or what would more naturally be presented in another form in interactional language become noun phrases that can be the subject of other verb s or prepositions (Eggins, 2004; Schleppgegrell, 2004). It is also an instance of grammatical metaphor. Non finite v erb A non finite verb (or a verbal) is a verb form that is not limited by a subject a nd, more generally, is not fully inflected by categories that are marked inflectionally in language, such as tense, number, and person. An example is: To succeed takes courage, foresight, and luck (Here to succeed is a non finite verb that serves the su bject of the clause). Register Register is defined as "the configuration of lexical and grammatical resources which realizes a particular set of meanings (Schleppgrell, 2004, pp. 45 46). Three variables determine a register: field (what is talked about, the subject matter of the discourse), tenor (participants in communication and their social relationships), and mode (expectations of how particular types should be organized, the format of language, such as the differences between speech and writing) (Hal liday & Matthiessen, 2004; Schleppgrell, 2004).
12 A bstract 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 ADOLESCENT S USE OF ACADEMIC LA NGUA G E IN HISTORICAL WRITING By Ting Shen December 2011 Chair: Zhihui Fang Major: Curriculum and Instruction Despite its importance of academic language research on academic language is often limited to academic vocabulary and focused on the English language learners. Informed by systemic functional linguistics, this study examined adolescents' use of academic language and the relationships between its u se and students' reading ability and their writing quality Eighty four ninth grade students from a public high school in a southeastern state of the United States participated in the study. Three types of analysis were conducted on students' expository es says. First, the essays were rated holistically on a scale of 1 5, with 5 meaning the highest quality. Second, each essay was analyzed for evidence of the following five academic language features: academic vocabulary, expanded noun phrase, nominalization, embedded clause, and lexical density. Third, statistical analyses were conducted to examine the relationship between the use of these academic language features and students' reading ability and writing quality. Linguistic analysis showed that the student s used all five academic features in their essays with varying degrees of proficiency. Whereas features such as expanded noun phrase, academic vocabulary, and embedded clause were used with regularity in
13 the students' writing, nominalization was used much less frequently and lexical density was low. Multivariate analysis of variance (MANOVA) indicated a significant relationship between the students' use of academic language and their reading ability, defined by their scores on a state reading achievement te st. Compared to students of lower reading abilities, students of higher reading abilities used significantly more nominalizations and showed higher lexical density in their writing. However, no significant differences were found between the reading groups in the use of academic vocabulary, embedded clause, and expanded noun phrase. Analysis of variance (ANOVA) showed that higher quality essays scored significantly higher than lower quality essays on expanded noun phrase, nominalization, and lexical density, but not on academic vocabulary and embedded clause. Backward stepwise regression analysis revealed that nominalization predicted 43% of the variances in writing quality and that academic vocabulary had a negative relationship with writing quality. These f indings suggested that the relationships between academic language use and reading ability and writing quality are complex and warrant further exploration.
14 CHAPTER 1 INTRODUCTION In educational research and practice, academic language refers to the language and literacy practices that are specific to the schooling context (e.g. Scarcella, 2003; Bailey & Butler, 2003). Mastering academic language means knowing how to use general and discipline specific vocabulary, specialized grammatical patterns, diverse language functions and discourse structures in different subject areas in order to comprehend and compose academic texts, acquire new knowledge and skills, and interact with various audiences about a topic. Successful participation in disciplinary learning depends on successful command of the academic language in the discipline (Schleppegrell, 2004). Academic language is important but challenging for adolescents At the high school level students move from a social register to a more complex arena of academic learning Students leave the sphere of learning material s that closely resemble their everyday lives and enter a more specialized type of learning. Adolescents encounter a gr eater range of disciplinary texts and subject matter that must be decoded and assimilated in order to acquire and display their disciplinary knowledge. The challenges that academic language present s to adolescents must be overcome in order for them to acqu ire the knowledge and skills being taught in all disciplines My study examines high school students' use of academic language in historical writing and the relationship between such use and students' reading ability and writing quality Background A cademic language is the language used in academic settings for purposes of teaching and learning. In a society where the need for literacy is growing, learning
15 academic language is the most reliable way of attaining school academic achievement, career adva ncement, and socio economic successes in the United States (Scarcella, 2003). Academic language has also been regarded as one of the key factors affecting the achievement gap that exists between high and low performing groups of students (Zwiers, 2007). I t poses great challenges to students, especially language and culture minority students, when they advance into secondary schools (e.g., Cummins, 1981; Scarcella, 2003; Schle ppegrell, 2003 ). Academic language plays a cr itical role in secondary schools due to the increasing language and literacy demands of secondary content area learning. When students advance from elementary to secondary school, the knowledge that they are expected to learn in content areas is removed from students' personal lives as well as everyday contexts and becomes more and more specialized. The language that constructs such knowledge also becomes more technical, abstract, information laden, and hierarchically organized (Fang & Schleppegrell, 2008). The lexico grammatical features and discourse organizational patterns of content area texts differ significantly from the language patterns in interactional language that students typically encounter during the elementary years or in everyday life. For many students who have little experien ce with academic language outside school, this challenge is more conspicuous and must be addressed. The importance of academic language to secondary students has no t been recognized and explored to a full extent (Scarcella, 2003). Despite the crucial role of academic language in secondary schooling, the existing body of research focuses on college learners rather than students in elementary or secondary school (Scarcella,
16 2003). A number of descriptiv e studies and linguistic analyse s study lexico gram matical features and teachers' expectation on students' writing by analyzing content area reading materials and classroom discourse (e.g., Bailey & Butler, 2007; Bunch, 2006; Schleppegrell, 2001,2004 ; Zwiers, 2007). But l ittle is known about mainstream stu dents' ability (beyond English Language Learners) to use academic language, such as how they cope with academic language features in academic writing and whether their academic language proficiency meets expectations in academic disciplines. Even less is k nown about academic language use in content areas such as history. Therefore, my study fills this gap by investigating how adolescents use academic language in their historical writing and whether such use is related to their reading ability and writing q uality. This will lead to an in depth understanding of students' abilities in coping with academic language features in the context of secondary history classrooms. Conceptualization of Academic Language Academic language is a complex concept that has been operationalized from a variety of perspectives and for a variety of purposes. Among them, the cognitive, linguistic, and sociocultural perspectives are the most important groundwork for conceptualizing a cademic language. Generally speaking, academic language is the language used in academic context to help students acquire and display knowledge (Chamot & O'Malley, 1994). Cummins (1981, 2000 a ), drawing on research with bilingual children, attempts to dis tinguish academic language as Cognitive Academic Language Proficiency (CALP), from everyday language as Basic Interpersonal Communicative Skills (BICS). By interpreting these two concepts into two intersecting continua, Cummins (1981, 2000a) views academic language as the construction of meaning that relies on mostly linguistic
17 resources and minimum contextual support. He defines academic language as "the extent to which an individual has access to and command of the oral and written academic registers of s chooling" (2000a, p. 67). Cummins also lists particular linguistic features such as complex sentences and precise vocabulary as characteristics of academic language. The theory of CALP and BICS lays the groundwork for a comprehensive and applicable definit ion of academic language. The conceptualization of academic language also draws on the sociocultural tradition of literacy that emphasizes differences in students' home language experience and school language. Gee (1996 ), who holds a sociocultural orienta tion, believes that Discourse is a combination of language and relevant social practices in a specific group. He claims that language and literacy involves social practices that go far beyond linguistic and cognitive processing. Academic language is regard ed here as a type of social practice or Discourse where social interaction plays an indispensible role (Gee, 1996). By differentiating primary Discourse as literacy learned at home and secondary Discourse as literacy learned in school and institutions, Gee (1996) regards academic language as a kind of secondary Discourse that is socially dominant, entails values and beliefs, and excludes others who have not acquired it. Acquisition of academic language thus relies on both acq uisition of cognitive abilities, linguistic codes and participation in specific ways of being and acting in academic context. The third perspective is a linguistic perspective, in which the most productive line of research draws on Systemic Functional Linguistics (SFL). SFL is a lingui stic theory that relates language to context (Halliday & Matthiessen, 2004; Schleppegrell, 2004). In SFL, language is viewed as a resource to make meaning. Linguistic resources are
18 exploited to express meaning s and serve functions based on the social conte xt By focusing on meaning, the symbiotic relationship between social context and language can be explained. In SFL, academic language is conceptualized as "a set of linguistic registers that construe multiple and complex meanings at all levels and in all subjects of schooling" (Schleppegrell, 2009). From this perspective, a cademic language proficiency is the ability to exploit a wide range of grammatical and lexical features, semantic, inter clausal, or intra clausal, to construct meaning and realize langu age functions such as constructing and communicating certain kind of knowledge, and describing complex ideas, higher order thinking processes, and abstract concepts (Chamot & O'Malley, 1994). SFL provides the field with a toolbox to analyze grammatical and discourse features of academic language that may guide academic language conceptualization, instructional methods, and test development. A cademic language compris es a set of linguistic registers to construe multiple levels of social practices and differe nt disciplines T he range and variety of social practices within academic community leads to different types of academic language such as Essential Academic Language (the basic features of academic language that are used across all content areas) and disci pline specific language such as discourse of history. Different types of academic language encompass different knowledge skills and linguistic features. Proficiency in a cademic language also relies on multiple and diverse competencies, for example, written or oral. Whereas students are engaged in academic reading and writing in school on a daily base academic language is used in oral situations such as academic discussion about global warming in a science classroom.
19 The cognitive and sociocultural perspec tives are crucial for academic language, and the research in these two dimensions receives more attention (e.g., Scarcella, 2003). On the other hand the important role of linguistic resources in construing academic language is not fully recognized and the current research from the linguistic perspective of academic language is relatively limited. Therefore my study adopt s the theoretical model of SFL and pays full attention to those linguistic features through which academic learning can be expressed. Statement of the Problem Academic language plays a crucial role in school success, as it is the foundation of academic learning. Without the proficiency and mastery in academic language, students are very likely to experience difficulty in both understandi ng content materials and demonstrating content area knowledge (August & Hakuta, 1997). Academic language has been increasingly cited as a reason for achievement gaps between ELLs and English proficient students and between high and low performing groups o f students (Zwiers, 2007). Therefore, understanding students' strengths and needs in academic language is the foremost step in designing subsequent instruction and/or remediation. The growing literature on academic language has been scattered across a nu mber of different areas, ranging from linguistic analysis of written and spoken texts to descriptive studies of classroom practices. In this body of research, there are a number of gaps that must be addressed. First of all, current academic language resea rch focuses on English language learners (ELLs) instead of the entire population A cademic language presents a challenge to all students, not just ELLs. There is strong evidence for the claim that
20 adolescents (both mainstream students and ELLs) are not abl e to read and write the specialized texts of s econdary schooling (e.g., Berman & Biancarosa, 2005; F illmore & Snow, 2000) Across the whole country, three quarters of 4th, 8th, and 12th graders performed below the proficiency level in a National Assessment of Educational Progress (NAEP) (Persky, Daane & Jin, 2003). The 2002 NAEP report shows that American students are not well prepared for academic writing Less than a third of stu dents in Grade 4 (28%), Grade 8 (31%), and Grade 12 (21%) scored at or above p roficient levels, and for all three writing samples collected only 2% wrote at advanced levels for all of them These are striking evidence s of students' limited academic language proficiency s Academic language is cha llenging to all students for a number of reasons. ELLs' academic language ability is significantly impeded not only by their limited foundation in conversational English but also by their lack of knowledge in specialized registers that tend to manifest the dominant cultures of those inside academia (Zwier, 2008). For many mainstream students, like ELLs, academic language is a second language because many students grow up in discourses that do not p repare them for specialized practices of academic reading, writing, and speaking. Fluency in conversational English, even when English is a first language, does not necessarily translate to a mastery of academic language. To develop proficiency in academic language, the process of teaching and learning must be involved (Gee, 1996). It also relies on sufficient exposure to academic texts (written or oral), extensive reading and writing experience in a variety of academic texts, years of immersion in academic vocabulary and linguistic features that characterize academic language, and opportunities to critically use academic
21 language in lecture, discussion, or cooperative learning. The assumption that every student with English as the first language can acquire academic language automatically is a misinterpretation of academic language nature. The lack of attention to mainstream students will considerably inhibit students' academic language development and content area learni ng and thus must be addressed by the research. Second, the need for my study is established by the fact that current research focuses on college level learners. C urrent research does not pay much attention to adolescent l earners (i.e., secondary students ) A n extensive literature focuses on English for special purposes and university learners rather than school children and the demands of secondary schooling (e.g., Adamson, 1990 ). When students advance into secondary schools, the knowledge becomes more and more specialized, and the complexit y of language increase s dramatically as well. Although the demands of academic language are extensive for secondary students, there is a lack of attention to students' needs in developing academic language in secondary schooling Teachers often spend most of the time in teaching content knowledge and much less time in developing academic language. As the experts in their respective content areas, teachers expect their students to display their learning in a way that is appropriate in their specific discipli ne. However, these expectations about language use are not always made explicit (de Oliveira, 2011) Some teacher is not trained to identify and explain the important academic language features to students, which presents more difficulty to secondary stude nts in content area learning ( Bruna, Vann, & Escudero, 2007 ) Due to these challenges, secondary schooling is a significant phase to prepare
22 students for intensive disciplinary learning at college level. Therefore, m ore attention must be paid to academic l anguage use in secondary schooling. Third, current academic language research looks at academic language primarily from the perspective of vocabulary as if academic language consisted solely of vocabulary. But academic language is more complex than that and should be examined using a more comprehensive approach. Academic vocabulary is the words students must master in order to comprehend the concepts and display their acquisition of these concepts in any specific discipline ( Beck, McKeown, & Kucan 2002). As lexical choice plays a critical role in academic language, researchers pay close attention to academic vocabulary instruction and believe that academic vocabulary intervention can have a positive effect on academic language development (CalderÂ—n, 2007 ; Snow, 2008). However, academic vocabulary instruction alone is not sufficient for academic language development. A mere focus on academic vocabulary will impede students' academic language learning (Bruna et al., 2007) The nature of academic language inc luding its grammatical features and discourse structures must be identified and taught to develop students' academic language (Bruna at al 2007). Current research has identified many of the distinctive language patterns in academic texts (e.g., Halliday, 2008 ; Schleppegrell, 2001 ). By comparing a corpus of oral and written texts, researchers (Biber, 1988, 1995 ; Halliday, 1985; Schleppgrell, 2001 ) have identified a number of lexicogrammatical features of academic language and discussed how these features are different from the language of everyday social language. Although we know that students must use these academic language features successfully to fulfill the reading and writing demands in academic communities, the degree of secondar y students'
23 success in coping with academic language in content areas remains an unanswered question and thus must be addressed. In addition, current academic language research focuses on published materials rather than students' actual use of academic la nguage It, therefore, does not shed light on how academic language u se relates to students' reading and writing ability, which is critical to the design of effective instruction and remediation. When studying academic language, researchers usually analyze published materials including state standards, textbooks, reading materials, and oral texts transcribed in teachers' instruction and teacher student interaction. It is significant to know secondary students' actual use of academic language and how the use interacts with students' reading abilities. Although we assume that students with higher reading abilities can handle academic language more successfully, current research has not reveal ed how students' use of grammatical and lexical resource s differ acro ss reading groups. A comprehensive analysis of students' writing allows more insight into how students with different reading abilities grapple with academic language demands and exploit features at semantics, syntax, and pragmatics levels to represent the ir learning in content area knowledge. Another unknown aspect is the relationship between this use of academic language features and overall writing qualities. In the current research, whereas some academic language feature(s) is/are usually analyzed as a n indicator of students' academic language ability (e.g., Coffin, 2006 b ), how academic language features affect the overall writing quality is unknown. We do not yet know how the use of academic language features differ across various writing qualities and predict the se writing qualities. Knowing the interrelationship between students' use of academic language
24 features and their reading abilities and academic writing qualities will contribute to the ongoing discussion regarding adolescent students' literacy development in content area learning. Therefore, it is a research gap that must be addressed. Last but not least, academic language research typically involves small scale studies (e.g., case studies of 1 5 students). Due to the variety and complexity of academic language features, researchers tend to employ descriptive studies in analyzing these texts. Although these studies provide in depth information about the demands of academic language larger scale studies with larger group s of participants may pr ovide a more comprehensive picture of adolescents' academic language competence s. A method that enables a more convenient and practical analysis and interpretation of academic language features in a large number of writing sample s should be generated to be tter guide operationalization and practice in academic language. Overall, the current research is still evolving to build a thorough understanding of academic language. Although a theoretical groundwork has been established to suggest the challenge of aca demic language for both mainstream students and ELLs, the essential role of academic language features for secondary mainstream students has not been fully acknowledged in either research or instructional practice. The current literature indicates the natu re of academic language and the important role of academic language features such as academic vocabulary. However, more attention should be paid to a comprehensive conceptualization of academic language and all linguistic features of academic language. The current body of empirical studies do es not provide a detailed picture of mainstream students' capability in employing academic language in
25 content area learning. We also know very little about the interrelationship between use of academic language feature s and students' reading abilities. Other questions of interest include how mainstream students use academic language features to represent their content area knowledge, how the disciplinary community evaluates these students' academic writing, and how thes e two aspects interact with each other. All of the areas mentioned above must be investigated in order to contribute to the conceptualization of academic language which would then inform instructional and professional development decisions A comprehensiv e and in depth knowledge of students' abilities in coping with academic language may help content area teachers recognize their students' strengths and needs in academic language development This knowledge may also guide teachers to model and scaffold aca demic language and content area instruction. Research Questions The overarching goal of my study is to examine evidence of academic language in ninth graders' history writing and explore if such use is in any way associated with their reading ability and w riting quality. Three research questions will be examined: What academic language features can be observed in 9th grade students' historical writing and what functions do these features serve in historical meaning making? Does academic language use differ according to students' reading ability? Does academic language use differ according to the quality of students' historical writing? That is, to what extent does the use of academic language predict the overall quality of students' historical writing? Sign ificance of the Study My study intends to fill the gap of existing academic language research that focuses on college students but not high school learners, academic vocabulary rather
26 than linguistic features, and the language resources that students must know instead of how much they know about them. My study will also shed light on the relationship between academic language use and reading abilities as well as writing qualities. Overall, by focusing upon secondary students' academic writing in world histo ry classrooms, my study is designed to lead to further discussion regarding adolescents' academic language development. First of all, my study enables educators and researchers to identify what adolescents know and can do with academic language in the current learning context so that instruction or remediation can be designed to meet the ir needs. The focal population in the present study i s secondary students who have diverse reading and writing abilities The participants were in their fir st semester of high school and experiencing the transition from middle school to high school when my study took place. My study offers insight in to secon dary students' actual academic language competences as seen in this learning context by examining how they cope with academic language in historical writing. My study examines secondary students' academic writing in real content area context and demonstrat es students' uses of essential academic language features such as academic vocabulary, embedded clause, expanded noun, nominalization, and lexical density. The results will address the question of whether secondary students can handle academic language fe atures successfully in historical writing. The analysis of students' acade mic language proficiency in my study may help educators and researchers understand the demands of academic language fe atures on secondary students. This can lead to meaningful discus sion regarding instructional practices that address students' needs.
27 Second, the current study emphasizes a variety of academic language features, rather than academic vocabulary only. Within the framework of SFL, beyond the analysis of academic vocabular y, my study examines essential academic language features including embedded clause, expanded noun, nominalization, and lexical density. A comprehensive picture with respect to students' academic language proficiency can then be obtained. Our understandin g of students' academic language proficiency may move beyond word level to sentence and discourse level. Through the analysis of a variety of academic language features, we can gain understanding regarding the roles and functions of each feature and its sp ecific challenge for secondary students. My analysis can then provide rich and constructive information regarding the complexity and abstractness of academic language. The knowledge of a variety of academic language features can provide teachers and studen ts with tools to analyze linguistic features of academic language so that they can be more explicitly and directly taught and learned in classrooms. Last but not least, the connection between the evidence of academic language use and reading and writing ab ilities will be one of the most important contributions of my study. By examining the relationship between academic language uses and reading ability, my study will shed lights on the actual challenges of academic language to students of different reading abilities. This can test the assumption that students with higher reading abilities exhibit better command of academic language compared to students of lower reading abilities. The results may shed light on whether students of high reading abilities still need support in academic language and what support different reading groups need specifically. In my study the relationship between writing quality
28 and students' use of academic language features will be discussed. My goal is to understand what academic language features can affect writing quality and what the actual influences are. Through these discussions, my study will help practitioners, researchers, and policy makers reconsider the academic language challenges for secondary students and how these st udents are doing with academic language. With this information, practices, programs, and policies can be modified to address students' needs in developing academic language and content area learning. Limitations and Delimitations One limitation of my stud y is in the sample characteristics as all participants were selected from one school and one grade level. I recruited the participants from three world history teachers' classes. The c onvenient sampling method limited the interpretation and generalization of final results The sample size wa s medium (N=84). Nevertheless, I purposefully selected a research site that wa s a public school with a diverse population. Participants in this study were equally distributed based on gender, socioeconomic status, ethn ic groups, and acad emic achievements to allow for maximum variation. This delimited the deficits in data sampling. Furthermore, no conclusion will be made to generalize to a bigger population. The findings of this study can only be interpreted and generali zed in a similar type of context. Secondly, a number of variables such as different teachers' personal styles, classroom instruction, and different levels of content knowledge were involved in explaining the differences in academic language use among stud ents, which may affect the interpretation of this study. In order to control these factors to the greatest extent, the researcher worked with three teachers who were involved in this study on their curriculum t o ensure they were teaching s imilar content. T he researcher also visited
29 every class frequently to observe the delivery of the content However, there were still other factors such as teachers' instruction involved. In three levels of classes (Regular, Honor s and Advanced Placement class), teachers t aught the same content at different depth. For example, the teacher of the Honor s history class conducted more in depth discussion regarding Greek culture compared to Regular class. Therefore, i t was impossible to eliminate the variance among different types of classes even when all three classes were learnin g the same content. This is, therefore, a factor that must be taken into consideration when the results of this study are interpreted. Last but not least, a limitation in my analysis was that I did not include all aspects of academic language and all academic language features. Since academic language is a complex construct, it is impossible to analyze all aspects and all features in one study, which may also affect the conceptualization of academic language. To address this limitation, I conducted literature review to explore the academic language features that best represent the challenge of secondary content area learning. The factor that my analysis only included a number of academic language fea tures should be in the foreground of any interpretation of the results in the present study. Overview of the Study T he purpose of my study is to examine the use of academic language in adolescents' historical writing and explore the relationship between t he students' academic language use and their reading ability as well as the writing quality. My study draws both its theoretical framework and analytical methods from Systematic Functional Linguistics to analyze and discuss 84 adolescents' historical essay s In Chapter 2, I will first review what academic language is (i.e., the three perspectives on academic language) and what research says about the relationship of academic language to
30 school learning and success. Then I will present the research methods u sed to study academic language in Chapter 3 and findings in Chapters 4. Discussions and implications of these findings for research and instruction will be discussed in Chapter 5.
31 CHAPTER 2 LITERATURE REVIEW In this study, secondary students' use of ac ademic language features in historical writing will be examined This chapter describes how the literature review was conducted, beginning with the conceptualization of academic language. It then presents a critical examination of each area of inquiry, ran ging from the cognitive perspective, sociocultural perspective, to sociolinguistic perspective Finally, it weighs the evidence in support of relationships between teaching practices and academic language features and includes a summary of this research ba se. The intent of this literature review is to build a comprehensive and accurate conceptualization of academic language and identify what is known and what has yet to be done in the study of academic language development. Conceptualizing Academic Languag e Three paradigms grounded in different theoretical frameworks are cognitive perspective: Cummins' basic interpersonal communicative skills (BICS) and cognitive academic language proficiency (CALP); sociocultural perspective: Gee's Primary and Secondary Di scourse; and linguistic perspective: Halliday's Systemic Functional Linguistics (SFL). In order to understand these three perspectives I will illustrate each perspective's basic descriptions, important studies and critiques and its relevance and contribu tion to understanding academic language. Cognitive Perspective on Academic Language Cummins's early work (1981) is one of the first approaches that define academic language. He focuses on the difference between social language and academic language in co gnitive and contextual demands. He uses the dichotomy between BICS
32 and CALP to discuss the nature of language proficiency in the field of second language acquisition (SLA). Cummins revisits his own work in 2000 but the primary theoretical framework remains the same. According to Cummins (1981, 2000a), BICS is less complex and abstract than CALP and usually accompanied by extralinguistic and contextual clues. It is usually used in less formal settings such as family conversations. CALP tends to be complex a nd abstract with limited contextual and extralinguistic clues. It is usually the language of school settings. Cummins (2000a) also claims that language supported by contextual clues is easier to understand than language without such clues, and language on familiar topics is easier than language that contains new or difficult information. Moreover, the speeds at which BICS and CALP are acquired vary as well. In his study of immigrant students in Canada, Cummins (1981) states that ELL students learn conversat io nal skills in English in about two years but typically needs five to seven years of exposure to English in school before these ELLs are considered as academically successful. In other words, students learning a second language need more time to acquire a cademic language than to acquire social language. On the basis of the dichotomy between BICS and CALP, Cummins (2000a) defines academic language proficiency as the extent to which an individual can access and master the oral and written academic registers of schooling. Gaining control on academic language means that students are capable of making complex meanings explicit in oral or written mode of language without resorting to contextual or paralinguistic cues such as gestures and intonation (Cummins, 2000 a). To achieve that, students must be acquainted with an ever increasing vocabulary, concept load involving
33 words specific for school contexts, different syntactic features and discourse conventions. Basically, students need to acquire all features of acad emic language to function successfully in academic tasks. Knowing the distinctions between CALP and BICS has numerous implications, particularly for ELLs. For instance, learning in content areas does not necessarily take place even though students can comm unicate fluently in daily life. In content area learning, language can be more abstract and complex with unfamiliar vocabulary and syntactic features and students may lack extralinguistic support. Also it is not sufficient for ELLs to learn academic langua ge and content knowledge in English language class. ELLs must lea rn content specific language in content areas. Modifications to the traditional ESL curriculum during the early 1980s have been made to combine content learning and academic language developm ent in order to provide ELLs specific instruction on CALP in various content areas. These modifications are one of the direct implications of Cummins' model. Cummins' theoretical approach to conceptualizing academic language has been criticized for a numb er of reasons. Martin Jones and Romaine (1986) point out Cummins' inadequate linguistic modeling. They claim that without conducting linguistic analysis on social and academic discourse Cummins fails to establish an adequate model of social language and ac ademic language. Criticism has also been made that the dichotomization between social language and academic language cannot explain the complexities of language and literacy. Bailey (2006) suggests that BICS is not necessarily less cognitively demanding th an CALP. For example, to contrive plausible excuses, negotiate and persuade in everyday life involves BICS, it takes more mental
34 effort and cognitive ability than comprehending a paragraph on Civil War in a 5th grade social studies textbook (CALP). BICS ca n exist within context reduced settings. Children's make believe play, for example, involves complex and highly abstract reasoning. Schleppegrell (200 4 ), on the other hand, points out that CALP is not necessarily more linguistical ly difficult than BICS. I t is also inaccurate to describe academic language as decontextualized (context reduced) because all language is produced and used within a context. What differentiate academic language from interactional language is the grammatical elements that are used t o construe the context (Schleppegrell, 2004) Furthermore, with the growing recognition of the sociocultural nature of literacy, Cummins has been criticized for disconnecting learning to read and write from the sociolinguistic experience of everyday life a nd for attributing higher status to CALP (e.g.,Wiley, 1996). F urther critiques ( e.g., MacSwan, 2000) argue that through the distinctions between BICS and CALP Cummins attribute ELLs' academic failure to their low cognitive/academic proficiency rather than to inadequate schooling, thereby promoting a "deficit theory" explanation for ELLs' underachievement. To address the critiques, Cummins (2000a) revisits his theory and discusses many of the above mentioned issues. First of all, he conceptualizes the diffe rences between BICS and CALP in linguistic terms to address the critic ism of Martin Jones and Romaine (1986). Citing Biber's (1988, 1995) detailed empirical analysis of a large corpus of spoken and written textual materials, he lists three major dimensions of the differences between written and spoken materials. These three dimensions are labeled as interactive vs. edited text (contrast features showing high personal involvement with those which allow editing; first and second person pronoun is an interacti ve feature);
35 abstract vs. situated content (features such as nominalization and passive voice make a text abstract), and reported vs. immediate style (past tense, perfect aspect, and 3rd person pronouns that can be referred to narrative and non narrative c ontexts and reported style). A number of linguistic features that are characteristic of written and spoken texts are also listed. For example, high personal involvement and real time constraints are characteristic of spoken text, whereas nominalizations an d prepositions contribute to detachedness in written texts. Cummins (2000a) suggests that these distinctions between oral and written language are consistent with the broad distinction between CALP and BICS. This type of linguistic analysis is a preliminar y analysis of linguistic differences between social language and academic language rather than an established linguistic model of academic language. While his early work is more situated in cognitive perspectives, Cummins incorporates the sociocultural th eory of literacy by claiming that language development and language proficiency should take into account the context in which language is used (Cummins, 2000a, 2000b). In different contexts, such as academic settings and everyday life, specific linguistic registers are required for adequate and successful functioning. Cummins (2000a) proposes that language proficiencies are "intervening variable" (p. 96) that me diates students' academic development. Language proficiencies are shaped by and specific to the contexts in which they are used. Although academic language is not linguistically or cognitively superior to any other forms of language, it is related to acade mic success in the context of schooling. Reading and writing skills are important to CALP, but CALP consists of more than reading and writing. For example, students must exhibit their CALP in an academic presentation. Cummins focuses on
36 how providing stude nts' access to academic registers of languages can help bilingual students achieve school success to fight for the coercive power relationship in this society. Cummins (2000a) also extends his focus on ELL to all students by claiming that the distinction b etween BICS and CALP reflects a pattern of language development that also exists in native English speakers (Cummins, 2000a). Cummins' analysis of BICS and CALP examines language proficiency through context, register, and cognitive as well as linguis tic features. His model has a profound impact in the field of second language acquisition. Cummins' model takes into account the different contexts of social language and academic language. However, the complexity of context such as different disciplines h as not been studied to explain the complexity and variability of language proficiency. For example, writing a narrative that describes Shakespeare's life and works in a language arts classroom and a report that presents the result of an experiment in a sci ence classroom pose distinct language demands on students, although both are complex, cognitively demanding tasks. Cummins also pays little theoretical attention to monolingual or mainstream students and thus its implication for students in mainstream cont ent area classrooms is very limited. Sociocultural Perspective on Academic Language Academic language is also conceptualized from a sociocultural perspective, with James Gee (1996) as the key representative. In exploring the use of academic language in sc hool settings, James Gee (199 6 ) conceptualizes academic language as a family of social languages that are related to each other. However, this social language is significantly different from the social language in Cummins' paradigm. For Gee, social
37 languag e refers to how language is used to carry out socially situated activity and realize socially situated identity. Gee's distinction between Discourse (with a capital D) and discourse (with a lower case d) is based on the definition of social language. Spec ifically, Gee (1996) defines "discourse" (with a lower case "d") as a social language or language in use through which activities and identities are enacted. D iscourse (with a capital "D") is a much wider concept involving non language elements. "Discourse (with a capital "D") means "a socially accepted association among ways of using language, of thinking, and of acting that can be used to identify oneself as a member of a socially meaningful group or a social network" (p. 131). It is an "identity kit" th rough which community members know how to act and talk in the specific community so as to take on a particular role that others will recognize. A very important implication of D/discourse is that discourse is a constituent element of Discourse. A discours e is a social language that is associated with a specific Discourse. Different discourses are reflected in different grammatical and lexical features. Knowing a discourse means knowing how to use these grammatical and lexical features that are characterize d by associated Discourse and thus become a member of this Discourse. On the basis of "discourse" and "Discourse", the concepts of Primary Discourse and Secondary Discourse are introduced (Gee, 1996). The Primary Discourse is one's "first social identity." It is "those to which people are apprenticed early in life during their primary socialization as members of particular families within their sociocultural settings" (Gee, 1996, p.137). All humans, barring serious disorder, get one form of discourse
38 throug h acquisition, which is usually their Primary Discourse. Primary Discourse varies among different socio cultural groups even among all native speakers of English. For example, lower socio economic black children use English to make sense of their experienc e differently from middle class children and therefore they have different Primary Discourses (Gee, 1996). Secondary Discourses are those to which people are apprenticed as part of participation and their socialization in a variety of local, state, and n ational groups and institutions outside early home and peer group socialization. Secondary Discourses build on and extend the uses of language acquired as part of our Primary Discourse. All Secondary Discourses involve uses of language, either written, ora l, or both. This is called "secondary uses of language." For Gee, literacy is the mastery of a Secondary Discourse (Gee, 1996). As Gee (1996) points out, Primary Discourse and Secondary Discourse constantly negotiate and contest in society. Primary Discour se serves as a framework for future acquisition and learning of other Discourse. Other Discourses influence Primary Discourse as well. Primary Discourses and Secondary Discourse are compatible to different extents depending on different social groups. Wi thin the framework of Discourse, by comparing the story telling transcripts of an Anglo American, middle class student and an African American student from a lower socioeconomic home, Gee (1996) suggests that it is a great advantage for learning when the S econdary Discourse is more compatible with the Primary one. The analysis shows that language features construe the social relationship and students' performance is shaped by their Primary Discourses. According to Gee (1996), many
39 African American people ha ve a Primary Discourse that emphasizes mutual participation and social networking. In this Discourse, authoritarian texts are not privileged and written words are not over people's voices. On the other hand, white, middle class culture is usually more conn ected to the western middle class and modern capitalism that have set the foundation of middle class primary socialization as well as modern formal schooling. Therefore, it is not surprising that the white students' story telling was more rewarded in the d escribed school context. Based on this analysis, Gee suggests that students' success in school is related to how much students' home based Primary Discourse is melded with school based Secondary Discourse. Students also benefit from this homology in the fu ture jobs and the social worlds. In terms of how people come by D/discourses, Gee (1996) illustrates different processes in terms of "acquisition" and "learning". Acquisition is a process of gaining something that is necessary for functioning in a particul ar situation by repeated long term exposure. Accordingly, learning "involves conscious knowledge gained through teaching" (1996, p. 138) but this learning does not have to involve someone called a "teacher." Learning involves reaching an understanding (or partial sense) of meta knowledge. It is tied to having a conscious awareness of the process by which one is gaining this understanding. While some cultures highly value acquisition in which children are exposed to adults modeling, some other groups highly value teaching by explanation and analysis. For anyone living in our society and culture, both acquisition and learning need to happen to get full access to a discourse and a Discourse. Nevertheless, D/discourses are mastered by acquisition instead of ov ert instruction. To fully attain a D/discourse,
40 students need to be acculturated or apprenticed into soci al practices by someone who has mastered this D/discourse via scaffolding and interaction. While modeling and instructional guidance are important, D/d iscourses are typically acquired through immersion in meaningful social practices. Still, in order to critique one D/discourse and further change it, learning must take place to facilitate meta level knowledge in both D/discourses. Within theoretical fram ework of Primary Discourse and Secondary Discourse, Gee (2004a) discusses further about academic language and attributes high significance to it by claiming that, although alphabetic code is important, more students fail in school in the long run because t hey cannot cope with academic language rather than because they cannot decode words. According to Gee (2004a), English is a language that is composed of a myriad of social languages. A cademic language is defined as a large family of related social languag es or discourses that are associated with Secondary Discourse of schooling. As language is always situated within particular ways of communicating and understanding information, academic language is associated with, but not limited to, grammatical patterns These grammatical patterns construct particular identities and activities in the context of academic disciplines. Gee (1996) suggests that disciplines such as mathematics, science, and art, have their own Discourses that students must learn in order to s ucceed in those disciplines. Since academic language represents a variety of discourses situated in academic Discourse, learning academic language is more than just learning language in use, such as discrete language rules. It is participating in a Second ary Discourse that
41 involves interaction with others who do not necessarily share the same knowledge and experience (Gee, 1996). In secondary content area classrooms, students learn to utilize a variety of lexical and grammatical choices to construct social practices and relationships that fit Discourse of disciplines. Gee (1996) and oth er scholars (e.g., Delpit, 1988 ) consider f eatures of academic language as meta knowledge or power code that constitutes a style that is designed to exclude the marginalized outsiders and to enhance the status of powerful insiders. Gee (2004a) points out that students will acknowledge the value of learning academic social language only if they value the socially situated identities and activities where academic language is si tuated To many students, learning academic language represents a big loss in their primary discourse and identity. Their life world or primary Discourse is disassociated with the middle class Discourse that historically share interest and values with some academic specialist domains. For these students, learning to value the identities and practices (but not deny primary identities) in an academic Discourse is the foremost step of learning academic language. Furthermore, Gee (1996, 2004a) emphasizes that s chools should teach students to reflect on and critique the dominant Discourse in school and the Discourse maps of their society and the wider world. Diverse Discourse s should be juxtaposed to each other in schools to allow students to understand them at a meta level so that students will be able to transfer and vary their Discourse, and create new Discourses. This meta knowledge, Gee (1996) believes, is the core ability that school should inculcate and explicitly teach. As Gee (1996) regards science as the subject that represents best the demands of academic language, he specifically analyzes the subject of science to study features
42 of academic language by conducting a preliminary analysis on the style of language. Two examples from a student's two drafts o n the topic of Albinism were provided: First Draft: Then to let people know there are different types of Albinism, I will tell and explain all this. Second Draft: Finally, to let people know there are different types of Albinism, I will name and describe several. I n Gee's analysis, how the author situated the meaning of tell and explain is inconsistent with the situated meaning of different types of Albinism. When discussing about Albinism in academic writing, apparently name and describe are m ore appropriate, if not the best, words that situate the meaning in an academic way. According to Gee, features of academic language, or meta level cognitive and linguistic skills must be built into the curricula and integrated with acquisition in practic al ways. Students' primary Discourse and the language exposure they encounter prior to school years play a part in their readiness for academic language. Those who do not have much home experience with interactive "book talk" and extended single topic pres entation must be exposed to a variety of written and spoken models of language in content areas and develop an academic identity. Nevertheless, Gee (1996) suggests that the language models of academic language cannot be overtly taught S tudents must be a pprenticed into social practices and interaction because scaffolding plays a crucial role in learning and acquiring academic language. Delpit (1998) criticizes this stance by rejecting the idea of apprenticing students of color and poverty in academic lang uage. Delpit (1998) maintains that every Discourse has its culture codes. For school Discourse, both the "superficial features" of middle class Discourse -grammar, style, mechanics, features -and the more subtle aspects of dominant Discourse would cont ribute to students' school
43 success. Those who do not have access to the dominant Discourse would never get these subtle aspects without explicit instruction about these codes. In more updated literature, Gee (2004a) addresses this issue by claiming that s ome explicit instruction is needed when interactive experience can not provide students with sufficient understanding of academic language. He provides example that, in face to face situations, teachers and students tend to use truncated language including deictic, vague references, and ambiguous structures. Interlocutors tend to use numerous pronouns to refer to persons within the context. The cause, in a cause effect, is sometimes implied as the interlocutors who both know the context. In written models o f academic language, expanded forms of academic language appear in the reading and writing in overt ways to make meaning understood by publics, which are valued by evaluation in school. Any pronoun must be specified to a participant in the text A ny cause and effect must be listed to build a cause effect relationship. Therefore, students who do not have much experience with this elaborated form of language must be involved in explicit discussion concerning the identification and uses of academic language. T o Gee (2004a) both explicit attention to expanded forms of academic language and interactive experience of academic language practice are necessary for students to acquire academic language. He also argues that schools are not doing enough to support stud ents in acquiring academic language and that students need extensive authentic interactive experiences with proficient member from academic Discourse community. Overall, Gee's theory of "Discourse" (with a capital "D") and "discourse" (with a lower case "d ") builds groundwork for discussing academic language. Academic
44 language is an array of language and linguistic feature s. It is a "discourse" (with a lower case "d") that is situated in Discourse (with a Capital "d") of secondary content areas. Direct and explicit instruction, interaction and scaffolding should be combined to develop academic language. As language patterns are associated with Discourses that are situated in c ontext, the Discourse of academic language varies with different disciplines and subject areas. To have full access to academic language and academic Discourse, students need to learn a range of power codes or meta knowledge including the lexical and gramm atical features of academic language. Nevertheless, Gee's analysis of D/discourse does not provide us with a tool to analyze these lexical and grammatical features of academic language and build connection between these features and their social functions. Functional Linguistics Perspective on Academic Language A more powerful model for conceptualizing academic language is in systemic functional linguistics, or SFL (Halliday & Matthiessen, 2004). SFL provides us with a definition of academic language, tools to analyze grammatical and lexical features of academic language in different content areas, and insights into teaching and learning academic language. In this section, an introduction of SFL including some significant concepts such as register and genre will be presented. A preliminary analysis of lexical and grammatical features of academic language will also be developed. SFL is a language theory that focuses on the context, social purpose of language and the relationship between text and context. It is introduced by Australian linguist M.A.K. Halliday (Halliday & Matthiessen, 2004). In this theoretical framework, literacy is viewed as a semiotic, a form of social action and a process of meaning making in which language and context co participate. Langua ge use is based upon and also constrained
45 by context. Language consists of various grammatical and lexical choices that construct different meanings for different purposes. SFL explains the symbiotic relationship between social context and language enabl i ng a direct and explicit connection between grammar, lexicon and discourse features and social function. From a SFL perspective, every text constructs three kinds of meaning the experiential (the content or main idea of a text), the interpersonal (the i nterpretation and judgment presented in a text), and the textual (the way a text is organized as a coherent message) (Schlepp e grell, 2004). These meanings are constructed in different configurations of language resources. D istinc t language systems allow d ifferent kinds of meaning to be made to meet a variety of purposes in various contexts. They can be used to represent personal experience of the world; establish and maintain relationship; and form texts. The way these language resources are used varies in relation to the context from which the text emerges. Language enable s the learner to achieve the social goals such as asking for directions or answering math questions (Christie & Derewianak, 2008). Experiential, interpersonal, and textural grammatical fe atures work together simultaneously to realize context of situation (Schlepp e grell, 2004). In SFL, the manifestation of the context of situation is referred to as register. Register is defined as "the configuration of lexical and grammatical resources whic h realizes a particular set of meanings (Schlepp e grell, 2004, p. 45 46). Three variables determine a register: field (what is talked about, the subject matter of the discourse), tenor (participants in communication and their social relationships), and mode (expectations of how particular types should be organized, the format of language, such as the differences between
46 speech and writing) (Halliday & Matthiessen, 2004; Schleppgrell, 2004). This means that register is characterized by some combination of gra mmatical and lexical features that differ in different contexts. Diverse registers are also more or less valued (Schleppegrell, 2004). In current U.S. society, academic register is highly valued particularly in the school context. Moreover, the degree of e ffectiveness of different grammatical choices realizing particular text types in certain situation also varies (Schleppegrell, 2004). In other words, a text that realizes the expected register in a certain context is considered more effective than the text that does not realize the register. Academic language: register of the school context Based on the construct of register, SFL sees academic language as means of making meaning in the school context (Schleppegrell & Colombi, 2002). Respectively, the field of discourse is realized in experiential resources; the tenor of discourse is realized in interpersonal resources; and the mode of discourse is realized in textual resources (Schleppegrell, 2004). Only through this "three way perspective" (Schleppegrell, 2004, p. 46) can the link between meaning and function be analyzed through grammatical expression (Schleppegrell, 2004). This three way perspective of SFL thus serves as the foundation for the following comparison between two texts. These three meanings o f academic language are realized through a variety of grammatical features, such as the use of clause combining strategies (the type of clause combination), nominalization (the use of a verb or an adjective into a noun so that the word can act as a constit uent of a clause), use of noun (simple or expanded noun), and lexical density (the measure of the density of information in any passage of text). These featu res will be discussed in detail later.
47 Text (1) is selected from a scientific text What's So Hot a bout Peppers ". This text is a selection from the book Preparing for FCAT Reading (Grade 9) (Spigler, 2005). In this book, a variety of reading texts mostly in science, history, and language arts are included to help students practice reading to prepare for Florida Comprehensive Assessment Test (FCAT), a standardized test of reading in the state of Florida. Text (1) represents the text that students encounter regularly in secondary schooling, and therefore serves as an example of academic register. Text (2) is selected from the transcription in Gee's (1996) analysis of the story sharing of an Anglo American, middle class seven year old girl, Mindy. It is considered an appropriate and successful example of sharing time, which in here is an example of inte ractional register. Please refer to Table 2 1 for the two texts. Overall, text (1) is denser, more complex, technical, and abstract, which is the type of written text that secondary students encounter in their content area classrooms. To the contrary, text (2) is clearly a dialogue. Whereas this is just an overall impression, SFL enables us to analyze the differences in register features between these two texts and provide evidences to support this impression. As the first step, both texts can be seg mented into clauses as in Table 2 2. Next, using a three way perspective of SFL, the grammatical features that realize three different kinds of meanings can be analyzed: Experiential meaning The experiential meaning refers to the content or main idea of a text. It is usually realized through vocabulary choices (technical vocabularies vs. non technical vocabularies; long nouns vs. simple nouns). In terms of vocabulary, Text (1) draws on an abstract and technical lexis (e.g., stimulus, endorphins, chemical, substan ce ), while
48 the lexical terms in T ext (2) are non technical and concrete (e.g., wax, water, gets to the size ). Interpersonal meaning The interpersonal meaning concerns the interpretation and judgment presented in a text. Mood structure is a major grammatic al resource that establishes the stance of a text. English has three mood options declarative, interrogative, and imperative. In addition to mood, there are word choices that differentiate the two texts along the interpersonal dimensions. For example, mo dality and other appraisal resources can enable language users to talk about possibility, certainty, usuality, normality, seriousness, necessity, and obligation. In addition the nouns in Text (1) tend to be long and complex, (e.g., a chemical that transmi ts the impulse interpreted by the brain as "Pain!" endorphins, the body's natural painkiller ) whereas simpler nouns and pronouns are frequently used in Text (2). Long noun phrases construe density. Through this, knowledge is displayed in a dense and compl ex way in academic text like T ext (1). Textual meaning The textual meaning has to do with the way a text is organized as a coherent message including clause combining strategies, thematic progression (the way that information is organized through sentence s that scaffolds the organizational structure) as well as nominalization. In Text (1), an embedded clause ( that transmits the impulse interpreted by the brain as "Pain!" ) is used. The use of embedded clause results in complex nominal phrases and dense sen tence structures. Another difference between these two texts is the way in which logical relationship is realized and conjunctions are used. In Text (2), Mindy relies on
49 conjunctions to make the point, such as linking to previous discourse ( and ) and introducing a condition ( until ). The text structure is marked by conjunctions as discourse markers. However, in Text (1), there are no causal conjunctions such as because ". Instead the logical connections are realized through nominal and verbal expression s such as causes, triggers and the stimulus The sentence such as Capsaicin causes a complex series of events to occur replaces lengthier sentence and looser structure such as a complex series of events occur because of capsaicin The conjunctive rela tionship is integrated into clauses. More integrated logical relationships also contribute to a dense presentation of information in academic text. To summarize, from the perspectives of experiential, interpersonal, and textual meaning, we can see that con text of school is realized through a range of lexical and grammatical choices, which makes academic language different from everyday language. Grammatical Features of Academic Language In the theoretical framework of SFL, analyses have been conducted to e xamine register features that are pervasive and characteristic in students' school reading and writing (e.g., Bailey & Butler, 2003; Schleppegrell, 2001). According to th is body of research the function of academic language is expository because academic texts focus on presenting ideas to be analyzed and interpreted (Schleppegrell, 2004). Academic langu a ge is complex, dense, and abstract, which is different from informal interactional language (e.g., Fang & Schleppegrell, 2008; Schleppegrell, 2001, 2004; S chleppegrell & Colombi, 2002). Grammatical choices are employed to construe these functions (Schleppegrell, 2004) Linguistic features of academic language include academic vocabular y embedded clause, expanded noun phrases, nominalization, and
50 a high lexi cal density. These discussions serve as the foundation for our discussion and provide a list of language features that is the most significant, prevalent, and also more universal to academic language. Academic vocabulary To communicate in academic discour se, it is important to know a large number of academic vocabulary words Vocabulary, as the most salient lexical resources across registers, constitutes a significant part in teaching ELL and struggling readers (Carlo et al., 2004). In academic language, abstract and generic vocabulary is one of the major devices that construe technicality and abstractness in academic texts (Fang, Schleppegrell, & Cox, 2006). Secondary students must be able to use different types of academic vocabulary words including tech nical vocabulary words that are specific to academic disciplines, non technical academic vocabulary words that are used across content area s, and multisyllabic words (e.g., Coxhead, 2000 ; Scarcella, 2003; Zwiers, 2008). One way to conceptualize academic v ocabulary is in terms of tiers or categories. A commonly accepted classification system frames academic vocabulary according to three tiers (Beck, et al. 2002; CalderÂ—n, August, Slavin, Duran, Madden, & Cheung, 2005). The first tier is non academic, conve rsational vocabulary such as flower or help ; the second consists of general academic words such as however and influence ; and the top tier is composed of content specific, technical vocabulary such as organism or s upplement This three tier approach has be en widely used. Among different tiers, researchers suggest that Tier 2 words, general vocabulary are the most important to teach (CalderÂ—n et al 2005; Snow, 2008).
51 Expanded noun phrase Other than academic vocabulary, long and complex nouns phrases are frequently used in academic writing. Structurally, nouns can be simple or complex. Simple nouns are nouns without modification, including pronouns (e.g., we, it ) and proper nouns (e.g., Flo rida ). Expanded noun phrase s have more than one modifier (e.g., a pleasant sensation of well being ) I n this example a pleasant is the pre modifier and of well being is the post modifier. Nouns can take on a range of functional roles including creating tax onomies, expanding information, condensing previously mentioned messag e, and building argument (Fang et al. 2006). As a powerful resource that contributes to the density of academic language, expanded noun phrases pack loads of information in fewer clause s with multiple pre and post modifications that include adjectives, other nouns, adverb, ed/ ing participles, prepositional phrases, and embedded clauses. For example, Text (1) uses embedded clauses (e.g. a chemical that transmits the impulse interpreted by the brain as "Pain!" ) to pack a lot of information into the nominal structure, whereas the lexical terms in Text (2) are mostly simple nouns (e.g., wax, water, size ). The ability to use expanded noun phrases in academic writing indicate to a great extent students' reading and writing abilities in academic register S tudents who are proficient in academic language use more expanded noun phrases (e.g., Hunt, 1965; Fang, 2008b). For example, Fang's (2008b) analy zes the use of nouns in third fifth seventh and ninth grade students' scientific writing samples. By calculating the percentage of every type of nouns including simple nouns and complex nouns in the low quality and high quality texts in these four gr ade levels, Fang (2008b) suggests that regardless of grade level, high quality texts use more complex nouns than low quality
52 texts and that more complex nouns are used in texts of higher grade levels than those in texts of lower grade levels. T he use of ex panded noun phrases is believed to be one of the strong indexes of academic language. Nominalization Nominalization is a type of grammatical metaphor. When developing academic writing the increasing use of abstractions, generalizations, impersonality, a nd lexical density derive s from the use of grammatical metaphors (Martin, 1997) Developing academic language means having a good control of grammatical choices to realize abstractness and technicality in academic writing. One of these choices is grammatic al metaphor, a pervasive academic language feature that expresses meanings in incongruent grammatical forms (Halliday, 1985) By using grammatical metaphor, an expression that could be construed more naturally in daily interactional language is realized in a more abstract and technical way. Halliday (1985) sees grammatical metaphor as a complex and significant step in a student's language development. According to Halliday and Matthiessen (2004), grammatical metaphor is conceived as an incongruent realizat ion of a given semantic configuration in the lexicogrammar. There are various types of grammatical metaphors. Grammatical metaphor includes interpersonal metaphor using a mental process of cognition, such as "using first person present tense mental process es of cognition ( I suppose ) instead of modal verbs to realize low probability (Martin, 1997, p. 28 ). Grammatical metaphor can also be interpersonal metaphor that uses question to represent an imperative statement. For example, when we say would you pleas e open the door," we actually mean open the door." Grammatical meta phor can also be experientia l. For example, construing
53 causality not as conjunctions ( because, so ) but as nouns ( reason ), verbs ( trigger ) or prepositions ( with ) is an example of grammatica l metaphor. As an essential type of experientia l grammatical metaphor, nominalization is the process by which verbs, adjectives, or what would more naturally be presented in another form in interactional language become noun phrases (Eggins, 2004; Schlepp gegrell, 2004). According to Eggins (2004), it is "situations where meanings typically (congruently) realized by one type of language pattern get realized by less typical (incongruent) linguistic choices" (p. 99). Nominalization particularly serves to reca st everyday language in a more specialized way and contributes to technicality, reasoning, and abstractness that characterize academic language. Nominalization contributes to the density in a text by allowing information that otherwise would be a whole cl ause to be packed into a noun phrase (Martin, 1997) By changing a congruent (verbal or clausal meaning) in to an incongruent expression (nominal element), nominalization can enable something that has been presented as a clause to be distilled into a noun p hrase. Nominalization is usually related to linguistic features such as complex noun phrases with pre and post modifiers as well. One of the cognitive functions of nominalization is reasoning (Schleppegrell, Achugar, & Oteiza 2004). Nominalization help s the reasoning to be developed in academic writing. In academic writing, nominalization usually distills the information previously mentioned in to a noun phrase. This noun phrase can then participate in a chain of reasoning. In historical writing, nominal ization may be used to represent a category of reasons that explain a historical fact, which can significantly increase the academic ness in the reasoning.
54 Nominalization also construes technical knowledge and contributes to the technicality in the text (Schleppegrell, 2004). Another main cognitive function of nominalizations is categorizing, through which nominalization constitute the basic principle of forming technical terminology. Nominalization also makes it possible to construe hierarchies of techni cal terms. Through nominalization, a step by step argument can be developed using complex passages packed in nominal form as themes. This also contributes to the technicality of a text. In written language, nominalization is one of the most important indicators of a more formal and abstract writing style (Connor, 1995). It is also regarded as the most important feature of students' development from an oral language style to academic writi ng (Colombi, 2002). N ominalization play s an important role in academic writing development. As students move through the grade level s, they are expected to use nominalizat ions in their writing to demonstrate that they understand the more abstract concepts in these subjects and also express the concepts in a formal way. Embedded clause In secondary school content areas, academic texts usually present information in a hierarchical structure (Schleppegrell, 2004). This is construed through textual resources su ch as the clause combining strategies which is how clauses are combined with different types of clauses such as embedded clauses and nominal structures. C lause combining strategies are central to register differences (Colombi, 2002). Clause is the primar y constituent of text in writing research from a grammatical perspective. It is defined as grammatical structures with finite verbs (finite verb is a verb that shows agreement with subject and is marked for tense) and can be categorized as
55 main, paratactic hypotactic, and embedded clauses (Colombi, 2002). A more detailed definition and examples from text (1) and (2) can be found in Table 2 3: Clause combining strategies play an important role in information structure and thus are characteristic of differen t registers (Colombi, 2002). In informal spoken language, we tend to make a sequence of clauses that are hypotactic or paratactic, whereas in academic register more embedded clauses are used to pack more information in less independent clauses. Therefore, relatively high percentage of embedded clauses in writing text is regarded as a marker of academic language. Clause combining strategies can also be highly related to writing qualities (Schleppegrell & Colombi, 1997). A well planned and tightly constructe d essay tends to have embedded clauses that are linked in logical ways. To the contrary, a more emergent organizational structure needs more extension and elaboration of ideas, which results in more uses of paratactic and hypotactic clauses. Studies have s hown that as students develop their academic language, their writing moves from more oral to more written styles that are characterized by reduced clausal structures (e.g., Colombi, 2002; Crowhust, 1990; Menyuk, 1988) and less parataxis and hypotaxis (Hall iday & Matthiessen, 2004). Clause combining strategies have been used as one of the most important indicators by researchers when exploring students' academic language developm ent trend (e.g., Colombi, 2002). Colombi (2002) examine two bilingual students' Spanish academic writing and use clause combining strategies, grammatical intricacy, nominal structure, and lexical density as measurement to determine students' growth in academic writing. The results suggest that students move from informal register tha t draws on hypotactic and paratactic clauses to more formal register that
56 draws on main clauses and embedded clauses. Because of the important functions that different clauses can realize, the use of embedded clause must be considered in this study. Lexi cal density Lexical d ensity is a measure of the number of lexical items or content words per non embedded clause in a text. Lexical items are the content words including nouns, adjectives, verbs, and some adverbs (Halliday, 1985). To the contrary, the gram matical items include articles such as the and a propositions such as in and on pronouns such as him auxiliary verbs such as had conducted conjunctions and and demonstratives this In conversation, we use fewer lexical items and more grammatical items But in academic writing, more lexical items must be included because we need to express the meaning through the language itself without relying on contextual or extralinguistic clues such as intonations (Christie & Derewianka, 2008) Lexical density reve als how tightly the lexical items have been packed in a text and how difficult a text is to read. It can be a summary measure of academic language as all three metafunctional meanings (experiential, interpersonal, and textual) work together simultaneously and result in different lexical densities of various registers. Eggins (2004) points out that lexical density is a linguistic feature that is the most sensitive to mode variation because it signals the most striking difference between spoken and written la nguage. Whereas spoken language usually has low lexical density and more complex structure with frequent conjunctive links, academic text s usually ha ve more content words and e mbedded clauses and therefore are more densely structured. In order to create a flow of information and facilitate reasoning and argument, academic texts
57 frequently use expanded noun phrase s, nominalization, and other grammatical metaphors that all contribute to packing more information. This may result in academic language's hig h lexical density. As Halliday and Matthiessen (2004) suggest, two content words per clause is typical for informal spoken language, whereas this density is around four to six in written language and even higher in scientific English. In text (1), there are 39 content words and seven non embedded clauses, and thus the lexical density is 5.57. In text (2), the lexical density is 15/7=2.14, which is much lower than the first one. To summarize, a number of language re s ources realize academic register and build a conceptual model of academic language Register, as the situational context in which a text is produced (Eggins, 2004), determines the social functions and language patterns that realize these social functions. Academic register thus decides and also relies on the grammatical and lexical features that we just discussed to realize different meanings. Table 2 4 is adapted from Schleppegrell's (2004) Form of Register Features of the Language of Schooling (p. 74). Th is form lists three types of situational expectations of academic language that correspond to the variables of field, tenor, and mode of the academic context. It also gives us a comprehensive picture of these grammatical and lexical features in three meani ngs (experiential, interpersonal, and textual) of academic language. Please refer to Table 2 4 for these grammatical and lexical features. Proficient mastery of these register features is significant for developing academic language and learning knowledge in secondary content areas. Building upon this, SFL provides strategies for evaluating the quality of students' writing (Fang &
58 Schleppegrell, 2008; Schleppegrell, 1998) and support secondary teachers to develop students' reading and writing abilities. Whe n analyzing students' writing, SFL is concerned about students' effectiveness in using linguistic features to construe the academic ness in historical writing. SFL also looks beyond the grammatical errors and focuses on students' strength. Students' writin g development is considered as a continuum of competency between an interactional, colloquial style and an advanced use of register features in academic writing. To learn the register features of academic language, the best pathway to learn language is th rough social experience or apprenticeship (Gee, 1996; Schleppegrell, 2004). It is significant for students to be provided with meaningful experience and context in order to learn these linguistic features and develop academic language. It is also essential to provide tools for both teacher and students to analyze academic texts a nd examine whether students are developing academic language features to meet the linguistic expectation of academic register. Furthermore we may discuss whether or not the classro om context and social practice are really helping students develop their academic language to facilitate content area learning. In this way, more likely the expectations of schooling can be reconsidered and the pedagogical approaches regarding reading and wring instruction can be informed (Schleppgegrell, 2004). Academic language: different content areas and different genres A cademic language is dynamic and varies with respect to factors such as purposes, topics, and situations. Different disciplines have different purposes and thus require the use of different genres to accomplish those purposes. Language choices also vary across genres (Fang & Schleppeggrell, 2008). This dynamic nature of academic language presents a challenge to students: learning to und erstand and use
59 these features to build different types of texts in different school subjects. For example, language in history and language in science have different purposes. In science, students are frequently required to write scientific report that de scribe the steps of experiment and discuss the application of results. In history, students usually write about historical events and analyze the cause and effect within events. Different language resources must be deployed to realize distinct purposes, to pics, and situations. When we discuss the language of discipline, it is important to know the concept of genre that reflects different purposes of language and variations within each discipline. The notion of genre describes particular types of texts that serve specific social purposes and characterize particular social contexts (Schleppegrell, 2004). According to Eggins (2004), genre is the context of culture, which is "the general framework that gives purpose to interactions of particular types, adaptabl e to the many specific contexts of situations that they get used in" (p. 63). Each genre can be identified by its communicative purpose and linguistic features. Genre reflects speakers' cultural expectations about what goes first, next and last. On the bas is of various purposes that different genres intend to achieve, texts of different genres can be realized through different lexical grammatical choices. Defining the features of a genre can never be exclusive or definitive, but it is useful and important t o know that a range of lexical and grammatical features characterizes various genres. Seven prototypical school based genres proposed by Martin (1989) are : recount, narrative, procedure, report, account, explanation, and exposition, all serving different purposes and having different linguistic features (see Table 2 5 ). Martin (1989) put seven prototypical school based genres: recount, narrative, procedure, report, account,
60 explanation, and exposition into three categories based on the purposes they serve: personal (to present personal experience, including recount, narrative), factual (to present factual information, including procedure, report, account), and analytical (to analyze and argue, including explanation and exposition). Within each category, as a genre advances, grammatical demands increase. For example, in analytical genres, exposition demands more academic register features than explanation does. Among three categories, students usually have much more experience in personal genres since most of the time in elementary years students read and write recounts and narratives When students advance into secondary schools, they have more frequent encounters with factual and analytical genres. For these students, the lack of familiarity with factual and analytical genres usually results in big challenges in content area learning. Report and arguments are the most advanced and demanding genres in the category of factual and analytical genres. Students encounter them almost everyday in secondary content ar ea classrooms. Students need to have good mastery of academic language to read and write successfully in these genres. Expository text is particularly important as the gate keeping milestone when students move from elementary to secondary levels. Report is to classify and describe by relating a set of facts. The register features include timeless verb, generic participants that are realized by technical noun phrases, and expanded noun phrases with embedded clauses and preposition phrases. Clause Themes usua lly focus on the abstract concept classified or described. Besides, report is usually written in a relative formal and objective way.
61 Argument, or exposition, proposes thesis, presents different views, and supports the thesis with evidences. This genre re lies more on expanded noun phrases and abstract nouns to create abstract participants and present complex argument. To create a flow of argument in a dense and hierarchical organizational structure, exposition draws on language resources such as nominaliza tion and embedding. To express attitudinal meaning, logical connectors and modal verbs need to be effectively used. Declarative mood and third person are used to realize impersonality. Nominalization changes actions into things, which make s the argument so und more objective and also contributes to the hierarchical structure. Academic register varies to different social context and different social purposes, topics, and situations it serves. Variability of academic language thus presents challenge s to stude nts: learning to understand these features and use them to create different types of texts in different school subjects. To gain control o ver academic register s secondary students are expected to learn these genres of schooling in order to read and write in various content areas. In addition, to function successfully in any content area in secondary school setting, students must be familiar with genres that are prevalent in this area. More importantly, students should be familiar with the grammatical and l exical re s our c es available so that they can construct texts that meet the expectations of the disciplines Developing Academic Language Across and Within Content Areas Overview of the Academic Language Instruction This section reviews research regarding ac ademic language demands across content areas, academic language instructional practices and teacher s' preparation and training in teaching academic language. A broad collection of materials with regard
62 to these areas was retrieved from professional journa ls, practitioner handbooks, and some online resources. The literature included has a clearly defined purpose, matches my goal of discussing academic language in the framework of SFL, and explores the pedagogic strategies to develop academic language in sec ondary students. Some studies included in this review may not employ a SFL theoretical framework. They were included because they emphasize the importance of lexical and grammatical features and their connection to academic functions. Although attempts wer e made to include mostly empirical studies, some noteworthy position papers were also included for this discussion. The body of recent research reviewed here has identified a number of features of academic language that can be made explicit to students, an d has demonstrated that students can learn to meet the linguistic expectations of academic contexts when being provided with appropriate instruction. Academic Language Features of Content Areas This subsection includes a discussion of literature regarding the academic language demands of different content areas, followed by instructional practices supporting the development of academic language. Different disciplines engage in different social pr actices. These differences are manifested in the ways language is used by disciplinary experts (Fang & Schleppegrell, 2008). Research suggests that different disciplines create, disseminate, and evaluate knowledge in diverse ways, and correspondingly, have their particular use of language (e.g ., Fang & Schleppegrell, 2008 ; Schleppegrell, 2004 ). Various characteristics of academic language in different content areas pose great challenges to secondary students (e.g., Coelho, 1982; Short, 1994).
63 In K 12 settin gs, the content areas in focus are language arts, science, math, and social studies. A number of attempts have been made to operationalize academic language for teaching, learning or test development purposes by discussing the language features in various reading materials such as textbook across core content areas (e.g. Bailey et al., 2007; Bruna, et al., 2007; Butler et al., 2004a; 2004b; Kidd, 1996). Some studies in this field focuses on a specif ic content area, such science and math These studies prese nt and analyze the kinds of language students need for successful school performance in various content areas Their findings contribute to the conceptualization of academic language and also provide concrete guidelines for teaching, learning, and assessme nt purposes. Science is one of the most challenging subject areas for secondary students. Language in this content area builds up specialized and abstract knowledge and involves translating or constructing common sense meaning into technical terms (Unswort h, 1999). Therefore, the science language is regarded as the best representation of academic language and receives the most attention from researchers (e.g., Gee, 2004a). In science, a variety of grammatical and lexical resources are used to formulate hypo theses, propose alternative solutions, describe, classify, use time and special relations, infer, interpret data, predict, generalize, and communicate findings (Chamot & O'Malley, 1986). The grammatical and lexical resources such as complex subjects, nomin alizations, passive main verbs, and complex embedding make science text s difficult to read for secondary students (Gee, 2008). When reading in science, students are challenged with a great deal of technical terms and non technical terms that have unique me anings in a scientific context. These terms are chained with pre
64 (adjective that modify the noun) and post modifiers (adjectives and embedded clauses that modify the noun). Students must also comprehend expanded noun phrase and nominalization to interpret a flow of information and scientific reasoning The discipline of history presents a very different discourse from science. History usually describes events in chronological order and explores cause effect relationship through classifying and interpretin g these events ( Coffin, 2006a, 2006b ; Schleppegrell, 200 5) H istorical texts can construct explanation and present points of view and present historical debates. Historical understanding involves building connections and interpretations as well as perspect ives among historical event, people, and movements. Oftentimes the attitudes of the authors are implied underneath the descriptions and explanations of historical events and actors I n many situations, the readers must interpret the attitude of author Th is interpretation positions readers in viewing history from some specific perspectives. Meanwhile, students are also expected to step back from their own interpretation and seek for a deeper understanding of issues (de Oliveira, 2011). When writing in the subject of history, students need to use academic language to achieve functions such as explaining, describing, defining, justifying, giving examples, sequencing, comparing, and evaluating (Short, 1994). Students are expected to demonstrate their knowledge through presentation of historical events in authoritative and structured ways (Schleppegrell, 2004). Students must be able to understand and display the cause effect relationship among events for explanation and interpretation Historical texts also uti lize a number of language patterns and linguistic features to construct historical meaning For instance, a variety of syntax types, including simple past, historical present, sequence words, active voice, temporal signals, and causative
65 signals are domina nt in histo ry texts and construct the time sequence and cause effect relationship (Short, 1994). Moreover, the pattern of information flow is usually construed by nominalization. Sequences and events are nominalized into noun groups so that more extended d escription, classification, and qualification can be unwrapped in the text. Furthermore, multiple embedded clauses, complex past tense forms, and extensive use of pronouns within an expository text structure are also important features (Chamot & O'Malley, 1994). As Schleppregrell (2009) concludes, noun groups that construe historical actors, verbs that indicate different types of processes, and prepositional phrases that refer to grammatical circumstances, are the fundamental features of historical texts. T he discourse of history will be elaborated further in the upcoming section. As for the discipline of mathematics, mathematics texts draw upon three semiotic systems to make meaning: language, mathematics symbolism, and visual display (O'Halloran, 2005). Ac curate reading and writing in mathematics is necessary because language plays an integral role as students engage in problem solving, participate in discourse practices around mathematical topics, and construct meaning during their lessons (e.g, Adams, 200 3; Spanos, Rhodes, Dale, & Crandall, 1988). The demands of mathematical language in secondary classrooms can be described at semantic and syntactic levels (Spanos et al., 1988). Examples of semantic features include technical vocabulary (e.g., coefficient ), ordinary vocabulary that has different meanings in math (e.g., square ), complex strings of words (e.g., least common denominator ), and synonymous words and phrases (e.g., add, plus, and combine ). In terms of syntactic features, comparatives structures (e.g., greater than/less than ), logical connectors (e.g.,
66 if then, given that ), reliance on the passive voice, and various uses of prepositions characterize mathematics language (Spanos et al ., 1988). Schleppegrell (2007) provides a summar y of key linguistic features in the mathematics register. These features of the classroom mathematics register include technical vocabulary, dense noun phrases, being' and having' verbs, conj unctions with t echnical meaning, and implicit logical relationships Schleppegrell (2007) also suggests that teachers must support students and teach these features explicitly so that they can move from interactional register to mathematic register. Finally, in literar y texts, students read various kinds of genres and the patterns they are expected to read vary as well. The literature particularly addresses aesthetic needs and language in literatures such as poetry, drama, and fiction is usually creative and artistic (L ukin, 2008). Although the language features are very diverse, cohesion, language conventions, rhetorical devices, and choices of words are usually emphasized in getting the meaning from literary texts. To summarize, academic language exhibits different fea tures across various content areas and thus poses diverse language demands on secondary students (Fang & Schleppegrell, 2008). The literature reviewed with respect to different academic language demands in different disciplines shows both subtle and distin ct differences across content areas. The literature emphasizes the needs for instructional practices that address both syntactic and semantic features specific to the discipline, teach discipline specific knowledge of language, and facilitate teachers' cla ssroom instruction to develop academic language
67 Writing in Academic Language In the previous section, the literature studies published materials such as textbooks and other reading materials. In terms of academic writing, the great majority of studies fo cus on the early years rather than the late childhood and adolescence (Christie & Derewianka, 2008). Few studies adopt a SFL framework to study the nature and development of academic language in students' writing. Several functional linguistic studies exa mine relative large samples of students' successful academic writing and explore the development of academic language and how adolescents employ grammatical choices to realize meaning in different contexts ( Christie & Derewianka, 2008; Perera, 1990). Accor ding to this body of research, adolescent writers use features such as more embedded and dependent clauses, longer sentences, an extended vocabulary, greater use of modals, and more uses of passive voice in their academic writing. In a large scale empirical study of Myhill (2008), the linguistic characteristics of writing in 13 and 15 year olds are one of the most important findings in adolescent writing. Myhill (2008) conducts detailed grammatical and discourse analysis of persona l narratives and arguments. Her analysis suggests that the texts written by adolescents have greater use of subordination and nonfinite clauses and shows more awareness to the audience. Christie and Derewianka (2008) take into consideration various content areas such as history and science and perform functional linguistic analysis on texts written by students of different ages. These texts are chosen by their teachers as the benchmarks of what are possible at each phase of writing development. In their boo k, one chapter is devoted to explore the linguistic resources required to provide explanations and develop arguments in the later years of schooling. According to the
68 linguistic analysis, adolescent students use a variety of linguistic resources for reason ing about causes, influences, outcomes, and consequences. Their texts are better organized and have a higher value of lexical density than young writers. General participants and specialized history vocabulary are more dominant in the writing. These studie s of developmental trajectories of academic writing focus mostly on benchmark essays and successful academic writing of adolescents and most of them take place in the Australian context. Nevertheless, they will serve as the background knowledge of my study so that we can contextualize adolescents' historical writing in the overall trajectory of academic writing development. More functional linguistic studies examine one to several writing samples by adolescents to understand students' ability of utilizing academic language features in their writing (e.g., Colombi, 2002; Schleppegrell & Colombi, 1997). Studies of this kind focus mostly on bilingual writers. They analyze discourse organizational patterns, clause combining strategies, grammatical metaphor, an d a variety of academic language features and how successfully these features realize meanings in content area learning. Some of these studies analyze students' uses of academic language features to examine the impact of academic language instruction (e.g. Huang, 2004; Spycher, 2007). Most of these studies do not examine the relationship between academic language use and reading ability and writing quality. One interesting study of McNamara, Crossley, and McCarthy (2010) explores how sophisticate uses of a cademic language features can predict writing quality. Specifically the study examines linguistic differences related to cohesion and linguistic sophistication between high and low proficiency writers, as indicated by their score on
69 an argumentative essay The participants are undergraduate students and 120 essays are collected in this study. The authors examine features such as lexical diversity, word frequency, and syntactic complexity. According to the findings, t hese three features are also the three m ost predictive fea tures of essay quality. The authors state that more skilled writers have greater working memory capacity to access and use less familiar words as well as more complex syntax in their writing. On the other hand, the cohesion is not predict ive of essay quality. The authors explain that, in this specific context, the raters do not seem need cohesion cues to understand students' expression. Overall the results of this study indicate that more skilled writers use more sophisticated language. Ho wever, some of the textual features of good student writing (e.g. cohesion) may not be the same as those features that are considered to be facilitative for reading. This study raises more research interest toward the connection between use of academic lan guage features and reading ability a well as writing quality. Studies of academic writing focus on the features secondary students must know to construct content area meaning at different phases. These studies offer a context for us to understand academic writing development. It is interesting and meaningful to know the connection between adolescents' actual abilities of using academic language features and the connection between these uses and reading/writing abilities. Instructional Practices to Develop Academic Language As defined earlier, in secondary school sett ing, developing academic language is a process of developing essential "multiple, dynamic, inner related competencies" and building knowledge of linguistic, psychological, social, and cultural components with the goal to accomplish a variety of academic ta sks in school context (Scarcella, 2003, p.3). Different instructional practices are based on different conceptualizations of academic
70 language. Many emphasize the importance of lexical component (i.e., academic vocabulary ) and other s pay more attention to grammatical component (i.e., grammatical features ) interwoven wi th higher order thinking skills and language functions (Scarcella, 2003). Most of the researchers view academic language as a type of challenging language specifically for ELLs, and they propo se instructional practices to address ELLs' needs (e.g.,Scarcella, 2003; Zwier, 2006) Below I will present some of these instructional practices. Academic vocabulary Academic vocabular y include s general academic vocabulary words that are used across dif ferent content areas, technical vocabularies that are specific to academic disciplines, non technical academic vocabulary words that are used across content areas, and multisyllabic words (e.g. Zwiers, 2008; Scarcella, 2003; Coxhead, 2000). Because of its essential role for secondary students, academic vocabulary has traditionally been the focus of instructional practices to develop academic language. The literature on academic vocabulary instruction has several common emphases, including ways of identifyi ng and classifying essential words (e.g., Cox, 2003), principles of effective academic vocabulary instruction (e.g., Adams, 2003), and explicit word study and practice across contexts (e.g., CalderÂ—n, 2007). For example, CalderÂ—n (2007) describes results o f a six month academic vocabulary intervention for 300 bilingual (English Spanish) third graders across eight elementary schools in El Paso, Texas. Everyday students receive 90 minutes of reading instruction. In this instruction, 30 minutes are devoted to academic vocabulary: pronunciation, meanings, and English Spanish cognates. The vocabulary taught is from school texts, and also classified based on a process adapted from the three tier model mentioned before.
71 According to CalderÂ—n (2007), the interventio n has a positive effect on academic vocabulary. Missing from the academic vocabulary literature are findings that demonstrate how secondary students cope with academic vocabulary in content areas and how effective academic vocabulary instruction is impleme nted and sustained in school settings. Moreover, a tendency in the literature is that academic vocabulary is equated with academic language. Despite of the significance of academic vocabulary, a mere focus on academic vocabulary is problematic, as vocabula ry acquisition cannot provide enough support for academic language development. This shed light on the importance of linguistic features and discourse structures in learning academic discipl ines ( Bruna et al, 2007 ; Schleppegrell, 2001, 2004 ). Bruna et al. (2007) present a cautionary example regarding the consequence of academic language instruction that is driven by a simplistic approach and focuses on vocabulary. Their exploratory study examines explicit academic language instruction in science classrooms and how a teacher's knowledge of academic language affects her instruction and students' opportunities for learning. The teacher is interviewed about the linguistic elements of her science lessons. H er ninth g rade all ELL science classroom instruction is o bserved over a four month period. Results reveal that this teacher's conceptualizations of academic language revolve mainly around academic vocabulary. In this classroom, academic language instruction is more like a vocabulary based instruction that ignore s important linguistic features, such as the unique grammatical and discursive patterns embedded in the genres of science. The author argues that, contrary to the goals of developing academic
72 language, this type of academic language instruction impedes stu dents' opportunities to talk and think like scientists. In this study, academic language teaching could not lead to students' real progress in academic language development because the classroom teachers' conceptualization of academic language only focuse s on academic v ocabulary Students could not get sufficient support in other components of academic language development. Therefore, w hereas academic vocabulary is necessary for learning in any content area, academic language development is dependent on th e development of all components, especially lexical and grammatical features and higher order thinking skills (e.g., Chamot & O'Malley, 1994 ; Kidd, 1996; ValdÂŽs, 2006 ). Other l inguistic features Linguistic features beyond academic vocabulary, pose enormou s challenge to secondary students. As we discussed before, a high degree of academic ness is realized through elaboration of noun phrases (Coffin, 2005) conjunctions (Spycher, 2007), different type s of noun phrases (Fang, 2008a), and clause structuring st rategies including nominalization and embedding (Colombi, 2002; Schleppregrell, 2001). These grammatical features realize higher order thinking skills and language functions in academic register s (Zwiers, 2007). In this section, I will review a body of res earch literature that advocates explicit teaching of linguistic features in a meaningful learning environment (e.g., Bailey et al., 200 7; Warschauer, Grant, Real, & Rousseau, 2004). Explicit teaching of linguistic features is supported by studies regarding the cultural codes that must be mastered for ful l participat ion in a Discourse or a culture (Delpit, 1998). Cultural codes are Language patterns and linguistic features that many students will never get without being explicitly taught (Delpit, 1998). Students must be
73 acculturated and apprenticed into academic Discours e and academic ways of communicating (Bartolom ÂŽ 1998 ; Gee, 1996). When students have little capital in and familiarity with the valued ways of using language, explicit and direct teaching of language demands is strongly r ecommend ed (Bartolom ÂŽ 1998 ). As B artolom ÂŽ (1998) points out, direct teaching "linguistic messages explicitly and precisely" is particularly needed for many students who lack experience of sharing knowledge with distant audience and do not see the need of producing clear and explicit acade mic texts (p.66). D ifferences between literacy requirements in primary and secondary levels are another reason that academic language features must be taught explicitly Strong early reading skills do not automatically develop into more complex skills tha t enable students to process specialized and sophisticated reading of literature, science, history, and mathematics (Shan a han & Shanahan, 2008). Narrative is the dominant focus in American elementary school settings. Many students are unfamiliar with the e xpository nature of academic language they encounter in secondary school content area learning. This unfamiliarity, demonstrated by functional linguistic analyses of student writing, means that students in secondary schools and even colleges often lack und erstanding of expected language use in performing given academic tasks (Schleppegrell, 1996; Schleppegrell, 2003). Therefore, instructional practices must explicitly teach students academic language features such as sophisticated genres, specialized langua ge conventions, disciplinary norms of precision and accuracy, and higher level interpretive processes (Schleppegrell, 2001). Literature supports the positive impact of explicit teaching of language features on academic language development (e.g., Aguirre M uÂ–oz el al, 2006a, 2006b;
74 Echevarria, S hort, & Powers, 2006 ; Hammond, 2006; Huang, 2004; Schleppegrell & de Oliveira, 2006; Spycher, 2007 ). These studies usually target features of academic language in a specific content area such as history and science. T hey report how the explicit instruction of these features influe nces students' (mostly ELLs') language and literacy dev elopment. For example, Spycher (2007) studies how one secondary ELL handles the challenges in academic writing tasks. The participant is a student who is learning the language and literacy skills in an English Development Class (ELD). This study uses sentence analysis organizer to examine the effects of instructional strategies that teach linguistic features explicitly. The student's two dr afts of one topic are analyzed in terms of three linguistic features: authoritative stance, conjunction, and references. Results suggest that student's first draft is written in an everyday style and does not meet the expectations of academic register. How ever, in the second draft, the student produces text that increasingly incorporates the linguistic features of academic language. This progress suggests that explicit instruction on the expectations of academic writing promotes students' academic language development. A nother study focuses on specific science content and the construction of academic writing in school science (Huang, 2004). The instructor teaches academic features of scientific language from the perspective of SFL in five weeks. Over these five weeks, the students write multiple drafts on several topics, and the author collects field notes, lesson plans, students' writing on classifi cation of matter, and discourse data from teacher student interaction. Results suggest the improvement of sub sequent drafts particularly in the use of reference and nominalization and the number of terms explained and exemplified. The author concludes that explicit instruction in particular
75 grammatical resources used in science writing plays an important role in this development. In these studies, we can see that a number of factors including content and language instruction are usually interwoven with each other. It is hard to disentangle the impact of explicit instruction as an independent variable from all othe r variables. As an exception, Echevarria et al. (2006) include a control group in their study of academic writing. In this study, the authors report the positive effect of the Sheltered Instruction Observational Protocol (SIOP) on middle school ELLs' expos itory writing. SIOP is a sheltered approach to language instruction that combines both content area learning strategies and ELL teaching strategies on ELL's academic language development. The study uses a quasi experimental design. One major difference bet ween the control and experimental group is that students in the experimental group receive explicit teaching of language objectives and linguistic features whereas students in control group receives nothing in this regard. In this model, at the beginning o f each lesson, the specific language objectives would be taught explicitly. The instructors teach about content specific vocabulary and text structures. The researchers collect and assess the pre and post test writing samples using the writing assessment from the Illinois Measurement of Annual Growth in English. Results reveal that the group taught using the SIOP model shows significantly better gains in writing than does the comparison group. Based on the findings, authors suggest that a consistent and sy stematic implementation of the features in instruction found in the SIOP model is effective in improving ELL's expository writing. This study offers more persuasive results with respect to the positive impact of explicit instruction of academic language fe atures.
76 Explicit functional grammar instruction has been criticized by a number of researchers (e.g., Bunch, 2006; GutiÂŽrrez 199) For example, Bunch (2006) claims that, given the wide range of ways in which language is used in academic settings, what ca n be taught explicitly in classrooms is quite limited. Moreover, teachers' explicit attention to and particular emphasis on form might lead to an artificial learning environment and students' misunderstanding that form is the substance of instruction itsel f. Moreover, Bunch (2006) emphasizes that learning academic language is gaining a membership in academic discourse communities. Explicit focus on language is helpful, but not enough. Students need to use and participate in "a community of discourse" rather than only learn through "an abstract study of language" ( GutiÂŽrrez 1995, p. 34). This type of critique result s from the misinterpretation of academic language features and how to teach them explicitly. Explicitly teaching of academic language is not teaching decon textualized and static grammar Neither is the focus on form and linguistic features in isolation (Sch leppegrell, 200 4 ). In explicit instruction, students are engaged into analyzing language features and understanding the functions of these features in disciplinary meaning making. Therefore, this is not an abstract study of language. Rather, this is "recog nition of the socio cultural and discursive bases of knowledge and learning" (Coffin, 2006 b p. 424). Learning linguistic features is part of the process of obtaining cultural codes to enter the community of academic Discourse (Delpit, 1998). Overall, the positive impact of explicit teaching of academic language has been suggested in the literature (e.g., Gibbons, 2003 ; Scarcella, 2003 ). In reality, the language features, the demands, the expectation posed by academic texts and school
77 discourse is usually implicit to students (de Oliveira, 2011) This is a particularly unfair reality for language and cultural minority students. Therefore, it is crucial to raise students' awareness of important components of academic language including lexical and grammatica l features and their functions. Students must know that academic language uses these components differently compared to everyday language. Students should understand the expectation of different disciplines and be engaged in actively analyzing language fea tures of these disciplines. Students must understand how resources of linguist functions can be employed to construct meaning in disciplinary communities. All of this must be achieved through explicit academic language instruction. Training Teachers to T each Academic Language Features Effective classroom instruction, as the most important factor in developing adolescents' academic language, is contingent on the quality of teachers (Campbell & Kmiecik, 2004; Shanahan & Shanahan, 2008). The studies mentioned previously mostly focus on the students' side. Here I will review a nother body of literature that pays more attention to teachers' knowledge and practice. Teachers' knowledge and practice of academic language In ter ms of academic language, a quality instruction is dependent on, among other things, teachers' knowledge of academic language demands and linguistic features specific to different content areas (Colombi & Schleppegrell, 2002; Schleppegrell, 2001, 2004). Tea chers should possess sufficient linguistic knowledge to conduct language development activity in content area discourse to help students in academic reading and writ ing ( Schleppgrell, 2006 ; Spycher, 2007 ). In order to help students fully develop academic l anguage and content knowledge, teachers should be
78 capable of making the linguistic expectations explicit to students and teaching the linguistic elements that are characteristic of and valued by academic registe rs (Aguirre MuÂ–oz el al, 2006a; Zwiers, 2007) Unfortunately, although content area teachers are usually experts in their disciplinary communities, many of them are not aware of the highly demanding nature of academic texts in their respective content area (Scarcella, 2003). Content area teachers ar e busy teaching their content and thus spend little time engaging students in language development activities beyond learning academic vocabulary words (Bruna et al., 2007). Researchers have noticed that many preservice teachers hold recalcitrant attitude toward teaching academic writing, although they are required to take a content area literacy course as a part of certification (Draper, 2002). Classroom observation and teacher interviews have been conducted to examine teachers' knowledge and practice reg arding academic language development (e.g. Bailey et al., 2007 ; Br una et al., 2007; Spycher, 2007 ). Solomon and Rhodes (1996) provide survey results of 157 ESL teachers' perspective on academic language Results suggest that the respondents (elementary and secondary ESL teachers) view academic language in terms of discrete aspects of language such as vocabulary or grammar. Some of them focus on functions of language such as comparing and contrasting, catego rizing, and sequencing events. This lack of knowled ge has not changed qualitatively and is still reflected in teachers' classroom practice (e.g. Bailey et al., 2007; Bruna et al., 2007). The descriptive study by Bailey et al. (2007) examines how teachers use oral language to support students' academic lang uage in fourth and fifth grade mainstream science classrooms. The language functions used by teachers and the
79 academic vocabulary growth of students are observed over several weeks across several different science lessons taught by teachers. Although some evidences of all language functions such as explanation, description, comparison, and assessment can be found in the classrooms, teachers use many academic vocabulary words without any substantial support. These teachers do not make full use of the opport unities to teach about academic vocabulary and language functions. Zwiers (2007 ) reports similar finding s in his study. He investigates the ways in which mainstream content area teachers teach academic language to non mainstream students. The author obser ves three teachers' classes approximately two days per week in four months and focuses on language events of teacher and student with respect to academic language features and functions. The teachers' practice including their explicit teaching of academic language, modeling and scaffolding, and less explicit uses of academic language by teachers and students are all recorded during the observation. A measurement of linguistic features of academic language is conducted to examine the "academic ness" of stude nts' utterances and both positive and negative factors in teacher' practices are identified. The occurrence frequency of academic language features such as nominalization, passive voice, complex sentences, cohesion, and coherence are calculated to measure the level of "academic language ness" in particular utterances taken from the classroom texts and student writing samples. The researcher analyzes the logbook and essays of students to track the change in thinking processes and academic expressions. Result s reveal that students make evident progress in the language of identifying cause and effect, taking other perspectives, and comparison but not the language of bias and application of history. The authors explain
80 that, when teachers understand the cognitive skill of their discipline and the language that support these skills students can be supported in their academic language development. When teachers do not have those skills or do not design activities that focus on those thinking skills and lan guage patterns, the instructional practice will impede students' academic language development. Studies have also looked into how teacher s take the roles of modeling (demonstrate the accurate use of academic language features) and scaffolding (explain and solve the challenges that academic language features pose) to develop academic language and learning content in content areas (Gibbons, 2003 ; Jacob, Rottern Berg, Patrick, & Wheeler, 1996). To read and write successfully in content areas, students should think and use language in the ways disciplinary experts do (Schleppgrell, 2006; Zwier s 2008). Content area teachers, as disciplinary experts, must model the ways of using disciplinary language and teach the specialized ways of thinking and using language via teacher student interaction. Gibbons (2003) study examines factors in classroom discourse that enables (or constrains) language development. The role of teacher student interaction in academic language development is also explored. This study is bas ed upon the constructs of mediation of sociocultural theory and mode continuum of SFL. The author investigates how teacher student talk in a content based (science) ESL classroom contributes to learners' language development. Data sources include audio rec ordings and transcriptions of 14 hours of discourse; environmental print around the classroom, such as posters, charts, and children's work, field notes, and interviews with teachers and students. Data analysis suggests a number of strategies that teachers may use to build
81 explicit linguistic bridges between two discourses (academic Discourse and everyday interactional Discourse). These strategies include recasting, signaling to the students how they can self reformulate, indicating where a reformulation is needed but handing this task over to the students, and modeling alternative ways of recontextualizing personal knowledge. Through linguistic and discourse choices that teachers make and discussion of relationship among language, meaning, and context that students are engaged in, students are able to successfully shift along the mode continuum from the students' current linguistic levels in English and their commonsense understandings of science to the educational discourse and specialist understanding Thi s study confirms further the important role of teacher's practice. It also suggests the value of teacher student interaction and explicit teaching of mode shifts from everyday language to science discourse in developing academic language. In brief teacher s' knowledge of academic language demands and linguistic features can have positive influence on students' academic language development When teachers are capable of making the linguistic expectations explicit to students and teaching the linguistic eleme nts that are characteristic of an d valued by academic registers, students' academic language can be successfully developed Professional development S cholars advocate preservice and inservice professional development as a key to provide teachers with high cognitive skills, disciplinary knowledge, and language knowledge so that their expertise in teaching academic language can be established (e.g., Gibbons, 2003 ; Lesley et al., 2007 ). A few studies report successful transfer from teacher training programs o n teachers' readiness and students' achievement in
82 academic language development (e.g., Aguirre MuÂ–oz el al, 2006a, 2006b ; Schleppegrell & de Oliveira, 2006 ; Walker & Bean, 2004). Research suggests that professional development activities have positive in fluence on teachers' attitude and knowledge as in the successful case presented by S chleppegrell and de Oliveira (2006) This study examines how teachers' knowledge of functional grammar influences instruction in a positive way. In this study, participant s are history teachers attending workshops in summer institute. They use SFL tools to analyze passages from history textbooks. Teachers receive help from linguistic specialists to understand the challenges of history in linguistic terms and deconstruct the meaning in history using these linguistic terms. The linguistic constructs discussed in the study include sentence constituents, the meaning relationships between the parts of a sentence, complex noun phrases and the multiple meanings they present, time m arkers and connectors that structure a text, and the use of reference that build s cohesion. Teachers' reports and the authors' observations suggest that these history teachers are able to use the kind of linguistic analysis to engage students in talking ab out the historical events and participants presented in the texts. The results of these teacher training activities also show positive impact on students' learning outcomes. In the discipline of language arts, Aguirre MuÂ–oz el al (2006a) uses a variety of assessment tools to investigate how teachers' training in SFL can support teachers' teaching and students' academic language development. Based on the perception that direct instruction on linguistic structures of academic language is usually not a compon ent of classroom instruction, the authors present a four day teacher training program on instructional strategies to incorporate functional grammar in
83 classrooms. A two day follow up comprised of four modules is to ensure the implementation in classrooms. Thirty two language art teachers participate in the study, among which 21 are in the trained group and 12 are in the comparison group. A research approach incorporating both quantitative analysis and discourse analysis is adopted in this study. Teachers' s urvey, interview, and observation data indicate that, before this training program, teachers in general are unable to adequately expose students to functional grammar concepts. After the training program, however, two groups are differentiated, particularl y in the writing instruction. T rained teachers offer instruction more directly and more specific to the needs of the ELLs. To the contrary, comparison teachers focus more on content and ideas, and on a broad and superficial level of writing instruction suc h as an overall essay structure. This differentiation in instruction also has an impact on students' learning outcome. In this study, the Language Arts Performance Assignment (LAPA) scores are used to examine students' writing samples as the outcome variab le. Qualitative differences are found in the level of functional grammar implementation between the trained and comparison groups. Students in classes with trained teachers have higher performance on LAPA than students in the classrooms with low implementa tion of functional grammar concepts. Furthermore, results from statistical analysis show that the level of functional grammar implementation is consistently the most important variable in predicting student performance on all four LAPA scores. It is more i mportant than other variables such as L2 processing strategies, teachers' experience and content expertise, or feedback assessments.
84 Studies of this kind show the positive impact that professional development can have on students' academic language development. The professional development activities must enable teachers to identif y discuss, and appl y a variety of academic language features that are specific to different content areas. A successful professiona l development program must prepare teachers for teaching academic language in secondary content area classrooms. Developing Academic Language in History Classrooms Many studies are specifically about the content area of history which is the focus in the current study In this section, I will explore studies that illustrate the high literacy demands of learning materials and practices in history classrooms, followed by the discussion of students' competencies in the discourse of history. A variety of acade mic language features that are prevalent in the content ar ea of history will be identified. A positive impact of instructional practices that teach these features will also be reported. Language demands in the discourse of history In the discipline of history, the interaction between the reader and the text is always important. Historical meaning often is hidden in the text of history (Wineburg, 2001). To read and write in history, the authors' motiv ation, rationale, and goals are involved and must be i nterpreted. As in all content areas, learning history means learning the language of history. However, compared with science, history receives much less attention despite its importance in American schooling (de Oliveira, 2011) The ability to read and wr ite is essential to academic success in school history. Paxton (1999) points out the challenge of language of history in reading and writing for secondary students and suggest s that a textbook centered history classroom is dominant in many secondary histor y classrooms Students must be able to read the
85 reading materials and access the language used by historians in this subject area. In historical texts there are a variety of academic language features that present challenges to students (Achugar & Schlepp egrell, 2004; Schleppegrell, 2004). Short (1994) an alyzes the reading demands that middle school English learners encounter in history curricula. According to her findings, students must cope with long expository passages filled with abstract concepts (e.g ., liberty, propaganda ) to engage in classroom tasks. Researchers using SFL and discourse analysis have also made major contributions to the understanding of the historical discourses and the language patterns common in school history texts (e.g., Coffin, 2006a; 2006b; Martin, 2002; Unsworth, 1999). For example, Coffin (2006a) argue s that history has distinctive textual forms that need to be learned by history students. In her book, a systematic and comprehensive analysis of historical discourse introduces how language of history represents the motifs of time, causality, and evaluation effectively through the lexical and grammatical choices. Time is a successive movement through space or continuous cycles. Each time movement has its own beginning, peak, an d ending, which plays a significant role in historical discourse. To Coffin (2006a), macro time constructs such as calendar time, chronology and their relation to historical narrative are a determining point of understanding historical perspective. Time is not only a logical system but also can function to make experiential, interpersonal and textual meaning. Coffin analyzes a range of grammatical and lexical resources for represent ing concepts such as the sequencing and segmenting of time. These linguistic choices include: h ypotactic temporal conjunction ( conjunctions of hypotactic clauses that describes temporal relationship. e.g., when, before )
86 c ircumstance of time ( when an event take places. e.g., last year ) c onjunctive a djunct (he temporal resources th at construe the beginning, continuation, and end phases of an event or activity e.g., then, first, finally ) s pecialized abstract technical terms (e.g., the Gold Rushes, the Second World War, the Cultural Revolution ) t emporal verbs or temporal circumstantial ( verbs in any clauses that construe temporal relationship between two events. e.g., ensued ) r elational clause ( clauses that describe the temporal condition of an event. e.g., when I was seven, after my sister was born ) Another core concept of history is causality. In history, historians do not just tell a chronology of events. Instead they use these events to explain how one thing leads to another. Causality is particularly important to students' understanding of historical tex t (Coffin, 2004; Achugar & Schleppegrell, 2005). Causality involves more significant connections than those of the temporal framework. In order to read and write successful history texts, students need to determine the causal effect from various events, or ganize these causal phenomena rhetorically, and assign different significance to different events (Coffin, 2006a). According to Coffin (2006a), there are four functions of causality that linguistic choices achieve in historical text. They can enable and de termine causation by using verbs such as enable, influence, result in or make Cause can also be packaged as a noun or nominalization by using result effect or reason to realize abstract causation. In brief, causality in historical d iscourse is represent ed through a variety of linguistic choices including: circumstance of cause (e.g., because of, as a result of, due to ), verbs (e.g., resulted, led to ), conjunctions (e.g., because ), conjunctive adjunct (e.g., therefore, thus ), and nominalizations (e.g. effect, reason ).
87 Coffin (2006a) also delineates an appraisal framework in terms of graduation of intensity, attitude, and engagement. In historical texts, the presence of agency is usually minimized and the real historical actors are given least promin ence so that the reader would not recognize them. Nevertheless, events and so called facts in all historical texts are interpreted through the authors' perspective. Even though it is important to recognize the author's stance and negotiate alternative judg ments and opposing perspectives within history, this appraisal framework is a suitable basis for pedagogical intervention to develop students' ability to read and write in history. To sum up, s tudents must be able to understand and use generalized, abstract nouns and specialized lexis in history Students should show diminished reliance on chronology and embedding events as part of explaining or arguing. Moreover, students must use evaluative lexis as a means of appraising historical processes in ter ms of their historical significance. These linguistic choices work together to construct the concept of time, causality, and appraisal in historical writing. Despite the important role of linguistic features around time, causality and appraisal, teachers i n history classrooms do not always articulate these linguistic features ( de Oliveira, 2011; Schleppegrell, 2005 ). This calls attention to why and how teachers can help students to focus on language to understand content or express their historical understa nding. Students must develop control on both form and function of texts to be successful in historical reading and writing. Developing academic language in history In the previous section, I discussed how to develop secondary students' academic language in different content areas. Among them, a number of studies address the content area of history specifically (e.g., Coffin, 2006a, 2006b;
88 Schleppegrell, 2005 ; Short, 1994; Zwiers, 2006 ). Studies have acknowledged the wide range of language demands including various genres and different language features as well as language learning opportunities in history classroom ( Bunch 2006; Schleppegrell, 2005). Also history teachers usually do not explain and instruct about these language resources before ass igning stu dents writing tasks in history classrooms (de Oliveira, 2011; Schleppegrell, 2005). Zwiers (2006)'s study explores how to scaffold the development of academic language, disciplinary thinking, and content learning for non native English speakers in middle school history class. Participants are 60 ELLs at early intermediate or intermediate proficiency levels of English attending a five week history based English language summer program. The teaching approach focuses on six dimensions of historical thinking: background knowledge, cause, effect, bias, empathy, and application by examining professional models of persuasive essays in history. The persuasive features and mortar vocabulary (general academic vocabulary) and brick vocabulary (content specific vocabul ary) in the essay are the focus in the teaching approach as well The data resources include the persuasive essays assigned by the teacher, audio recording of lessons, and student logbooks. The analysis shows that scaffolding in higher order thinking and l anguage features leads to students' development in cognitive and communication skills. This finding suggests that in order to teach students to learn to think, read, write, and talk about history like a historian, historical language needs to be modeled, s caffold, and practiced in the ways that historians think. A nother study by Coffin (2006b) uses a pre and post test research design to evaluate the impact of teaching learning cycle' on students' writing. This approach
89 encourages students to consciously reflect on how language functions in creating historical meaning. In this approach, students are engaged in three phases. These phases are: deconstruction phase (students are introduced to model historical text of a specific genre through a range of activi ties and teacher input. After that they would explore the grammar and lexis in text such as nominalization), joint construction phase (students and teachers jointly negotiate and publicly write up texts representing the same genre), and independent constru ction phase (student work individually or in small groups to rework on the target genre). Selected students' texts written before and after the implementation of the cycle are analyzed. Results demonstrate the progress in students' control of text organiza tion. A more purposeful organization and a clearer text structure could be seen in post intervention texts, whereas improvement at grammatical features is not that evident. The study indicates that teacher training have successful influence on students' writing development and "teaching learning cycle," a pedagogic strategy can encourage conscious refection on how language functions to create historical meaning. The study of de Oliveira (2011 ) investigates the language resources used by eig hth and eleventh grade students when writing an exposition in school history. Using a qualitative methodology, the study analyzes questionnaire data from 44 history teachers, interviews with four focus history teachers, and essay data from these four teach ers' classes. Sixty three strong essays and 26 weak essays are selected and rated by students' teachers. T hese essays are analyzed and the linguistic patterns that realize the language of history are identified. The language resources that successful 8 th g rade writers use include thematic choices to signal the organizational structure of texts and
90 cohesive resources such as pronouns and demonstratives to establish relationships between elements of discourse. T he most distinguishing difference between succes sful essays and essays is in the way elaboration is used. Elaborating relationships are used to reiterate the main semantic notions presented in the texts. Less successful students use little of these elaborating resources Also the r esources used at the e ighth grade level are amplified at the eleventh grade level. T his study shows that construing evaluative meanings is especially relevant for writing an expository in history. Elaboration, or the way examples, clarifications, additional details, and explana tions are provided, is the most distinguishing feature between essays considered "strong" and "weak." The findings also suggest that teachers have expectations on the way in which historical information is presented and developed. All these studies examine students' academic language in the content area of history. These are mostly small scale and qualitative studies. The foci are the language features of historical discourse including evaluation, time, and causality. These studies try to raise the awarenes s of teachers and students to how language resource constructs knowledge in history. The results also suggest that appropriate instructional practices have the potential of preparing students for reading and writing in historical discourse. Gaps in the Li terature Overall, t he growing literature on academic language ranges from linguistic analysis of written and spoken texts to descriptive studies of classroom practices. Three theoretical frameworks described in this study are : Cummins's initial cognitive a pproach of BICS and CALP, Gee's sociocultural perspective that illuminates the nature of academic language as the cultural practice of various disciplinary communities and
91 linguistic perspective of SFL that provides linguistic analysis tools for connectin g academic features to their functions and context. Various studies are grounded in different frameworks and a range of conceptualizations to explore the instructional practices that address the challenge of academic language in secondary students. Adoptin g SFL, academic language is defined as a variety of language resources that realize more formal and abstract styles and construe academic register in different disciplinary communities. Nevertheless, a number of gaps exist in the current body of research. First of all, research literature with respect to academic language has been growing, but much has been written for ELLs, within the field of S econd L anguage A cquisition and T eaching E nglish to S peakers of Other L anguage Academic language presents challenge s to all students, not just ELLs. Recent reports (e.g., Berman & Biancarosa, 2005) have found that adolescents (both mainstream students and ELLs) are not able to read and write the specialized texts of secondary schooling These reports provided striking evidence of students' limited academic language proficiency in academic reading and writing. Although mainstream students may encounter tremendous challenges in developing academic language to function successfully in content area classrooms, this field is highly underdeveloped and thus must be addressed. Therefore, a cademic language should be a concern for educators of students from a broad ra nge of backgrounds and from content areas such as science, mathematics, social studies, and language arts. Second, the need for my study is established by the fact that current research focuses on college level learners. C urrent research does not pay much attention to adolescent learners (i.e., secondary schools) A n extensive literature focuses on English
92 f or special purposes and university learners rather than adolescent students and the demands of secondary schooling ( e.g., Adamson, 1990) a lthough the demands of academic language are extensive for secondary students. T here is a lack of attention to stude nts' needs in developing academic language in secondary schooling Therefore, m ore attention must be paid to academic language use in secondary schooling. Third, current academic language research looks at academic language primarily from the perspective of vocabulary as if academic language consisted solely of vocabulary. Students must master academic vocabulary words in order to comprehend the concepts and display their acquisition of these concepts in any specific discipline ( Beck et al. 2002). As lex ical choice plays a critical role in academic language, researchers focus more on academic vocabulary instruction and believe in the positive effects of academic vocabulary intervention on academic language development (Snow, 2008; CalderÂ—n, 2007). However a cademic language is more complex than academic vocabulary only and should be examined using a more comprehensive approach. A mere focus on academic vocabulary will impede students' academic language learning. The nature of academic language including it s grammatical features and discourse structures must be identified and taught to develop students' academic language (Bruna, et al. 2007). The importance of these academic language features must be acknowledged to a full extent and also incorporated in th e instructional practice In addition, current academic language research focuses on published materials including state standards, textbooks, reading materials, and oral texts transcribed in teachers' instruction and teacher student interaction rather than students' actu al use of academic language The current research does not shed light on how academic
93 language u se relates to students' reading and writing ability Understanding students' actual ability of coping with academic language is criti cal to the design of effecti ve instruction and remediation. Based on students' actual capabilities in handling academic reading and writing, i t is also significant to know secondary students' actual use of academic language and how this use interacts with students' reading abilities. A number of studies that examine adolescents' academic writing focus more on successful benchmarks in students' essays and pay little attention to those students of different proficiency levels (Christie & Derewianka 2008) Al though we assume that students with higher reading abilities can handle academic language more successfully, current research is unable to reveal how students' use of grammatical and lexical resource differ across reading groups. A comprehensive analysis o f students' writing allows more insight into how students grapple with academic language demands and exploit features at semantics, syntax, and pragmatics levels to represent their learning in content area knowledge. Another gap of research is the relatio nship between this use of academic language features and overall writing qualities. In the current research, whereas some academic language feature(s) is/are usually analyzed as an indicator of students' academic language ability (e.g ., Coffin, 2006), we d o not know how academic language features affect the overall writing quality. Knowing the interrelationship between students' use of academic language features and their reading abilities and academic writing qualities will contribute to the ongoing discus sion regarding adolescent students' literacy development in content area learning. Therefore, this research gap must be addressed.
94 Last but not least, academic language research typically involves small scale studies (e.g., case studies of 1 5 students). Due to the variety and complexity of academic language features and discourse analysis researchers tend to employ descriptive studies in analyzing these texts. Most of the current studies are limited to case or small scale study, and thus cannot offer spe cific suggestions to effective research based interventions. Although these studies provide in depth information about the demands of academic language larger scale studies with larger group s of participants may provide a more comprehensive picture of ado lescents' academic language competence s. A method that enables a more convenient and practical analysis and interpretation of academic language features in a large size of writing sample s should be generated to better guide operationalization and practice in academic language. Moreover, quantitative studies are needed to study students' use of academic language in order to inform the direction and magnitude of students' academic language development. Overall, a lthough a theoretical gro undwork has been estab lished to conceptualize academic language the essential role of academic language features for secondary mainstream students has not been fully acknowledged in both research and instructional practice. The current literature should pay more attention to a ll linguistic features of academic language and also provide a more comprehensive and detailed picture of mainstream students' actual ability in coping with these features in content area learning. It is important to know the interrelationship between use of academic language features and students' reading abilities and writing qualities The answers to the questions above may help content area teachers recognize their students' strengths
95 and needs in academic language acquisition and also guide them to mod el and scaffold academic langua ge and content area instruction. In Chapter 2, the theoretical framework for the study was presented. In addition, a review of the relevant literature illustrated key theoretical features. Next, the literature on teaching and learning academic language was reviewed. In Chapter 3, the method to answer the research questions will be presented.
96 Table 2 1 Academic t ext and w ritten t ext Text (1) Text (2) Capsaicin causes a complex series of events to occur in the body. It triggers nerve terminals to release "substance P," a chemical that transmits the impulse interpreted by the brain as "Pain!" The stimulus also causes the brain to release endorphins, the body's natural painkillers. Endorphins produce a pleasant sensati on of well being, which may explain, in part, why people eat hot peppers even though the initial effect is burning. M: And uh, I I tried it with different colors, with both of them out, one came out, this one just came out blue, and I don't know, what th is color is. (p.116) M: But you have, first you have to stick it into the wax, and then water, and then keep doing that, until it gets to the size you want it. (p. 117) Table 2 2. Academic t exts and i nteractional t ext in clause l eve l Text (1) Text (2) (1a) Capsaicin causes a complex series of events to occur in the body (2a) And uh, I I tried it with different colors, with both of them out, (1b) It triggers nerve terminals to release "substance P," (2b) one came out, (1c) a chemical that transmits the impulse interpreted by the brain as "Pain!" (2c) this one just came out blue, (1d) The stimulus also causes the brain to release endorphins, the body's natural painkillers. (2d) and I don't know ( 1e) Endorphins produce a pleasant sensation of well being, (2e) what this color is (1f ) which may explain, in part, (2f) But you have, first you have to stick it into the wax, (1g) why people eat hot peppers (2g) and then [sticking it into] water, (1h) even though the initial effect is burning. (2h) and then keep doing that, (2i) until it gets to the size
97 (2j) (and) you want it. Table 2 3 Clause types Type Definition Example Main clause The only clause in a simple sentence, the initiating clause in a paratactic sequence, or the dominant clause in a hypotactic clause complex. Capsaicin causes a complex series of events to occur in the body. Hypotactic clause Hypotactic clauses are dependent on but not constituents of an other clause. They are traditionally called: subordinate clauses that participate in discourse structuring. why people eat hot peppers Paratactic clause Paratactic clauses are linked to the main clause with a coordinating conjunction or merely juxtaposed. Direct quotations are also included. Interactional language relies heavily on paratactic clauses to link from clause to clause. and then [sticking it into] water, Embedded clause Embedded clauses are constituents of a larger clause in which they are embedded. a chemical [that transmits the impulse interpreted by the brain as "Pain!"] Table 2 4. Register f eatures of a cademic l anguage Situational Expectations (Context) Grammatical and Lexical Features (Register) Display Knowledge in a dense and abstract way Experiential Features S pecialized, and technical, and abstract nouns Long and Complex Nouns Nominalization Lexical d ensity (Overarching feature of academic language) Be authoritative Interpersonal Features Declarative mood structure and the use of Thematic patterns to realize authoritativeness and distance Structure text in expected ways Textual Features Clause combining strategies Methods of Thematic development
98 Table 2 5. Genres of s chooling and t heir l anguage f eatures Genre Purpose Register Features Recount Retells a sequence of events, based on personal experience. Personal pronouns; additive and temporal conjunctions; past tense; mainly action verbs (doing verbs). Narrative Reports and evaluates problematic events and outcomes. Variety of verb tenses; embedded clauses; expand noun phrases; variety of verbs (including doing verbs, thinking verbs, being verbs, and etc.). Procedure Reports a sequence of events with general participants. Timeless; simple present tense; action verbs (doing verbs). Report Organizes and classifies information by relating a set of facts. Timeless verb; generic participants that are realized by technical noun phrases; and expanded noun phrases with embedded clauses and prepositional phrases, being verbs. Account Tells why things happened in a sequence. Nominalization of events; relational verbs realizing causal relationships. Explanation Explains and interprets a phenomen on. Logical organization; timeless verb; expanded noun phrases; being verbs; variety of clause Themes. Exposition Proposes thesis, presents different views, and supports the thesis with evidences Expanded, generalized and abstract noun phrases, nominalization; logical connectors and modal verbs; declarative mood and third person; Makers of contrast, classification and condensation. Note: This table is adapted from the Table of Some Genres of Schooling (Schleppegrell, 2004, p. 85).
99 CHAPTER 3 METHODOLOGY The purpose of Chapter 3 is to describe the methods used in the collection and analysis of data. Specifically, I describe r esearch site, data collection procedures, and data analysis methods including all the variables a nd statistical analysis Quantitative method s and linguistic analysis were employed based on the research questions of the present study. The goal was to exam ine adolescents' use of academic language in historical writing and determine whether such use is in any way associated with their reading ability and wri ting quality. Specifically, my study sought to address the following research questions: What academic language features can be observed in 9th grade students' historical writing and what functions do these features serve in historical meaning making? Does academic language use differ according to students' reading ability? Does academic language use diffe r according to the quality of students' historical writing? That is, to what extent does the use of academic language predict the overall quality of students' historical writing? R esearch S ite In this study, a local high school BG was selected as the research site. BG is a public high school located in a north central Florida town. It was the largest public high school in the town and served over 2600 students from 9th to 12th grades. In 2006, B G had 24 student s for every full time equivalent teacher, while the Florida average was 16 student s per full time equivalent teacher. In 2007, the percentage of White, Black, Hispanic, Asian American, multiracial, and Native American were, respectively, 64 %, 21%, 8%, 5%, 3%, and 1%. Of all the students, 52% were male. Twenty percent were eligible for free or reduced price lunch, which was much lower than the 45% of the state average student economic level. Moreover, 14% of the school population was identif ied
100 as talented or gifted, whereas 13% of the school population participated in special education programs BG high school consistently received high marks by state and federal standards. It received an A' rating for th ree consecutive years. In 2009, New sweek ranked BG among the top 5% of high schools in the nation. On the 2009 FCAT, 64% of BG high school ninth graders met or exceeded standards in Rea ding. This was higher than the school di strict 's average of 53%, and higher than the Florida state average of 47%. However, BG did not make Adequate Yearly Progress ( AYP ) in 2009. Under the provision of No Child Left Behind, a school makes AYP if it achieves the minimum levels of improvement determined by the state of Florida in terms of student performance an d other accountability measures. BG high school was located in a university town near a renowned public university. The school had close connection with the u niversity through student teachers and research faculty. More than half of the high school teacher s had a Master degree and they formed a faculty team with national reputation. Data Collection The data were collected from 13 ninth grade classrooms at the BG high school in the fall of the school year 2009 2010. This grade level wa s selected because stu dents in ninth grade are transitioning into high school where academic language is particularly challenging and important in learning disciplinary knowledge Background For ninth grade, world history was a mandatory course. When students moved into high s chool from middl e school, their eighth grade teachers made recommendations on their placements in all core subjects. For world his tory, there were
101 three levels: R egular, H onor s or A dvanced P lacement (AP). However, this decision could be overwritten based on students' individual needs and parents' opinions. The AP class was college bound and its students took the AP test at the end of the school year, 2010 May. The Honor track students were those who were at an advanced level in the subject of world history but not ready to take the college level class yet. The regular students were considered as those who were the lowest achieving students in this subject. According to the principal and teachers at BG high school, the decisions regarding placement were clos ely related to their 8th grade Florida Comprehensive Assessment Test ( FCAT ) scores. For example, in the current regular world history class, most students had a FCAT reading score at Level one or two. Those who were recommended to regular classes but had a reading score of four were usually moved into honor or AP classes. For the purpose of this study, thre e teachers, recommended by the P rincipal of BG were contacted and they consented to collect data in their classrooms. All three teachers possessed a Ma ster degree and teaching experience of more than five years. The AP teacher had a Master degree in Social Studies Education and twenty years' experience of teaching. The Honor s teacher had a Master degree in Social Studies Education too and had taught for five years. The Regular history teacher's Master degree is in Reading Education and she had taught in this subject for ten years. The class size varied from 15 to 28 student s. All the classes, except AP world h istory, were exclusively for ninth graders. The number and size of each class ensured that there w as a minimum of 25 students in each level of the world history class.
102 T he curriculum for ninth grade history was aligned with ninth to twelfth grade state s tandards for history. While utiliz ing historical inquiry skills and analytical processes, the curriculum included significant events, figures, and contributions of medieval civilizations (Byzantine Empire, Western Europe, Japan), Islamic, Meso and South American, and Sub Saharan African ci vilizations. Students also needed to analyze the causes, events, and effects of the Renaissance, Reformation, Scientific Revolution, and Age of Exploration and the causes, events, and effects of the Enlightenment and its impact on the American, French and other Revolutions. Also students must understand the development of Western and non Western nationalism, industrialization and imperialism, and the significant processes and consequences of each, recognize significant causes, events, figures, and consequen ces of the Great War period and the impact on worldwide balance of power as well as significant events and people from the post World War II and Cold War eras T he impact of classical Greco Roman civilization on modern western civilization was one of the f irst chapters in this curriculum. Different levels of world history classes differed to the depth and breadth of learning but the content and skills covered were similar. All these teachers believed their curriculum and instruction should build a nurturing and motivating environment for all the students. They also believed that history learning must involve a good comprehension of the course materials and an active discussion about important t opics. Teachers' instruction used mostly whole class lectures and some discussions. Students needed to read the textbook at home and bring their understanding of the textbook to school. In classes, they listened to teachers' synthesis of textbook and then finished worksheets. In terms of assign ments, the AP
103 teacher gave free r esponse questions every week and students practice d t hem as an important part of the AP world history exam. The Regular students and Honor s students, on the other hand, rarely did free response questions in their regular world history class because the teacher s believed students were not capable of answering free response questions successfully in the subject. The regular history teacher usually didn't require students to write essays in her class. A week before the writing task, she instructed stude nts in Regular class how to write essays in history. Her instruction focused on cohesion and organization. Students were advised to write five paragraphs and provide evidence for any comparison they made. Students in Honors class, on the other hand, didn't get any specific instruction regarding how to write an essay in history. Two weeks before the writing assignments, in order to get permission for students participation in this study, three world history teachers sent consent forms approved by Instituti onal Review Board (IRB) and the local school district to all ninth graders' parents or guardians. In total, 190 forms were distributed and 160 of them were returned to the teachers. Among the parents or guardians 148 consented their children's writing sam ples and FCAT data being used in this study. Through the entire study, all participants' names were anonymous and all personal information was kept private and confidential. Participants for my study were 84 ninth graders (49 male, 35 female) enrolled in a high school world history class. Please refer to Table 3 1 that summarizes participants' features. Sixty four students' data were disre garded in my study because (a) 12 of them came from other countries or private schools and thus did not have a FCAT scor e; (b ) 29 students did not complete their essay and thus were not included; (c )
104 23 students in AP classes came from other grade levels (10, 11 and 12 th ) and were not included either. This group of 84 ninth graders was an academically diverse group with va rying levels of reading proficiency. According to a recent state high stakes reading assessment (i.e., Florida Comprehensive Assessment Test), 27 of the students were at levels 1 and 2 (low achieving), 29 at level 3 (average), and 28 at level 4 & 5 (high a chieving). Procedures Participants were asked to write an essay to compare and contrast the impact of Greek culture and Roman culture on modern societies including European or American modern societies. The writing was completed in class (50 minutes) as p art of the course assignments. The procedures are described as below. Since 2009 August, through emails, phone calls, and meetings, the researcher contacted three classroom teachers recommended by the principal and discussed the possible genres and topics that participants wrote about. Since the present study compares the different use s of academic language by student s of different reading abilities, the prompt about which participants write must be identical. T he writing tasks must grow out of the content of each level's curriculum as well As mentioned earlier, the history classes of these three levels covered the same historical period and many similar topics but they were discussed in different breadths and depths. To align the instruction across readin g and writing groups i n an October 2009 meeting with all three teachers, a prompt was created to address the content taught in all three levels of world history classes. All three teachers ensured that these topics were discuss ed in their classes in detail The following prompt was decided after the meeting : to compare and
105 contrast the impact of Greek culture and Roman culture on modern societies in America or Europe. Teachers decided t he writing session date based on their curriculum delivery and ti me arrangements. All teachers met together in the department meeting and decided to conduct the writing assignments in the first week of December in 2009. By the time this writing task was conducted, all three levels had completed the unit of Greek and Rom an civilization. Therefore, student parti cipants in all three levels had been taught about Greek and Roman civilization and were familiar with the content. In all three levels of classes, student participants were notified abo ut this writing task in the beginning of the learning unit so that they could be prepared for the content. The classroom teacher was visited by the researcher before and after the writing session and the researcher observed each class for at least four class periods during th e unit. The writing sessions were also observed in order to record the context of these writing tasks and provide contextual understanding for further analysis. The participants' academic writing samples in world history classes were collected in December 2009. The classroom teacher in the natural context of history instruction managed the writing session. That is, in one class period of world history class (regular, honor and AP), each student was told by the history teacher to compose one text that insta ntiated the genre of historical comparative essay. As I mentioned before, AP students practiced writing in the genre frequently but they didn't receive specific instruction before this task. Honors students didn't receive any instruction either. The regula r students did receive some instruction with respect to how an essay could be organized.
106 In the writing session, the teachers delivered similar directions across different levels. After the participants came to each classroom and the bell rang, the teache r in each classroom handed each student a piece of paper that outlined the prompt and directions to finish the prompt. After each student in the classroom got the prompt paper, the teacher gave the following instruction: Boys and girls, today you will wri te an essay that compares and contrasts the impact of Greek culture and Roman civilization on the modern societies in America or Europe. You will finish the essay in this class period. Please look at the directions: you should spend 5 minutes organizing or outlining your essay. Your essay should have a relevant thesis and support that thesis with appropriate historical evidence. Your essay should address all parts of the question, make direct, relevant comparisons, and analyze relevant reasons for similarit ies and differences. Please include all the elements of a good essay and write as much as you can. Also if you need more time after this class period to finish the essay, you can let the teacher know and move to a quiet space to finish it. After this ins truction, no further information about task requirements, editing, or revising was provided. During the writing task, the participants worked independently in their original classrooms. Teachers provided lined paper and participants could ask for additiona l paper. All participants in the AP, honor, and regular world history classes were given a full class period (50 minutes) to write. Fifteen participants in regular world history classes did take longer time ( though less than 70 minutes) to try to complete their essays and five of their essays were included in the analyzed data set. Those disregarded writing samples were due to a lack of FCAT data and incompletion, as explained previously. After each writing task, the researcher made copies of each student' s writing. Then, I typed these texts into a computer as electronic files for further analysis. The corpus of writing samples was the primary data set in this study. The typing was
107 verbatim including spelling and grammar errors. When there was illegible wri ting within text surmises were made on participants' intended meaning in the conte xt of the task. T hese surmises were marked in the bracket right after the indistinct part. Data Analysis : Variables A history teacher and a l anguage a rt s teacher rated e ac h student text holistically for overall writing quality (1 5, with 5 being the highest) using an instr ument adapted from a state high stakes writing assessment rubric. T wo trained scorers coded t he essays for presence of academic language features includin g academic vocabulary, embedded clause, expanded noun phrase, nominalization, and lexical density. For each of the language features, a ratio score was computed by dividing its frequency of occurrence over the total number of non embedded clauses in the te xt. Finally, statistical analysis was conducted to examine the relationship between each of these language features and reading scores a s well as holistic rating of writing quality. The details regardi ng all variables are provided below. Reading Abilities To answer the three research questions posed at the outset of Chapter 3, the participants were grouped based on their reading abilities as measured by participants' achievement level in Florida Comprehensive Assessment Test (FCAT) reading test. All student participants reading scores from the FCAT in 2009 spring were requested from the school as a measurement of the participants' reading abilities. As the latest version of Florida's statewide assessment program, FCAT measures student performance on selecte d benchmarks in reading, mathematics, writing, and science that are defined by the Sunshine State Standards (SSS). Members of a school's staff, in grades 3 11, administer the FCAT to participants on regular school
108 days. The FCAT Reading section uses various written materials to measure broad areas of reading, language, and cognitive ability and assess across four content clusters: reading comprehension in the areas of words and phrases in context, main idea, compari son/cause and effect, and reference and research. Six to eight reading passages with sets of 6 11 questions based on each passage can be categorized as two types. Informational passages are taken from magazine and newspaper articles, editorials, or biograp hies and provide facts about a particular subject. Literary passages are taken from short stories, poems, folk tales or novels and are written primarily for the reader's enjoyment. The FCAT reading test is administered within a two week period in the spri ng and uses a scale with results between 100 and 5 00. Participants' performance on the FCAT reading test is described as Achievement Levels, based on both scale scores and developmental scale scores, ranging from 1 (lowest) to 5 (highest). Levels 1 and 2 r eflect below grade level performance in reading, with Level 1 being the lowest indication of reading performance. Levels 3 and above represent proficiency in reading comprehension at or above grade level standards. Student s scoring in Level 4 and 5 indicat e that they ha ve success with the most challenging content of the Sunshine State Standards (SSS). A student scoring in Level 5 answers most of the test questions correctly, including the most challenging ones. In my study, the FCAT reading test was selecte d as the measurement of participants' reading abilities, as it is a statewide assessment of participants' reading ability in school based texts at the elementary, middle, and high school levels. This test is directly linked to the Florida SSS and the compe tencies found in Florida's System of
109 School Improvement and Accountability. Therefore, it shows participants' mastery of academic language that is required by school and academic setting. Based on their FCAT reading scores, participants were placed into th ree groups: low achieving reading group of achievement level 1 and 2; average achieving reading group of achievement level 3; and high achieving group of achievement level of 4 and 5. Reading Ability was an independent variable in the current study. Three groups were: 27 students in the low achieving group (at levels 1 and 2), 29 in the average achieving group (at level 3), and 28 in the hi gh achieving group (at level 4 and 5). Writing Quality Holistic scoring was em ployed as a measure of the essays' writ ing qualities. In the present section, I will discuss the raters, rubric, training and reliability, and the writing groups engendered. Raters In order to provide information on participants' mastery of historical content and writing crafts, all the writing samples were graded holistically by both a world history teacher who was familiar with the content of world history and knew the curriculum in all three levels of world history classes (regular, honor, and AP) and a l anguage a rts teacher who had an in dep th understanding of writing requirements in Florida secondary schools. In terms of content, the evaluation focused on the accuracy (whether the historical fact and evidence provided in the essay was true or false) and the completeness (whether every aspect of the question was addressed including substantive thesis and appropriate evidence). Participants' writing competency in organization and mechanics were also taken into consideration.
110 Rubric Holistic quality scoring was used in my study as a subjective a ssessment to provide measurement on the overall quality of participants' academic writing. The scoring rubric was based on FCAT Writing Rubrics, Grade 10, downloadable from FCAT website (there is no FCAT Writing test in Grade 9). FCAT writing rubrics evalu ates focus (how clearly the paper presents and maintains a main idea, theme, or unifying point), organization (the structure or plan of development: beginning, middle, and end; and the relationship of one point to another), support (the quality of details used to explain, clarify, or define depending upon word choice, specificity, depth, credibility, and thoroughness), and convention (the punctuation, capitalization, spelling, and sentence struct ure). In this study, the genre i n which student s wrote was a c omparative essay. After consultation with all three teachers, the importance of analyzing, comparing, and contrasting was also included in the rubrics by expanding the scope of focus and taking into account the differences and similarities that student wri ters elaborated and exemplified. Overall, writing quality of all texts was evaluated using a 5 point holistic scale, with a score of 1 representing the lowest quality and a score of 5 the highest quality. Each score level had different requirements on con tent, organization, and mechanics. Below is the holistic scoring rubric on the quality of writing in each essay. This Rubric was adapted from FCAT Writing Rubric Grade 10 downloaded from FCAT website. 5 Points Content: The essay addresses all parts of the question and substantiates thesis with appropriate and accurate historical evidence. It makes at least 2 relevant, direct comparisons between or among items under consideration and analyzes at least one reason for a sim ilarity or difference identified in a direct
111 comparison. Organization: The organizational pattern provides for a logical progression of ideas. Effective use of transitional devices contributes to a sense of completeness. The writer may use creative writin g strategies. Mechanics: The writing demonstrates a mature command of language with freshness of expression. Sentence structure is varied F ew, if any, convention errors occur in mechanics, usage, punctuation, and spelling. 4 Points Content: The essay addresses all parts of the question and substantiates thesis with mostly appropriate and accurate historical evidence. It makes at least 2 relevant, direct comparisons between or among items under consideration and analyzes at least one reason fo r a similarity or difference identified in a direct comparison. Organization: The writing's organizational pattern provides for a logical progression of ideas. Effective use of transitional devices contributes to a sense of completeness. Mechanics: The w riting demonstrates a mature command of language, and there is variation in sentence structure. The response generally follows the conventions of mechanics, usage, punctuation, and spelling. 3 Points Content: The essay addresses most parts of the questio n and partially support thesis with some appropriate historical evidence. Most of the evidences are accurate historical facts. It makes at least 1 relevant, direct comparison between or among items under consideration and analyzes at least one reason for a similarity or difference identified in a direct comparison. Organization: The ideas are loosely related. An organizational pattern is apparent, and it is strengthened by the use of transitional devices. Word choice is adequate, and variation in sentence s tructure is demonstrated. Mechanics: The response generally follows the conventions of mechanics, usage, punctuation, and spelling. 2 Points Content: The essay addresses some part of the question and partially support thesis with some historical evidenc e. Some of the historical facts are accurate. It makes some comparison between or among items under consideration. Organization : The writing may lose focus by including extraneous or loosely related
112 ideas. The organizational pattern usually includes a begi nning, middle, and ending, but these elements may be brief. Mechanics: Word choice may be limited, predictable, or vague. Errors may occur in the basic conventions of sentence structure, mechanics, usage, and punctuation, but commonly used words are usua lly spelled correctly. 1 Point Content: The essay addresses little or none of the question and provides some historical evidence. Organization: The writing may lose focus by including extraneous or loosely related ideas. The response may have an organiz ational pattern, but it may lack a sense of completeness or closure. Mechanics: Limited or inappropriate word choice may obscure meaning. Frequent and blatant errors may occur in the basic conventions of sentence structure, mechanics, usage, and punctuati on, and commonly used words may be misspelled. Training and reliability The two raters were a world history teacher who taught the subject across all levels and a l anguage a rts teacher who was familiar with both the prompts i n this study and the writing requirements in the state of Florida. Both of these teachers had three years of teaching experience and a Master degree with coursework in literacy. Raters received a 50 minute training period in which they learned to use the 5 point scale by studying the rubrics and applying the rubrics on the examples of participants' writing. In a 30 minute meeting, two raters discussed with the researcher their understanding of the history content and their expectation on content, organization, and mechanics in the writ ing. The training also in volved the rating of a set of five benchmark writing sampl es, which had been gathered based on the original world history teachers' grading. These benchmark essays were selected by the three world history teachers and the researche r in a meeting and were believed to meet the
113 requirements for each level (1 5) ( see Appendix A for these benchmark writing samples). In another fifty minute session, the raters graded with the researcher together on the texts. An inter rater agreement of 1 00% on all five writing samples was achieved after the training session. After the training, the two raters rated the texts independently. Following that the raters met with the researcher to discuss their r atings in two 50 minute sessions. The raters dis cussed any divergence in scoring and explained to the group why he/she gave that score. After the discussion, the disagreement would be accepted if it still persisted. The final writing score thus averaged two teachers' scores. This decision was based on t he assumption that a teacher of history a nd a teacher of language a rts had slightly different perspectives in grading. Both of their perspectives were respected and accepted The Pearson coefficient suggests the correlations between two raters' rating was .88, which was relatively high. This confirms the outstanding inter rater reliability in the rating of writing quality. Writing groups As indicated above two teachers, one high schoo l history teacher and one high school la nguage arts teacher, graded the writing samples (N=84). They gave a score (from 1 to 5) for each text based on the same rubric we provided. In the analysis, I averaged the two sets of scores that the two teachers provided, which would be the final writing score of each essay. The mean final essay score for the entire sample was 3.18 (SD = 0.96). This final score was the dependent variable of Writing Quality. Students' essays were then placed into three groups. Please see Table 3 2 for the details. To ensure the equivalent size of three groups, the 84 graded essays were split into three groups based on the split of 1.0 2.5; 2.6 3.5; 3.6 5.0, resulting in a low
114 proficiency group of essays that received scores between 1.0 and 2.5 (n = 28), an average proficienc y group of essays that received scores between 2.6 and 3.5 (n = 29), and a high proficiency group of essays that received scores between 3.6 and 5 (n = 27). Among those essays in the low proficiency group, the majority had scores of 2.5 (n = 14); among the average proficiency group, the majority had scores of 3.0 (n = 18) and scores of 3.5 (n = 11); among the high proficiency group, the majority had scores of 4 (n = 16). The major goal of this grouping was to gain understanding the different uses of academi c language features in essays with distinct writing qualities. Academic Language Features A number of academic language features were the dependent variables in my study. They were selected based on th e research reviewed in Chapter 2 As in the previous literature review on academic language, a number of academic language features including embedded clause, high lexical density, and expanded nouns phrase are important features in academic language. Adolescents are expected to use these features proficien tly in their academic writing in their content area learning Therefore, the academic language features examined in the language analysis included academic vocabulary, expanded noun phrase embedded clause, and nominalization. Lexical density was also incl uded in the analysis as this study used this as a feature to examine students' ability of using lexical items rather than relying on grammatical items in their academic writing. After collecting all participants' writing texts, I read through the corpus of texts in order to get a basic understanding of the range of texts produced by par ticipants. Before all linguistic analysis, the first step was to segment each text into clauses: embedded (clause that serves as the participants of another clause), paratact ic ( clauses that are
11 5 linked to the main clause with a coordinating conjunction or merely juxtaposed ), hypotactic (clauses that are dependent on but not constituents of another clause) non finite clauses (clause with non finite verbs), and main clause ( the only clause in a simple sentence, the initiating clause in a paratactic sequence, or the dominant clause in a hypotactic clause complex) These clauses were numbered After this step, the first feature to be analyzed was academic vocabulary. To perform t he examination of academic vocabulary in the corpus of text, I used Coxhead (2000)'s A cademic W ord L ist (Appendix B) This list was selected because it contains 570 word families that constitute a specialized vocabulary with a good coverage of academic tex ts, regardless of the subject area. This list represents a broad range of academic texts and the words were selected based on a large corpus of text. It also serves primarily educational purposes and thus matches the goal of my study. In the preset study, the full set of text was then coded by using a macro in the software of Microsoft Office. The Marco was written to highlight every occurrence of academic vocabulary in the corpus of essays. Each occurrence of the same word (or its derivations) was counted as an individual occurrence. The researcher then calculated the occurrences of academic vocabulary in each essay. A ratio score was computed by dividing its frequency of occurrence over the total number of non embedded clause in the text. Next, other var iables of language features were analyzed. Overall, all the language features of interest were in square brackets and notes w ere marked when necessary. After that, each incident of language feature was sorted out and categorized
116 into four sets: academic vo cabulary, embedded clause, expanded noun, and nominalization All values in these analyses were accurate to the second decimal place. In order to perform a noun analysis, any expanded noun phrase in any position was identified. Based on the previous literature review, expanded noun phrases w ere defined as those noun phrases that had post modifiers such as an embedded clause or a prepositional phrase. The total number of expanded noun phrase s in each text was computed. The ratio of the number of expand ed noun phrase s to the total number of non embedded clauses in each text was the dependent variable of expanded noun phrase Embedded clause s in the texts were also analyzed. Each clause in any text was already identified as the different types of clauses A tally of the embedded clauses used in all academic writings and by each participant was maintained. The ratio of the number of embedded clauses to the number of all clauses was another dependent variable. Based on the notion that nominalization is a very important language resource of the academic ness in the historical discourse nominalization was selected as a dependent variable. In this analysis, nominalization w as sorted out and analyzed. Nominalizations such as influence, r eason or effect that also contributed to an external causality in the text was included in the identification of all nominalizations. The ratio of the occurrences of nominalizations to the number of non embedded clauses in each text was another index of dependent variable exam ined in the analyses. I then calculat ed the summary feature: lexical density. To calculate lexical density, content lexical items or content words including nouns, verbs, adjectives, and important
117 adverbs in each text were counted. The number of non embedd ed clauses was also collected. Lexical density was calculated as the number of content words/the number of non embedded clauses. Inter rater Reliability To ensure the inter rater reliability in the linguistic analysis of the corpus, a second rater other than the researcher performed linguistic analysis. Although the second rater was familiar with Schleppegrell's (2004) work on academic language an d the genres of school history as well as the goals and questions of the present study, he still received spe cific information as to each academic language feature. One fifth of all texts were rated by the second rater, meaning that he analyzed 17 texts for all grammatical features ( expanded noun phrase nominalization, embedded clause, and lexical density). An a greement at 100% was achieved for each language feature between the two raters except for nominalization at .88 and lexical density at 93, which were all better than chance. Statistical Analysis To answer the first research questions, I used descriptive statistics t o determine the range, mean, and standard deviation ( SD ) for each linguistic feature. Different means and standard deviation of each feature were compared. Then various analyses of variance procedures were carried out to investigate the obser ved differences further. Fo r the second research question a one way Multivariate Analysis of Variance (MANOVA) was conducted to explore differences in using academic language features among participants of different reading abilities. I planned to examine the differences in academic language features including academic vocabulary, lexical density, expanded noun phrase embedded clause, and nominalizat ion broken down by reading abilities.
118 MANOVA is a generalized form of univariate analysis of variance (ANOVA). It helps answer the following questions : Do changes in the independent variable(s) have significant effe cts on the dependent variables; What are interactions among the dependent variables ; and, What are interactions among the independent variables? MANOVA is useful in experimental situations where at least some of the independent variables are manipulated. It has several advantages over ANOVA (Har well, 1988). First, by measuring several dependent variables in a single experiment, there is a better chance of discovering which factor is truly important. Second, it can protect against Type I errors that might occur if multiple ANOVA's were conducted independently. Additionally, it can reveal differences not discovered by ANOVA tests. MA NOVA is a substantially mor e complicated design than ANOVA and therefore there can be some ambiguity about which independent variable affects each dependent variable. Thus, many potentially subjective assumptions must be made. These assumptions will be addressed in the data analysis. Moreover, one degree of freedom is lost for each dependent variable that is added. The gain of power obtained from decreased error varian ce may be offset by the loss in these degrees of freedom (Harwell, 1988). Fo r the third research question, m ultiple r egression (backward step wise) was conducted. It is a statistical technique that allows us to predict someone's score in one variable whil e controlling for the effects of other variables Multiple r egression is the most effective at identifying the linear relationship between a dependent variable and a combination of independent variables when its underlying independence assumptions
119 are sati sfied: each of the metric variables are normally distributed, the relationships between metric variables are linear, and the relationship between metric and dichotomous variables has equal variance (Cohen, Cohen, West, & Aiken, 2003). In Chapter 3 the me thod s of data collection and data analysis w ere described. Eighty four writing samples were included in the corpus of texts and the occurrences of five academic language features in each text were identified and computed. Statistical analyses including MAN OVA and multiple r egression were then conducted to examine the association between these features and students' reading ability and writing quality. The findings from these analyses will be present ed in Chapter 4
120 Table 3 1 Features of students Feature s Group Number of students. Students Participants Non participants 84 64 Gender Male Female 49 35 Class level Regular Honors Advanced placement (AP) 32 37 15 Table 3 2 Writing groups Writing group. Writing scores. Number of students. 1 (low proficiency) 1.0 2.5 28 2 (average0 proficiency) 2.6 3.5 29 3 (high proficiency) 3.6 5.0 27
121 CHAPTER 4 FINDING In Chapter 4, the findings of three research questions will be presented. This chapter is organized as three sections. Each section addresses one research question and describes the test of the h ypothesis posed for answering the research question Examples of academic features are also provided. Research Question I: Descriptive Statistics Research Question I explored the academic language features that students used in their historical writing. Regarding the use of academic language features, Table 4 1 shows the mean scores and standard deviations of each dependent v ariables (using SPSS) : academic vocabulary, expanded noun phrase embedded clause, nominalization, and lexical density. By referring to the Table 4 1 of descriptive analysis, we can further analyze how students used academic language features in their essa ys. Academic Vocabulary Academic vocabulary refers to specialized academic words that are prevalent across different academic disciplines. In this study, the Academic Vocabulary List by Coxhead (2000 ) was used. The descriptive analysis suggests that, in th e corpus of these high school students' academic writing, the average academic vocabulary per non embedded clause used was .41 with the standard deviation of .20. In other words in every ten non embedded clauses, only four academic vocabular y words were used on average. The maximum academic vocabulary in students' essay s was .96. T he academic vocabulary words students used were limited to a small list and students repeatedly used the words on this list T able 4 2 shows t he most frequently used academic v ocabular y words in the corpus of essays with the numbers indicating
122 the frequency of the word in the corpus of essays In total, 1392 occurrences of academic vocabulary (including multiple occurrences of the same word and its derivations) were found in th e corpus of essays. The 20 words list ed above constituted over 75% of the total numbers of academic vocabulary words in the corpus Multiple occurrences of words such as culture, cultural, influence, similarity and civilization constitute the majority of t he academic vocabulary that students used. These words were also the keywords of the lesson with respect to Roman and Greek culture that repeatedly occurred in teachers' instruction and they were part of the prompt Overall, academic vocabulary could be s een in most of the students' historical writing. But the academic vocabular y words used by students were limited to the repeated uses of a small set of words. Embedded Clause As Table 4 1 shows, in the corpus of essays, the average use of embedded clause per non embedded clause was 0.11 with the standard deviation of 0.06. In other words, approximately one in every ten clauses had an embedded clause. By examining the use of em bedded clauses, we understand how the texts were organized. In some clause structures, clauses were connected one after another, as in the following extract: 1a Their systems of Government and partial equality of women most likely inspired governments like the United States to look to Greek government for ideas (Main Clause) 1b whereas Rome's government was less influential in that respect (paratactic Clause) 1c because it was patriarchal, like most ancient, even current nations (Hypotactic Clause)
123 In th is clause complex, the p aratactic clause 1b compared the United States in the hypotactic clause 1a to Roman government and clause 1c explained further about C lause 1b The main clause was chained with the hypotactic and paratactic clauses, resulting in a c omplex and lengthy sentence. Embedded clauses, on the other hand, enabled authors to pack many lexical items in a clause that served as a part of another clause, resulting in a lexically condensed sentence. Among all 84 essays in this study, only four essays did n o t employ any embedded clause. On average, e ach essay used at least four embedded clau se Some essays used as many as 12 embedded clauses In the corpus of essays, embedded clause occurred in any position in a sentence It can be a subject, for example: 2 [ All of the things that contributed to their government and culture ] have paved the way for influences today. (Essay 21) It also occurred in the object position of a sentence, as in : 3 Army and militaries is [a thing that many mans do] (Essay 1) 4 The rivalry between the communities led to [warfare that devastated Greek society]. (Essay 77) The embedded clause could be used to modify a prepositional phrase too, for instance: 5 Unlike people in America they had to do the most out of them for each [the thing we are doing today] (Essay 50) 6 While one made theirs realistic with [stable, wrinkles thing s that are signs of age and time]. (Essay 44) Not all embedded clause s had the same functions in student's writing though. In some essays, student writers used embedded clauses to contribute to the textual
124 meaning through a leading sentence to start compar ing and contrasting the Greeks and the Romans, such as the thing the reason the way in the essay, for example: 7 [The last difference I see] between the Greeks and the Romans is in their trade (Essay 42) 8 [The next reason I am going to talk about] is the influence of religion. (Essay 9) When we removed these two embedded clauses, the experiential meanings in these two example s would not change. What changed was the interpersonal meaning because of the removal of I and a first person narrative i n the text. Please see Example 7a and 8a 7 a The last difference between the Greeks and the Romans is in their trade. 8 a The next reason is the influence of religion In many other essays, embedded clause was mostly used to realize experiential meaning and add to the lexical density in the text, for example: 9 Many other Greeks states remained as a "biligarchy" [(that is) ruled by a few], but in America we have a balanced central government [(that is) composed of three branches: the executive, leg islatives, and the other one]. (Essay 34) By using the embedded clauses "(that is) ruled by a few and (that is) composed of three branches: the executive, legislatives, and the other one ", Example 8 integrated information that could ha ve been expressed b y two long independent clause s, hence increasing the density of content in the text. The embedded clauses served as definitions of the participants biligarchy and a balance central government r espectively They played an important role in both experientia l and textual meanings. Overall, embedded clause was a prevalent feature in the cor pus of text. Most participants used it in their writing to contribute to the organization al structure in the text.
125 The use of e mbedded clauses reduced the numbers of non em bedded clauses, and therefore resulted in a denser structure and a higher lexical density. Expanded Noun Phrase Expanded noun phrase defined as a noun with post modifier(s), was the most commonly seen feature in the corpus of text. In all the essays, th e average number of expanded noun phrases in each essay was 9.92 and the maximum number was 28 in some essay. The average number of expanded noun phrase per non embedded clause, on the other hand, was only 0.24 with a SD of 0.13. It means student used one expanded noun phrase every four non embedded clause. Below a number of examples of expanded noun phrase s are provided and analyzed. In the following two examples, the post modifiers were nominal structures connected by prepositions, as in the square brack ets : 1 The most diverse things [about the two] were their cultures. (Essay 56) 2 Boys [as young as the age of seven] started training. (Essay 82) In a text of comparison and argument, it was natural to begin a clause with noun phrases that were modified by an adjective attributive like post modifier s, as in Example 1 about the two modified the most diverse things, and Example 2 as young as the age of seven was used to modify boys When nouns were placed in the first place with modifications, phenomena and chronology can be categorized, labeled and described effectively with little difficulty. Sometimes the post modifiers were nominalizations connected by prepositions. In the two examples below, the post modifier was represented a prepositional phrase includ ing nominalizations:
126 3 But by looking within the two one can see differences and similarities [of equal depth]. (Essay 43) In other examples, embedded clauses were used to modify the nouns, for instance: 4 For part of the time Rome was taken over by a dictato r [named Julies Caesar]. (Essay 84) 5 They had tragedies and comedies [that flourished Greek entertainment] (Essay 67) Expanded noun phrase s increased the density of lexical items in the text by packing lexical items together This feature was commonly pair ed with the use of embedded clause. For example, E xample 4) listed above could be rewritten as: 4 a For part of the time Rome was taken over by a dictator (main clause), and the dictator's name is Julies Caesar (paratactic clause) Written in this way, the text structure became more intricate and lexical density would be lowered down because the new sentence consisted of two non embedded clauses and more grammatical items such as and and the Using an embedded clause as a post modifier increase d the cohesiveness and tightness of the structure and creates a flow of information. With the use of this post modifier, the definition of a dictator was established within the clause and the readers would be able to focus on main action taken over in this histo rical event, which also represents better cohesion in the academic writing. Together with embedded clauses, expanded noun phrase s enabled this sentence to accommodate a great deal of lexical materials and provided a more dense presentation of information. Overall we can see that in the corpus expanded noun phrase s were frequently used. O n average, one expanded noun phrase was used in every four noun embedded clause s Expanded noun phrase was used mostly as a subject, object, or in a
127 prepositional phrases. The use of them contributed to a more tight structure and better cohesiveness in the historical writing. Nominalization Nomina lization expresses a meaning in nouns that could be naturally used as verbs or adjectives Among all 84 participants, 20 of them did not use any nominalization in their writing S ome essays used more than 10 nominalizations. In other words, approximately one fourth of the students did n ot exhibit any use of nominalization in their writing. The re were also students who showed frequent use of this feature. As Table 4 1 shows, in the corpus of essays, the average use of nominalization per non embedded clause was 0.06 with the standard deviation of 0.06. On average, there were only 6 nominalization s being used in 100 non embedded clauses. The nominalizations used by 64 students had different functions. First, students used nominalization to realize abstractness. For example, in the Example 1a: 1 a A large number of Greeks left where they lived to go to a new place in distant lands, so they could have good farmland and [the growth of trade]. (Essay 44) The more congruent way of organizing this sentence could be: 1b A large number of Greeks left where they lived to go to a new place in distant lands, s o they could have good farmland and [the trade could grow]. In Example 1a, the growth of trade became a more general and abstract concept than a concrete narration such as the t rade could grow Also, the causality ( Greeks left --the growth of trade) could be built within one independent clause, which also contributed to an abstract causal connection between two historical events. The former usage is commonly seen in academic writing. The same function can be observed in E xample 2 :
128 2 After the collapse of Myce naean Civilization, Greece entered a difficult period in which the population declined and food production dropped (Essay 83) Both the use of the collapse of Mycenaean Civilization and food production increased the abstractness of the text. Second, the frequent use of nominalization rendered texts more formal and objective. Through nominalization, actions could be realized into entities and the actor and goal as modifiers could be omitted. In some specific situations, the writer was willing to use nomina lization to de emphasize the actor or agent. Other times the writer did not know whom the actor or agent was, or the writer regarded the event more important than the actor or agent. Please refer to Example 3 : 3 As far as religion goes, both were polytheist ic, meaning they believed in the existence of multiple gods. Mythology can be traced in both societies. The Romans borrowed the Greek gods and basically gave them names of their own, but bo th of their mythological ideas and concep ts were along the same lin es However, unlike Greece, the spread of Christianity developed throughout Rome, despite their efforts to abolish the new religion. (Essay 56) In this paragraph, the event the spread of Christianity was more important than who spread it. Since the topic or the theme was the most prominent part to the writer, with nominalizations, the writer took the actor out of the flow of sentence. The reader's attention was less likely to be fixed on the actor ( who spreads Christianity ), but on the event ( the spread o f Christianity) in the subject position. In addition, the use of nominalization contributed to the textual meaning, the flow of text, and the overall organization. By changing verbal or clausal meaning into a nominal element, nominalization enabled somethi ng that had been presented as a
129 clause to be d istilled into a noun phrase. And then this phrase could participate in a chain of reasoning a s in Example 4 : 4 All adult males were allowed to vote on issues, thus giving them power over their own fate. In Rome, there was not even partial democracy. Unlike Greece, Rome's Government was less lenient and equal. Woman had many more rights under Greek government (in some parts) than Rome. They had jobs in trade, and in general had a mere equal standing with men than Rome. Their systems of Government and partial equality of women most likely inspired governments like the United States to look to Greek government for ideas, whereas Rome's government was less influential in that respect because it was patriarchal, like m ost ancient, even current nations. (Essay 1) In this example, the use of "partial equality of women" distilled the whole comparison regarding women's societal standing in Greece and Rome into a noun phrase. This noun phrase then became the subject of the f ollowing clause. It served as the starting point for the argument regarding how this system of inequality influenced the western society, which was the focus of the essay. This us e of nominalization could demonstrate that the writer expected the reader to be able to connect with the information that was distilled into the phrases. Last but not least, in many situations, nominalizations were used to rea lize causality, as in Example 5 and 6 5 First, Greek religion is [a huge factor and influential factor] on modern America (Essay 78) 6 In conclusion you can see the classical Greek have [many affects and influence] on modern Americans (Essay 75) Rather than using because or other causal conjunctions, the students used nouns such as factor, affect, and influence to construe the cause effect relationship in a more intact and objective form. If a conjunction was used, a clause of cause led by because and the other clause must be involved in Example 6. Moreover if the conjunction of because
130 was used, more i ntricate description must be involved to describe the effects of multiple affects and influences, which would obscure the causality in this sentence. Across the corpus of essays in the present study, the most common nominalization was influence. Students usually used the word of influence to start their essay by stating the influences of Greek and Roman cultures. Then they used several paragraphs to describe these influences. The type of idea flow was popular in the corpus of text, through which a cause effect relatio nship was built. This causality then led the overall organizational structure, as in the following paragraph : 7 The [long term influence] also had some alikes and differences. The Romans were influenced by the Greeks and Latins. The Gree ks were influenced by the minoiuas. They both had architecture, engineering, literature, and a writing system. The Greeks had more art than the Romans did. The Greeks were influenced in art and politics. They even made more routes to trade good with. The R omans had a road system that was used for trade. They were both influenced in science and astronomy. The cause effect relationship construed by the nominalization of influence led the flow of reasoning in this paragraph. In this paragraph, various aspects of these causalities were listed by focusing on influences The causality in different aspects of life including architecture, engineering, literature, writing system and etc. was listed to organize the flow of argument in the whole paragraph. Overall, n ominalization served the function of building abstractness, contributing to objectivity by embedding implicit perspectives, distilling information, and creating the flow of argument. Nevertheless, the feature of nominalization was not prevalent in the corp us of the text. Most students used very few nominalizations in their writing.
131 Lexical Density The descriptive analysis suggests that, in the corpus of these historical essays, the average lexical density was 4.86 with the standard deviation of 0.87 It means that in every non embedded c lause, an average of less than five content words or lexical items was used. According to Christie and Derewianka 's (2008) study of writing development, the lexical density starts around 3 to 4 in childhood and rises to ab out 5 by early adolescence. In late adolescents when students are 17 years or older, this can rise to about 6. This lexical density of 4.86 indicated a low density and a low use of content words across the corpus of text ( Halliday & Matthiessen, 2004) A cademic Language Features in the Text Descriptive analysis of the examples of student texts can also demonstrate students' use of academic language features. Below I will discuss all the five features in the context of three essays. Academic vocabulary In Essay 1 written by a student in Reading Group 1 (low reading ability group) we can see how academic vocabulary (in bold ) was used in students' writing: 1 A Wow! I didn't know, in 800 B.C., the Greeks had dramas Is that what "today" Americans call Soap Operas? Well, like it or not, but we Americans do new things that ancient people already done. In other terms, the Greeks have influenced modern America in a lot of ways like, Politics, Arts, and Literature. Let's find out how they influenced America. Can you believe that the Greek Philosopher Aristotle was the person who favored the constitutional government as the best form for most people? Aristotle is the student of the greatest philosopher of western civilization Plato. Plato's ideal state was, for philosophers to become kings of their countries. But, unlike Plato, Aristotle didn't seek an idea state. So, Aristotle at the constitutions of 158 states and found 3 good forms of government: monarchy, aristocracy and constitutional government. Did you know that?
132 Like I said in the beginning, the Greeks used to have dramas Dramas for them were called tragedies. It was called that because they examined such problems as the nature of good & evil, the rights of the individua l and etc. Today we use drama as Soap Operas. It deals with tragedies sometimes. Then Greek comedy developed after tragedies. Comedy tried to make a point, while making laugh their robes off. So should we laugh now, or cry later? For today Americans, literature is the most important. Why? Well, math, reading, science, astronomy, and etc. Greeks were doing that stuff before were even born. In this essay, although 13 occurrences of 6 academic vocabular y words could be found, the author repeated using word such as philosopher and drama. In terms of the word choice, the student many times selected the non academic vocabulary to represent their meaning. For example, the students used words such as influence rather than impact or new rather than innovative. The latter words both belong to Coxhead's academic vocabulary list (2000) The second essay was from a student from reading group 2 (academic vocabulary words are in bold) : 2 Romans and Greeks are two great people in the past. Their culture and government was so alike, but different, the Romans with their everlasting writing and the Greeks with their marvelous statues. The Romans and the Greeks are so alike and they are different. The Romans gods were originated from the Greeks. All the god s were exactly the same, but they had different names. They still saw Zeus or Jupiter as the god of the heavens and Hera or Junow was still his wife. A difference is they worshiped different gods as better than other. For the Greeks they disliked Ares or M ars because they hated war, but the Romans loved him because they enjoy war. Their government was very different. While the Greeks stayed as a democracy the Romans government changed. They went from a Republic to an empire. For the Greeks there was voting among the people. The Romans had people that were elected a position to make the laws and rules.
133 Something in their government that was the same was that only men that were citizens of the Empire could vote or be in an election for a position. The culture s of both empires influence us today. They both influence people in architect. The Greeks were known for their statues that showed strength and power, but no face expressions. The Romans showed statues that were doing an activity and they did have face exp ressions. One difference was the Romans Latin language was used more than the Greeks. Today it is used in all the languages and writing. That is why the Romans and Greeks are the same, but very different. Even though they share the same gods they worship them different. Also they have a different government system, but the only allowed men to elect. For their influence today they both had statues, but they were shaped different. Also the Romans Latin language has lived on when the Greeks has not. In Essa y 2 only one academic vocabulary word was used and it was used twice. The word same was used repeatedly But unlike its synonym, similar the word of same is not an academic vocabulary word In another example : While the Greeks stayed as a democracy the Romans government changed. A synonym of while, whereas could be used to express the similar meaning as an academic vocabulary: Whereas the Greeks stayed as a democracy the Romans government changed. Based on these examples, we could see that studen ts were able to present certain concepts but many times they failed to use academic vocabula ry words to present these concepts. In Essay 3 (all academic vocabulary words are in bold ): 3 The Romans and Greeks greatly influenced later civilizations even th e present day. Their accomplishments, ideas and examples were revolutionary and many of them are incorporated into modern day society. These two great empires were powerful, important parts of history. Their governments were predecessors to some of the mos t influential and
134 successful governments today, like the United States and many others. Three of their most well known influences on modern culture are their religion, government, and the level of equality they had with their empires. Greek and Roman Gods have influenced an unimaginable number of literacy words. Authors of thousands of different genres of books have taken inspiration from the Greek and Roman Gods. The Greek Gods were the forefathers of the Roman Gods. They are incredibly similar in their p owers, attitudes and human like qualities. The Roman Gods were believed to be a sort of copycat of the Greek Gods, but nevertheless were a great influence on the world. Both the Roman and Greek gods were temperamental, based on common necessities of life and or unexplainable events or forces, and had many Gods and Goddesses. Thus they backed each other up, and strengthened their influence on future peoples. So the Roman and Greek gods had similar characteristics, formed for different reasons. Both Greece and Rome were broken up into city states polis Their forms of governments in this way were quite similar H owever in some parts of Greece Sparta these were at least a partial democracy. All adult males were allowed to vote on issues thus giving them powe r over their own fate. In Rome, there was not even partial democracy. Unlike Greece Rome's Government was less lenient and equal. Woman had many more rights under Greek government in some parts than Rome. They had jobs in trade, and in general had a mere equal standing with men than Rome. Their systems of Government and partial equality of women most likely inspired governments like the United States to look to Greek government for ideas, whereas Rome's government was l ess influential in that respect because it was patriarchal, like most ancient, even current nations. Slaves were imperative to the survival and well being of both Greece and Rome. They allowed Greek people the time to vote, sculpt, and study science, and philosophy They also were one third of Rome's population and thus were very influential to their wealth and survival providing labor for agriculture etc. S laves allowed the creation of the famous Greek philosophy art, science, government and sculpting w hich are still used today. Slaves also allowed the Romans to survive and prosper, keeping their civilization alive, and allowing it to expand its power and influence, thus allowing it to have a greater influence on the present day. Greece and Rome were tw o major civilizations from the past whose influences were so great that they survived and prospered even the present times. Many other civilizations have left their mark on the world through countless means. Each one adds a piece of culture and knowledge t o our ever growing supply, and allows us to learn from the past and improve our present.
135 In this essay, written by a student whose reading achievement level was 5, 28 occurrences of 18 academic vocabulary words could be seen. V arious vocabularies could be found in this essay. Essay 3 demonstrated more occ urrences of academic vocabulary words and more diverse use of them, compared to Essay 2 and Essay 1. In Essay 3, we could see a variety of academic vocabulary words such as creation, issues, expand rather t han the repeated uses of culture, philosophy or similar. The categories of academic vocabulary words were different in three essays. In all three essays, many of the academic vocabulary words were general nouns that replaced concrete people or action. For instance, individual in Essay 1, culture in Essay 2, and adult in Essay 3 are all examples of this kind. The use of these academic vocabulary words contributed to the abstraction in the historical writing. Essay 3 also used academic conjunction words such as nevertheless and whereas while Essay 1 and 2 did not utilize any of these conjunction words. It also suggests a more effective use of academic vocabulary words in Essay 3. Expanded noun phrase As for expanded noun phrase s, Table 4 3 shows the expanded noun phrase that could be found in these three essays. A student from advanced reading group wrot e Essay 3 in which 21 expanded noun phrases were used. Essay 2 used 10 expanded noun phrases, and Essay 1 written by the students from the lowest reading grou p used only seven of them. These expanded noun phrase s served different functions in the essays. Sometimes, det ailed or extensive information wa s packaged into a single noun that refers to a series of events. For example, the expanded noun phrase predecessors to some of the most influential and successful governments today semantically refers to
136 many events and activities. Through the use of these expanded noun phrases, a series of actions and events were condensed into one element. Another functi on of expanded noun phrase was to expand explanations. Expanded noun phrases are used as grammatical participants to substitute real people to act in the historical events. In Example 1: 2 Their systems of Government and partial equality of women most likely inspired governments like the United States to look to Greek government for ideas, whereas Rome's government was less influential in that respect because it was patriarchal, like most ancient, even current nations. The expanded noun phrase partial equalit y of women did not only expand the explanation but also realized the causal relations in the sentence that the factor of partial equality of women caused modern governments to learn from them. We can also see that in these essays the expanded n oun phrases also took the function t o structure reasoning and contribute to the flow of information in a text. As we could see in Example 2: 2 Slaves allowed the creation of the famous Greek philosophy art, science, government and sculpting whi ch are still used today. The reasoning was structured between slaves and the creation of the famous Greek philosophy, art, science, government and sculpting which are still used today in a way that the existence of slaves became objective and the actor of slave system was much less prominent. Effective uses of expanded nouns rely on the realization of these three functions, which we could see in the Essay 3 that was written by a student with relatively higher reading ability. In Essay 1, the student mostly used expanded noun phrases to package information. In Essay 2, the student used some of them to expand explanation such as
137 t he cultures of both empires The most successful and effective use of expanded noun phrase was Essay 1. The expanded noun phrases made this essay more dense and abstract compared to the first two essays. Embedded clause Table 4 4 provides a summary of the analysis of embedded clauses in these three writing samples (embedded clauses are in bracket ). In these three essays, embedded clauses were used as the subject, object, and prepositional phrase in clauses units. They all contributed to the experiential meanings. There were only slight differences. According to the analysis, although E ssay 3 used fewer embedded clause compared with Essay 2, the Essay 3 exhibited more accu rate use of embedded clause. The Essay 2 repeated using that rather than who for embedded clause th at modifies people. The Essay 1 used only two embedded clauses. In terms of the functions of these embedded clauses, in Essay 3, all three embedded clauses served as a role that enabled the connections and causal relations between different events and actions can be developed within the clause, which significantly contributed to the explanation of causes and consequences in this historical writing. On the con trary, Essay 1 and Essay 2 used embedded clauses more like modifiers to define some nouns. Overall, through the use of embedded clauses, the objects that they modified were highlighted and emphasized. A less grammatical ly intricate style (fewer hierarchic al levels of clauses were used) and a more lexically condensed style were achieved. Moreover, causal connections and logical connections can be achieved within the clauses through the uses of embedded clause.
138 Nominalization The average use of nominalizati on per non embedded clause increased dramatically from Reading Group 1's .17 to Reading Group 2's .53. This average use of nominalization was even higher at .61 for Reading Group 3. Table 4 5 shows the descriptive analysis of these three writing samples in detail (the nominalization was in the bracket in the sentence): We can see a vast distinction in the se three essays. In the Essay 3 nominalizations could be found in the subject, object, and prepositional phrases. We could see abstract and technical ter ms such as necessity, equality, and survival. We can also see how the student used the nominalization survival and then elaborated other aspects to make people survive including labor for agriculture. The advanced reading student relied on nominalization t o develop the comparison among Greek, Roman civilization, and m odern cultures, whereas Essay 2 used only one nominalization and Essay 1 used none. As for the functions of nominalization, we can see the important role of nominalization in developing causality in the essays. For example, in sentence 3f, one third of Rome's population caused their wealth and survival as abstract and general participants, in which an abstract causal relationship was established. We can also see that the student started to reflect on the significance of cause and consequence in historical events. Lexical d ensity The lexical density of Essay 3 was 6.71, while Essay 2 's lexical density is 4.07 and the Essay 1 was 4.15. The low lexical density of Essay 2 was mainly due to its
139 clause combining strategies: relatively simple clauses were used and the percentage of embedded clause was low. Ove rall, different academic language features including academic vocabulary, embedded clause, expanded noun phrase s, and nominalization can be found across the corpus of essays. On average these texts exhibited a limited use of academic features and a low le xical density. There are many oral register features in low achieving essays, suggesting that the students were drawing on oral registers, but more successful essays draw more on academic registers. The discussion regarding these features and the interpret ation of them based on theoretical framework and literature review will be elaborated in Chapter 5 of Discussion and Implications. Research Question II: MANOVA Results The second research question examined whether academic language use is related to readi ng ability. The hypothesis one was th at the academic language use would not be changed with the change of reading ability. In all the analyses, academic language feature s were dependent variables and r eading ability was the independent variable. To examin e this research question and test the hypothesis, I first tried to find out if variables (general features: academic vocabulary, expanded noun phrase s, embedded clause, and nominalization; and lexical density) were correlated to each other. A moderate corr elation among the dependent variable s further confirmed that a one way multivaria te analysis of variance (MANOVA ) should be conducted for all five academic language features.
140 Correlation Before MANOVA, correlation was compute d to examine the association a mong dependent variable s within each set. This was also to confirm one of the factors why a MANOVA was conducted. Table 4 6 provides a matrix of the correlation coefficient for the five dependent variable s. The relationship between any two of these five d ependent variables was investigated using Pearson Product moment correlation coefficient. All variables were correlated to each other positively. According to the strength of correlation defined by Cohen (1988), Lexical Density has moderate correlation wit h Nominalization (r=.522) and Academic Vocabulary (r=.559), Expanded noun phrase (r=.496), and Embedded Clause (r=.363). The correlations among dependent variable s ranged from .123 to. 475 and suggested the existence of moderate correlations. The moderate correlation among dependent variables further confirmed that a MA NOVA was more appropriate statistics for the present study than a number of simple analysis of variance (ANOVAs). The primary reason was: as dependent variables were correlated with each othe r, the findings from separate ANOVAs would be redundant and difficult to integrate. Meanwhile, the family wise error rate becomes high; the odds of finding something are significant simply because chance rises with repeated use of the same sample of data. MANOVA Inferential statistics were then used to test mean differences for significance through a one way MANOVA. A post hoc test of multiple comparisons (ScheffÂŽ) was selected based on the results of the Levene's statistic (Coakes & Steed, 2007). The acad emic language features including academic vocabulary, expanded noun phrase embedded clause, and nominalization were the dependent variable s in my study
141 Additionally, I inc luded lexical d ensity as a dependent variable in the overall MANOVA analysis. Altho ugh lexical density was believed to have an overarching nature and was related to all academic language use, as other a cademic features, it represents the similar underlying theoretical construct and represents the lexical items that are used in academic w riting Also f or statistical purpose, it was more meaningful to include it in the whole model to control the overall alpha level to some constant (Bray & Maxwell, 1982). Before conducting MANOVA, its assumptions were tested. The assumptions of MANOVA include univariate and multivariate normality. First, all of the dependent variable s must be distributed normally. This was visualized with histograms in SPSS. The Shapiro Wil k test was used to determine the normality of all dependent variables. Second, any linear combination of the dependent variable s must be distributed normally. Pairwise relationships among the dependent variable s were checked out for nonlinear relationships using scatter plots and the assumption was confirmed. MANOVA also requires that the "covariance matrices" be homogeneous. Computations in MANOVA require the use of matrix algebra, and the matrices of the covariances (the variance shared between any two va riables) have to be equal across all levels of the independent variable. The hypothesis was that the covariance matrices of the dependent variables are significantly different across levels of the independent variable. This homogeneity assumption was teste d a nd held using Box's M (Table 4 7 ) and further explanation was provided with multivariate results (Table 4 8 ). Box's M test (Table 4 7) demonstrates the test was significant (which means that there were significant differences among the reading gro ups in the covariance matrices). Since high power (.912) was already identified, a s in Table 4 8, this result of
142 significance is not a problem. MANOVA is robust to violations of homogeneity of covariance when groups' sample sizes are equal, as in the current stu dy (Stevens, 1996). However, when Box's test finds that the covariance matrices are significantly different across levels of the independent variables it may indicate an increased possibility of Type I error (Stevens, 1996). So I made a smaller error regio n and changed my interpretation of the results according to the new significance level of .01. After all the assumptions were tested a one way MANOVA was conducted using SPSS 16.0. The MANOVA results revealed there was a significant multivariate main eff ect between reading ability and academic language use 2.252, p <. 01, partial eta squared = .212. Power to detect the effect was .912. Thus hypothesis 1 (the academic language use did not change with the change of reading abi lity) was rejected (see Table 4 8 ). This indicates that there was a significant difference among the three levels of the independent variable Since a significant multivariate main effect was obtained for the independent variable univariate F tests were then conducted to look at each dependent variable in turn to see if the independent variable had a significant impact on them separately. As I was doing five tests here and I would require an experiment wise alpha rate of .05, I divided it by five to get an acceptable confidence level for each of the five tests, so the alpha level was set to p < .01. By that criterion, there was a significant univariate results were between reading ability and lexical density and nominalization. Significant univariate main effects for reading were obtained for lexical density, F (2, 81) = 5.569, p <.005, partial eta square =.121, power = .843; and nominalization, F (2, 81) =7.476, p <.001, partial eta square =.203, power = .935. The re sults are displayed in Table 4 9 :
143 In or der to examine the differences between reading abilities in lexical density and nominalization, considered two at a time, I looked at the results of the Levene's statistic to explore what kind of post hoc test were appropriate for follow u p analysis with r espect to the r eading ability (Table 4 10) The Levene's statistics for lexical density that had significant univariate ANOVAs were non significant, meaning that the group variances were equal, so I could use the SheffÂŽ tests for comparing pairwise group means. The Levene's test for nominalization w as significant though. Again, as the MANOVA with same sizes of groups is robust to violation of normality, they were kept in the further analyses (Stevens, 1996). Since I was doing two significance tests (K (k 1)/2) looking at the pairwise tests comparing the use of academic language features by reading ability, I used the smaller confidence level again to protect against inflated alpha error. I divided the .05 by 2 and set .025 as my error level. By this standard, Table 4 11 shows that for mean scores for lexical d ensity were statistically significantly different between Reading 1 and Reading 3 (P < .025) but not for Reading 2 and Reading 3 (P = .658) or Reading 1 and Reading 2 (P = .068). Mean scores of Nominalization were statistically different between Reading 1 and Reading 2 (P < .025) as well as between Reading 1 and Reading 3 (P < .025), but not for Reading 2 and Reading 3 (P = .457). To summarize, a one way MANOVA suggests there was a signific ant multivariate main effect between overall academic lan guage use and reading ability Thus the null hypothesis that the academic language use did not change with the change of reading ability was rejected. Although a multivariate main effect was obtained, a significant
144 univariate main effect could only be foun d for the features of nominalization and lexical density. Research Question III: Regression Research Question III first examined the different uses of the focal academic language features among high medium, and low quality essays, based on the grouping indicated by the holistic score s of the essays. The second part of Research Question III looked at the prediction of academic language features to writing quality. In order to answer the first part of this research question, I first compared different mean s of academic language features among different groups. Then, I conduct ed a regression analysis using writing q uality as the dependent variable and academic language features as the predictor variables to determine which academic language features were the most predictive of writing q uality and account for the largest amount of variance associated with writing q uality. ANOVA Table 4 12 shows the comparison of means in terms of all five academic language features selected for this study. As it can be seen fr om the Table 4 12 higher writing quality groups shows more use of all academic language features compared to lower writing quality groups. ANOVA for the low proficiency (i.e., scored 1.0 2.5), average proficiency (i.e., scored 2.6 3.5), and high proficie ncy (i.e., scored 3.6 5.0) essays are reported in Table 4 13 The low average and high proficiency essays were significantly different in terms of lexical density (the number of content words per non embedded clause): F (2, 81)=22.369, p .001; expanded noun phrase (the number of expanded noun phrase per
145 non embedded clause): F (2, 81)=3.954, p .05; nominalization (the number of nominalization number per embedded clauses): F (2, 81) =29.397, p .001; but not for academic vocabulary (the number of academic vocabulary pe r non embedded clause); and embedded clause (the number of embedded clause per non embedded clause). Multiple Regression A backward stepwise regression analysis was conducted to examine which of the five variables was/were predictive of holi stic essay ratings using the continuous score. These five variables were regressed against the holistic evaluations for the 84 evaluated essays. The backwards stepwise regression began with an examination of the combined effect of all of the independent va riables on the dependent variable. One by one, independent variables were removed, and a new analysis was performed (Table 4 14 ). The order of removing was lexical density, nominalization, embedded clause, academic vocabulary, and expanded noun phrase The selection rule was based on the significance of each feature to secondary students implied by the literature in Chapter 2. Before running the Multiple Regression, the assumptions were checked. The variables were checked for outliers and multicollinearity. The outliers' values demonstrated that there were no independent errors caused by residuals. In SPSS 16.0, I used the Explore command to look at the normality of all the residuals. All of the results from the Explore command suggested that the residuals w ere normally distributed -the skewness and kurtosis were near 0, the tests of normality were not significant, the histogram looked normal, and the Q Q plot looked normal as well. Based
146 on these results, the residuals from this regression appeared to conf orm to the assumption of being normally distributed. Then, independent variables were checked for multicollinearity and the absolute values of Pearson correlation were all less than 0.8, collinearity was unlikely to exist (Field, 2005). Coefficient values demonstrated that the model's data did not suffer from multicollinearity. The "tolerance" was an indication of the percent of variance in the predictor that could not be accounted for by the other predictors, hence very small values indicate that a predict or was redundant, and values that were less than .10 may merit further investigation. As the results showed, the tolerance values were all acceptable (Field, 2005). VIF stands for variance inflation factor, is (1 / tolerance). When a variable's VIF values were greater than 10, it may merit further investigation. All VIF values were a little over 1, the threshold for multicollinearity (Field, 2005), but much smaller than 10, which suggest that multicolli nearity was not violated in my study. The regression an alysis ( see Table 4 15 ) shows that the indices of nominalization, lexical density and academic vocabulary had statistically significant effect on essay ratings, F (7, 46) = 10.54, p < .001, r = .69 R = .56, adjusted R = .55. Thus, the three indices comb ined (nominalization, academic vocabulary, and lexical density) accounted for 56% of the variance in the evaluation of the 84 essays examined. Among three predictors, nominalization accounted for 43.1% of the variance in writing quality controlling for oth er factors. Lexical density accounted for 10.4% of the variance and academic vocabulary accounted 2.9% of the variance controlling for other factors. Overall, the statistical results reveal that essays of different writing qualities used academic language features in different ways. Higher quality essays used significantly
147 more nominalization and academi c vocabulary and they also displayed higher lexical density. We could not detect significant differences in using the feature s of expanded noun phrase and e mbedded clause though. Among all the five independent variables, three factors of writing quality were found: nominalization, lexical density, and academic vocabulary. Nominalization accounts for 43.1% of the variance in writing quality, lexical density ac counts for 10.4% of the variance and academic vocabulary accounts 2.9% of the variance controlling for other factors. To sum up the results of the d escriptive statistical analysis addressed the three research questions. First of all, analysis indicate s th at the use of five target academic language features could be seen across the entire sample. The most commonly seen features were academic vocabulary word and expanded noun phrase. Embedded clauses could be seen in students' writing as well. N ominalization was the least commonly seen feature in the writing samples. The average lexical density in the writing samples was relatively low The functions of these features varied in students' writing. For the second research question higher reading groups used more academic language features including nominalization and lexical density. A one way MANOVA suggested that there was a significan t multivariate main effect between reading a bility, which means students with different reading abilities used academic lang uage features differently Compared to students with lower reading abilities, students of higher reading abilities used significantly more nominalization s and showed higher lexical density in their writing. For the third research question, c ompared to lower quality essays, higher quality essays used significantly more academic language features including expanded noun
148 phrase, nominalization, and lexical density T he regression analysis shows that the indices of nominalization, lexical density and academ ic v oca bulary significantly predicted writing q uality. The finding revealed a significant impact of academic language features including nominalization, lexical density, and academic vocabulary on writing quality. T he three indices combined (nominalization academic vocabulary, and l exical density) account for 56% of the variance in the writing quality of the 84 essays examined. Among these three features, nominalization could predict most of the variances in writing quality. Whereas the nominalization and lexical density had positive impact on writing quality, the presence of academic vocabulary seemed decreasing the essay quality. For every unit increased in academic vocabulary, we expected a 1.124 unit decrease in the writing score, holding all other var iables constant. In Chapter 5, these findings will be contextualized, discussed, and interpreted with the research questions and reviewed literature regarding academic language in secondary schools.
149 Table 4 1. Descriptive statistics Variable N Minim um Maximum Mean Std. d eviation Academic v ocabulary 84 .00 .96 .412 .198 Embedded c lause 84 .00 .27 .10 6 .058 Expanded n oun phrase 84 .05 .57 .241 .125 Nominalization 84 .00 .25 .062 .058 Lexical d ensity 84 3.25 8.00 4.86 3 .873 Valid N (listwise) 84 Table 4 2 Frequent word and their occurrence frequency Academic vocabulary word. Frequency of occurrence affect aspect c ivilization c ollapse c ulture/cultural decline despite drama establish impact influence isolate major military participate philosophy/philosopher period similar/similarity somewhat unique 14 10 49 5 240 2 3 52 11 17 177 3 24 38 8 45 9 188 8 4
150 Table 4 3 Expanded noun phrase analysis Essay Expanded noun phrases 1) 1a. new things that ancient people already don e, 1b. the person who favored the constitutional government as the best form for most people 1c. the student of the greatest philosopher of western civilization 1d. kings of their countries 1e. the constitutions of 158 states 1f. the nature of good & evil 1g. the rights of the individual 2) 2a. two great people in the past 2b. the god of the heavens 2c. voting among the people 2d. people that were elected a position to make the laws and rules 2e. Something in their government that was the same 2f. men that were citizens of the Empire 2g. an election for a position 2h. The cultures of both empires 2i. people in architect 2j. statues that were doing an activity 3) 3a. predecessors to some of the most influential and successful governments today 3b. Three of their most well known influences on modern culture 3c. the level of equality they had with their empires 3d. Authors of thousands of different genres of books 3e. inspiration from the Greek and Roman Gods 3f. the forefathers of the Roman Gods 3g. a sort of copycat of the Greek Gods 3h. a great influence on the world 3i. common necessities of life 3j. their influence on future peoples 3k. power over their own fate 3l. jobs in trade 3m. a mere equal standing with men 3n. Their system s of Government 3o. partial equality of women 3p. well being of both Greece and Rome 3q. labor for agriculture 3r. the creation of the famous Greek philosophy, art, science, government and sculpting which are still used today 3s. a greater influence on the present d ay 3t. two major civilizations from the past whose influences were so great that they survived and prospered even the present times
151 3u. their mark on the world Table 4 4 Embedded clause analysis Essay Reading Group Embedded clause 1 1 1a. Well, like it or not, but we Americans do new things [that ancient people already done]. 1b. Can you believe that the Greek Philosopher Aristotle was the person [who favored the constitutional government as the best form for most people]? 2 2 2a. The Romans had people [that were elected a position to make the laws and rules]. 2b. Something in their government [that was the same] was that 2c. only men [that were citizens of the Empire] could vote or be in an election for a position. 2d. The Greeks were known for their statue s [that showed strength and power, but no face expressions]. 2e. The Romans showed statues [that were doing an activity] 3 3 3a. Three of their most well known influences on modern culture are their religion, government, and the level of equality [ they had with their empires]. 3b. Slaves allowed the creation of the famous Greek philosophy, art, science, government and sculpting [ which are still used today]. 3c. Greece and Rome were two major civilizations from the past [whose influences were so great that they survived]
152 Table 4 5 Nominalization analysis Essay Reading group. Nominalization 1 1 N.A. 2 2 2a. Their culture and government/ was so alike, but different, the Romans with [their everlasting writing] 3 3 3a. [Their accomplishments], ideas and examples were revolutionary and many of them are incorporated into modern day society 3b. Three of their most well known [influences] on modern culture are their religion, government, and [the level of equality they had with their empires] 3c. Both the Roman and Greek gods were temperamental, based on [common necessities of life] and or unexplainable events or force s and had many Gods and Goddesses 3d. Their systems of Government and [partial equality of women] most likely inspired governments like the United States to look to Greek government for ideas, whereas Rome's government was less influential in that respect because it was patriarchal, like most ancient, even current nations 3e. Slaves were imperative to the [surviva l] and well being of both Greece and Rome. 3f. They also were one third of Rome's population and thus was very influential to [their wealth and survival], providing labor for agriculture etc Table 4 6 Correlation analysis Academic v ocabulary Lexical d ensity Expanded n ouns Nominalization Embedded c lause Academic v ocabulary 1 .559** .475** .255* .203 Lexical d ensity .559** 1 .496** .522** .363** Expanded noun phrase s .475** .496** 1 .260* .471** Nominalization .255* .522** .260* 1 .123 Embedded c lause .203 .363** .471** .123 1 **. Correlation is significant at the 0.01 level (2 tailed). *. Correlation is significant at the 0.05 level (2 tailed).
153 Table 4 7 Box's t est of e quality of c ovariance Box's M 61.532 F 1.868 df1 30 df2 20648.806 Sig. .003 Table 4 8 Multivariate t ests Effect Value F Error df Sig. Observed p ower b Intercept Pillai's Trace .977 645.999 a 77.000 .000 1.000 Wilks' Lambda .023 645.999 a 77.000 .000 1.000 Hotelling's Trace 41.948 645.999 a 77.000 .000 1.000 Roy's Largest Root 41.948 645.999 a 77.000 .000 1.000 READING Pillai's Trace .249 2.216 156.000 .019 .906 Wilks' Lambda .761 2.252 a 154.000 .018 .912 Hotelling's Trace .301 2.286 152.000 .016 .916 Roy's Largest Root .248 3.875 c 78.000 .003 .928 a. Exact statistic b. Computed using alpha = .05 c. The statistic is an upper bound on F that yields a lower bound on the significance level. d. Design: Intercept + READING
154 Table 4 9 Tests of b etween s ubjects e ffects Source Dependent v ariable Type III sum of s quares df Mean s quare F Sig. Observed p ower b Reading Academic v ocabulary .139 2 .069 1.799 .172 .366 Expanded n oun .039 2 .020 1.269 .287 .268 Embedded c lause .001 2 .001 .169 .844 .075 Nominalization .030 2 .015 7.476 .001 .935 Lexical d ensity 7.650 2 3.825 5.569 .005 .843 a. R Squared = .043 (Adjusted R Squared = .019) b. Computed using alpha = .05 c. R Squared = .030 (Adjusted R Squared = .006) d. R Squared = .004 (Adjusted R Squared = .020) e. R Squared = .156 (Adjusted R Squared = .135) f. R Squared = .121 (Adjusted R Squared = .099) Table 4 10 Levene's t est of e quality of e rror v ariances F df1 df2 Sig. Academic v ocabulary .878 2 81 .420 Expanded n oun .433 2 81 .650 Embedded c lause .580 2 81 .562 Nominalization 6.576 2 81 .002 Lexical d ensity .189 2 81 .829 Tests the null hypothesis that the error variance of the dependent variable is equal across groups. a. Design: Intercept + READING
155 Table 4 11 Multiple c omparisons ( Scheffe ) Dependent v ariable Reading Reading Mean d ifference (I J) Std. e rror Sig. Lexical d ensity 1 2 .522 .222 .068 3 .724 .224 .007 2 1 .522 .222 .068 3 .202 .220 .658 3 1 .724 .224 .007 2 .202 .220 .658 Nominalization 1 2 .045 .014 .008 3 .063 .014 .000 2 1 .045 .014 .008 3 .018 .014 .457 3 1 .063 .014 .000 2 .018 .014 .457 Based on observed means. The error term is Mean Square(Error) = .003. *. The mean difference is significant at the .05 level. Table 4 12 Compare means of five academic language features Writing Academic v ocabulary Lexical d ensity Expanded noun phrase Nominalization Embedded c lause 1 Mean .378 4.260 .191 .025 .093 N =28 2 Mean .406 4.816 .255 .035 .103 N =29 3 Mean 454 5.538 .279 074 .121 N =27 Total Mean .412 4.863 .241 .044 .106 N =84
156 Table 4 13 ANOVA Dependent variable Sum of s quares df Mean s quare F Sig. Academic v ocabulary Between Groups .082 2 .041 1.041 .358 Within Groups 3.179 81 .039 Total 3.261 83 Lexical d ensity Between Groups 22.517 2 11.259 22.369 .000 Within Groups 40.769 81 .503 Total 63.287 83 Expanded n oun Between Groups .116 2 .058 3.954 .023 Within Groups 1.183 81 .015 Total 1.299 83 Embedded c lause Between Groups .010 2 .005 1.554 .218 Within Groups .271 81 .003 Total .282 83 Nominalization Between Groups .119 2 .060 29.397 .000 Within Groups .165 81 .002 Total .284 83 Table 4 14 Variables entered/r emoved Model Variables e ntered Variables r emoved Method 1 Nominalization Stepwise (Criteria: Probability of F to enter <= .050, Probability of F to remove >= .100). 2 Density Stepwise (Criteria: Probability of F to enter <= .050, Probability of F to remove >= .100). 3 Voca bulary Stepwise (Criteria: Probability of F to enter <= .050, Probability of F to remove >= .100).
157 Table 4 15 Model summary Model R R s quare Adjusted R s quare Change Statistics R square c hange F c hange df1 df2 Sig. F Change 1 .657 a .431 .424 .431 62.133 1 82 .000 2 .732 b .535 .524 .104 18.192 1 81 .000 3 .751 c .564 .548 .029 5.258 1 80 .024 a. Predictors: (Constant), Nominalization b. Predicto rs: (Constant), Nominalization, Density
158 CHAPTER 5 DISCUSSION AND CONCL USION Based on the linguistic analysis and statistical finding s presented in Chapter 4, Chapter 5 presents discussions of findings and these discussions are contextualized in the previous literature review. Implications of future research and practice will also be considered. Discussion of Findings Adolescents' Use of Academic Language The statistical findings in Chapter 4 suggest th at the ninth graders made limited use of academic language in the ir historical writing. In this section, findings for e ach academic language feature are discussed in light of the data collection context and the previous research literat ure Academic v ocabu lary For academic vocabulary, the student participants use d academic vocabulary infrequently In order to understand this low occurrence of academic vocabulary, it is important to understand how academic vocabulary is conceptualized in terms of tiers or categories. As discussed in Chapter 2, a commonly accepted classification system frames academ ic vocabulary according to three tiers (Beck, et al. 2002; CalderÂ—n et al. 2005): non academic, conversational vocabulary (Tier 1); general academic words (Tier 2); and content specific, technical vocabulary (Tier 3). Tier 2 words, or general vocabulary are considered the most important to teach (Snow, 2008). As mentioned earlier, the Academic Wor d List compiled by Coxhead (2000 ) consists of 570 word families that occur reasonably frequently over a very wide range of acade mic texts, which is the Tier 2 a cademic vocabulary. In the corpus of essays, students utilized
159 many non academic, conversational vocabulary words rather than general academic words, which resulted in a more interactional registers rather than formal academic registers. In the corpus of essays, the limited use of academic vocabulary not only showed in the actual number of academic vocabulary words, but also the repeated use of certain academic vocabulary words. The results revealed that students used a number of academic vocabulary words repeatedly. Words such as influence can be seen almost in every essay, which was a result of instruction regarding the content in this academic writing. The ability to use a diverse selection of academic vocabulary words is a very important factor of acad emic writing quality (McNamara et al., 2010) The students in the present study did not exhibit sufficient abilities in this regard. Secondary students usually have the underlying conceptual understanding of what these words present but lack the knowledge and ability to select the most precise word to represent the concepts they already know (McNamara et al., 2010). The relatively low use of general academic words in the current study correspond s to Snow (2008) wh o advocates more teaching of Ti e r 2 general academic vocabulary. These findings clearly confirm the challenge of Tier 2 academic vocabulary to students and indicate a need for more attention to these words in content area literacy instruction. Expanded noun phrase In the corpus of essays, the feat ure of expanded noun phrase was prevalent and all students used it in their writing. Research has pointed out that in academic register, expanded noun phrase s are frequently used es pecially in high quality writing In Fang's (2008b) analysis of science wr iting, complex nouns constitute approximately half of all the nouns in high quality essays. Secondary students used more expanded noun
160 phrase s compared to students of lower grade levels (Fang, 2008b). In my study, since students were 9 th graders, it is not surprising to see the frequent use of expanded noun phrase s in their academic writing. According to the study conducted by de Oliveira (2010), expanded nouns are used to package information, expand explanations, and structure reasoning in textbooks. We ca n see many examples of expanded nouns that served as these functions in the corpus of essays. Since we do not have a reference group to compare the average use of expanded noun phrase s, it is more interesting and meaningful to see the distinct use of thi s feature by different reading groups, which will be discussed in the upcoming section that addresses the second research question. Embedded c lause E mbedded clause plays an important role in academic language It could constitute and reflect the structure of the text (Schleppegrell & Colombi, 199 6 ). In historical writing, the use of embedded clause can build causality within the clauses and contribute to the abstraction ( Christie & Derewianka, 2008 ). In the corpus of essays, we could see students used t he feature of embedded clause frequently. Almost every student used this feature in their historical writing. Students used this feature in their writing to realize experiential interpersonal, and textual meanings. Embedded clause was deployed to construct c omplex nominal phrases, dense sentence structures and a well planned and tightly constructed essay. For adolescents, the use of embedded clause is an especially important feature in their historical writing. Colombi (2002) studied bilingual students' Spani sh writing and her finding reveals that students used more embedded clauses when their writing was developed toward more academic styles. She also stated the use of embedded clauses
161 is crucial to differences between interactional register and academic regi ster (Colombi, 2002). In spoken language, clauses are used and chained one after another, often resulting in very long sentences. Academic language, on the other hand, packs a large number of lexical items in one clause, resulting in more uses of embedded clauses. Christie and Derewiank a ( 2008 ) analyzed adolescents' historical writing and suggest that, at this stage, students started having less grammatically intricate structure, which means less hierarchical levels of clauses and more embedded clauses are used. The findings in this study confirmed the dynamic developmental nature in using the feature of embedded clauses in adolescents' historical writing. Whereas many students used embedded clauses to construct a well planned and tightly constructed essay some other students used more of other clauses such as hypotactic and paratactic clauses and more dependent structure in their writing. The former is more prone to the characteristic of successful historical writing at late adolescence, while the latter sh ows the feature of early adolescents' historical writing ( Christie & Derewianka, 2008 ). What was missing in the previous literature was the use of embedded clauses to realize interpersonal and textual meanings. In this study, as described in Chapter 4, mo re than 10 students used embedded clauses as a tool to construe textual and interpersonal meaning such as the next thing I am going to describe or the reason you may know. Embedded clauses of this kind are usually seen in interactional registers In th e corpus of essays, they actually realized an interactional style as if there was an immediate audience, which contributed to both the grammatical intricacy and lexical complexity (by adding more non embedded clauses and meaning that was irrelevant to the focus in the sentences) but not academic ness of language in academic
162 writing. This use of embedded clauses suggests that students still drew on the asset of oral language to construe a more interactional style in their academic writing. For adolescent wri ters, one of the most important functions of embedded clause in historical writing is to build connections within clauses through abstraction ( Christie & Derewianka, 2008 ). In this way, the clause structure is grammatically simple but logical relations can be even more intricate. However, the embedded clause is always used to realize this function in adolescents' essays, which must be noticed when we analyze the role of embedded clauses in academic writing. Nominalization As the examples in Chapter 4 show, students used nominalizations to build general participants to replace actual events and actions. On the basis of this, definition, categorization, and comparison can be developed. Students also utilized nominalization to distill information, build embedd ed perspective, and create the flow of the text. In the present corpus of essays, nominalizations were used to construct density of information, abstractness and causality. Students commonly packed series of events into a single noun or a noun phrase as th e starting point of a clause, which contributed to the flow of the text. In this way, further explanation could be created and reasoning could be structured. These findings support the role of nominalization in academic language. As the literature suggests (Christie & Derewianka, 2008 ; Eg gins, 2004; Schleppgegrell, 2004 ), nominalization is the key resource of grammatical abstractness and specialized knowledge in secondary content areas. On average, o nly six nominalizations were used per 100 clauses in the c orpus of essays, which was the lowest use among all academic language features in the present study. Nominalization was used much more infrequently than other academic language
163 features such as expanded noun phrase in students' essays. In the previous subs ection, we could see that all students used expanded noun phrase s in their writing, suggesting expanded noun phrase was a relatively familiar academic language feature for these secondary students. To the contrary, nominalization was a much more challengin g feature to the adolescents and not all of them were adept in using it. Students actually exhibited very distinct abilities in using this feature. Whereas one fourth of the students did not use any nominalization in their writing, about one fourth of the students used more than 10 nominalizations accurately in their historical writing The limited uses of nominalization in the current students correspond to the existing literature that states nominalization is an important but challenging feature for adol escents (Christie & Derewianka, 2008). Nominalization is regarded as the most important feature of students' academic writing development from an interactional style to academic registers (Colombi, 2002). The feature of nominalization signals the onset of adolescence in literacy development (Christie & Derewianka, 2008). In the analysis of historical writing by Christie and Derewianka (2008), students at mid adolescence (13 15 years) start producing abstract nominalizations. At late adolescence (16 18 years), students utilize nominalization more and more as participants and circumstances to contribute to abstraction in their writing. As we may see in the writing samples, there were students who used nominalization successfully in their historical writing to construct causal accounts of events and build interpretations and arguments based on historical events and artifacts, whereas some other students did not use this feature at all. If we lo ok at this phenomenon from a developmental perspective, ninth grade is a
164 stage where students learn to use the feature precisely in their writing. This may explain the internal diversity of using the feature of nominalization in the corpus of essays. As s tudents advance through the grades, they are expected to use nominalizat ions in their writing to demonstrate that they understand the more abstract concepts in these subjects and also express the concepts in non congruent language with abstraction, general ization, judgments and opinions. The challenging nature of nominalization best represents the demands of increasingly technical and abstract historical language in high school when many struggling learners are left behind (Christie & Derewianka, 2008). Due to its challenging nature, the scarcity of nominalization in the corpus of essays is understandable, which also calls for more attention to this feature. Lexical d ensity The lexical density suggests the use of lexical items and grammatical items. Lexical items are the content words including nouns, adjectives, verbs, and some adverbs (Halliday, 1985). The grammatical items include articles such as the and a propositions s uch as in and on pronouns such as him auxiliary verbs such as had conducted conjunctions and and demonstratives this In conversation, we use fewer lexical items and more grammatical items. But in academic writing, more lexical items must be included b ecause we need to express the meaning through the language itself (Christie & Derewianka, 2008) Overall, students in the present study used a relatively low rate of lexical items. Many students frequently included pronouns and demonstratives rather than g eneral nouns, which not only resulted in an interactional register, but also lowered down the lexical density.
165 Halliday (1985) points out that written language has a higher lexical density than speaking and lexical density accounts for modality difference s (between oral and written language) and developmental differences (between younger writers and older writers) According to Halliday and Matthiessen (2004), most academic writing in subject areas demonstrates lexical density values of six or higher In t he present study, an average lexical density at 4.86 was found in the corpus of essays. The highest lexical density was 8, whereas the lowest one wa s 3.25. A lexical density at 8 wa s a high value of lexical density and typical of academic writing. And a de nsity at 3.25 represents interactional styles. On average the lexical density of 4.86 was not a high value of lexical density and indicated students' academ ic writing did not distinguish itself quite well from modes of non academic genres or interactional register. It is even more interesting to compare the average lexical density in the present study with another analysis of adolescents' historical writing ( Christie & Derewianka, 2008 ). In thei r study, the lexical density in the genres of explaining and arguing exhibit a movement from 5 to between 6 and 7. This further confirmed the low use of lexical items in the present corpus of essays. Through the analysis of academic language features and their functions, we can see students used all features in their writing. Not all the features served successfully the functions of academic register though. Sometimes, students used features that construed interactional styles. It must be noted that all t he analysis took place in the context of historical writing. Students used the academic language features to realize a style that was valued in conveying historical meanings. For example, it was highly valued in historical writing that students used nomina lization to construe causality within
166 the clause. Therefore, nominalization is a prominent feature for adolescents in history class. This acknowledgement of context is even more important when we discuss the connection between academic language use and rea ding and writing abilities. Academic Language Features and Reading Abilities A significant multivariate effect was fo und between reading ability and academic language features. Among all the features, there was significant main effects between nominaliz ation and lexical density and reading ability. No significant univariate effect was found for other academic language features including expanded noun phrase embedded clause, and academic vocabulary. We usually assume that better readers have better comm and of academic languag e features. The findings in my study confirmed this assumption part ia lly and suggested overall better readers significantly used academic language features in their writing. This corresponds to the descriptive analyses of three essay s in Chapter 4. Better readers such as the student who wrote Essay 3 exhibited more frequent, diverse and precise uses of academic vocabulary words more expanded noun phrase s and more accurate use of embedded clauses that contribute to the experiential m eaning in their writing Lexical density indicates to the number of lexical items and a great extent the academic ness of any essay. Better readers' essays exhibited higher lexical density, which also confirmed their overall better command of academic lang uage. Better readers in the seconda ry school setting in the present study em ployed more nominalizations and their essays exhibited higher lexical densities. Compared to other features, nominalization is a feature characteristic of adolescent literacy deve lopment (Christie & Derew ianka, 2008). At this point of ninth grade, students' abilities of using nominalization to construct academic ness in their writing diversify.
167 Better readers are more motivated to read and have more reading experience in expository text which provide them with more knowledge of academic language and higher proficiency in using challenging features of disciplinary learning such as nominalization. Therefore, they are expected to show better uses of nominalization in their writing, as in the present study. In terms of academic vocabulary, better readers are assumed to possess content knowledge and thus more academic vocabulary They would be assumed to possess more diverse range of words, greater working memory capacity or greater ski ll and knowledge that facilitates them in retrieving their academic vocabulary during writing (Just & Carpenter, 1992; Raynor & Pollatsek, 1994). Greater use of academic vocabulary in speech or writing is commonly thought to reflect reading abilities, ling uistic skills, or even a speaker's socioeconomic status (McNamara et al. 2010). Therefore, the use of academic vocabulary would be significantly different across differen t reading groups. However, an non significant relationship between academic vocabular y use and reading ability suggests that students of higher reading abilities did not necessarily use more academic vocabulary compared with students of lower reading abilities. A possible explanation is that academic vocabulary words of Tier 2 do not diff er according to reading abilities. For this group of adolescents their command of Tier 2 academic vocabulary was similarly low and therefore did not show significant differences among different reading groups. In content areas, secondary students usuall y understand the conceptualization that academic vocabulary presents but are unable to find an appropriate word to repres ent it. This explanation again corresponds to Snow's (2008) argument about the needs of teaching Tier 2 academic vocabulary.
168 Another ex planation is because the list of academic vocabulary words used by these students was so confined, it was not a valid representation of students' reading abilities. As we could see, 75% of the occurrences of academic vocabulary used by the students were l imited to repeated uses of 20 words. These words were the focus of the lessons regarding Roman and Greek culture content that students were learning in their world history classes Through these uses of academic vocabulary, we could hardly understand the role of reading abilities in this phenomenon. It implied that most students immediately applied the words that they learned rather than empl oying their knowledge of academic vocabu lary to construct a well written historical essay. Most importantly, this insignificant relationship points to the unreliability of equating developing academic vocabulary (especially Tier 2) with developing reading and writing in academic language. Since lower reading groups do not necessarily show more limited vocabulary knowledge, the reciprocal relationship between reading ability and academic vocabulary must be examined with scrutiny. For the feature of embedded clause, students of high reading abilit ies used even fewer embedded clauses compared with their peers of average reading abilities. The analysis of embedded clauses in the corpus of text s shed light on this insignificant main effect between reading ability and the number of embedded clauses. Wh ereas embedded clause is an important feature of academic language (Colombi, 2002), the finding in my study demonstrated that most of these secondary students accumulated a great deal of experience with embedded clauses The functions o f these embedded cla uses should actually merit more attention. In academic writing, students must know how to use embedded clause to create lexically condense clauses and concisely
169 describe and define concepts. When embedded clause was used to contribute to the textual meanin g and realize oral register, it should not be counted as a feature of academic language. In my study, the capabilities of using expanded noun phrase phrases in academic writing did not show close connection to students' reading abilities. Neither did stud ents' use of expanded noun phrase phrases differ ed significantly across different reading groups. Although there was no significant connection, it does not mean that expanded noun phrase is not an i mportant feature in academic language. An explanation of t he insignificant association between expanded noun phrase and reading ability is that this group of high school students was already familiar with the use of expanded noun phrases. Therefore, even the low reading groups could employ the features relatively successfully. It must be noted that the high reading group used more expanded noun phrase s compared with the average reading group, and the average reading group also used more expanded noun phrase s than the low reading group, although these differences w ere not statistically significant. Academic Language Features and Writing Qualities The finding in Chapter 4 clearly demonstrates that, significantly more use of nominalization and expanded noun phrase could be found in essays of higher qualities, compa red to lower quality essays. T he finding also showed that higher quality essays had significantly higher lexical densities and the lexical densities significantly had a significant impact on the quality of the essays. The feature that plays the most predi ctive role in writing quality was nominalization. The use of nominalizations predicted significantly over 40% of the variances in essay qualities in the presen t study. The finding regarding nominalization
170 corresponds to the previous research that the use o f nominalization is a key feature of historical writing (Coffin, 2006a) The frequent occurrences of nominalization increased the lexical density by reducing the number of non embedded clauses, thus contributing to the high quality in essay writing. The us e of nominalization also packed more information in fewer clauses, which resulted in an intact structure in the academic writing. Because of the important role of nominalization in predicting an abstract and formal writing style, Coffin (2006b) argues tha t nominalization is a dominant feature in the arguing and explaining genres in history writi ng. From the discourse analysis in my study we could also see that a high quality essay em ployed many nominalizations and these nominalizations played an essential role in the abstractness and infused perspective, which are highly valued by historians and the language of history. The important role of nominalization in predicting writing quality also confirmed with the previous literature stating that nominalizatio n is one of the most important indicators of a more formal and abstract writing style ( Christie, 2002; Christie & Derewianka, 2008). Nominalization can achieve linguistic complexity, abstraction and technicality in academic writing, which is also characte ristic of secondary academic learning. In the context of historical writing, nominalization realized causality in a subtle and implicit way. By using nominalization, students managed to represent a series of events and cause effect relationship occurs with in clauses rather than using conjunctions such as because to represent causality across clauses. Both this implicit causality and intact structure contributed to a formal and abstract writing style and a higher writing quality.
171 This analysis of nominalization suggests the uniqueness of historical discourse. Based on Coffin (2006b), in historical discourse, language features are used to construe causality in an abstract way in academic reading and writing. Successful writing in an expository text must demonstrate effective use of nominalization and abstract causality. Lexical density, as a fundamental feature of the text organization and syntactic structure, was considered to have close connection with writing qualities as well. I t account ed for 10% of the variances in writing qualities. This kind of connection between writing quality and lexical density was primarily based on the dense organization of lexical items in students' writing. As McNamara et al. (2010) stated, proficient writers are assumed to have the ability to write more complex sentences because proficient writer may have either greater working memory capacity or more knowledge of syntactic structures. Thus, proficient writers would be expected to have the capacity to write in more complex or sophisticated language, resulting in high lexical density. They are expected use more embedded clauses, resulting in fewer noun embedded clauses and higher lexical density. Lexical density is also a result of multiple features an d functions including academic vocabulary, embedded clauses, expanded noun phrase s, and nominalization. Embedded clause cannot explain fully the complexity and, furthermore, quality of the essays, as discussed before. However, lexical density can significa ntly impact essay quality because lexical density takes into account the content words in the computation, and also accounts for the quality of embedded clauses. Because embedded clauses that contribute to the experiential meaning comprise more content wor ds than those that construe textual or interpersonal meanings, high lexical density can be found in
172 academic writing in which language is more complex and sophisticated lexical items are tightly packed. On the other hand, the impact of lexical density on writing quality was much lower compared to the feature of nominalization. If we want the lexical density to have a more significant impact on essay quality, it is critical to consider lexical richness (the range of students' vocabulary use) on the basis of lexical density. Recent studies (e.g., McNamara et al. 2010; Johansso n, 2008) examined lexical diversity in their measurement of lexical development. Johansson (2008) concludes that lexical diversity accounts for more grade level differences and is a bet ter measure to use for detecting differences between age and grade levels when the richness of lexical items (content words) are also considered. This also connects to the feature of academic vocabulary. Although academic vocabulary predicted little writin g quality in the corpus, the results could be changed if we examined the richness and diversity of academic vocabulary instead, which will be elaborated later. In terms of academic vocabulary, there was no significant difference in using academic vocabulary for three writing quality groups. However, according to the results of multiple regression statistics, academic vocabulary had significant impact over the variances of essay quality but it only predicted 2% of the differences. By comparing the m agnitude of the coefficients, we can see the effect of academic vocabulary in predicting essay quality was only lower than lexical density and nominalization but higher than all other variables ( .262). However, the presence of academic vocabulary seemed d ecreasing the essay quality
173 The finding suggests that a simple count of academic vocabulary occurrences has little to do with writing quality. Other features such as the diversity of academic vocabulary may be more important. The study of McNamara et al. (2010) examined the relationship between writing quality and linguistic indices including the diversity of words used by the student authors. In their finding s lexical diversity show significant differences as a function of essay quality, or in other word s, the greater lexical diversity, the higher the essay quality. Essay quality was judged largely by the sophistication of the writing including more complex sentences, less frequent words, and a greater diversity of words. In the current study, over 75% of the occurrences of academic vocabulary words used by student writers were limited to the repeated uses of a list of twenty words. More repeated uses of certain academic vocabulary words implied more frequent words and a lower diversity of words, thus resu lting in lower essay qualities. Statistical findings in my study did not show a significant effect for the feature of embedded clause, although Co lombi (2002) and Schleppegrell and Colombi (1997) both found the connection between the use of embedded claus e and writing qualities in their case studies. As we discussed in the literature review, a well developed academic writing relies on the use of embedded clauses to achieve grammatical condensation, whereas a more emergent organizational structure needs mor e exte nsion and elaboration of ideas. An interactional register usually uses more paratactic and hypotactic clauses. Studies demonstrate that when students develop their academic language, their writing moves from more oral to more written styles that are characterized by reduced clausal structures (e.g., Colombi, 2002; Crowhust, 1990; Menyuk, 1988) and fewer parataxis and hypotaxis ( Colombi, 2002 ). Clauses combining
174 strategies and the use of embedded clauses have been used as one of the most important indi cators by researchers when exploring students development trend of academic literacy (e.g., Colombi, 2002). The use of embedded clause could not predict writing quality in the presen t study. A s we saw in the last subsection, when examining the connection between students' use of embedded clauses and reading ability no significant difference was found either. To explain this phenomenon, we must consider that students' use of embedded clauses plays different roles in their writing. In the corpus of the essa ys, many students used the embedded clauses to realize textual and interpersonal meaning such as the reason I think This type of embedded clause did not contribute to a chain of reasoning or a condense structure. Since the study of Schleppegrell and Colom bi (1997) did not list any embedded clause of this kind, it's understandable that they found the connection between the use of embedded clause and the academic ness of students' writing. The finding in this study suggests that when we use embedded clause a s a characteristic of academic language development, we must also examine the meanings that these embedded clauses construct and the functions they realize Another feature that was irrelevant to writing quality was expanded noun phrase s. Previous litera ture has found high quality texts use more complex nouns than low quality texts and that more complex nouns are used in texts of higher grade levels than those in texts of lower grade levels (Fang, 2008b). But that's in the context of science writing and the definition of complex noun was slightly different from the definition of expanded noun phrase in my study. My study may offer more insights into the role of expanded noun phrase s in wri ting quality in th e context of ninth grade history
175 classrooms. Most of the ninth grade students exhibited knowledge of this feature. T he frequency of using it did not really affect writing quality. A s all groups of students' use in expanded noun phrase s could fulfill teache rs' expectation, expanded noun phrase 's role in p redicting writing quality would be non significant. Moreover, the analysis in Chapter 4 demonstrated there were different types of post modifi ers. A post modifier could be (a) a noun connected by a prepositi on ( the most diverse things [about the two]), (b) a nominalization connected by a preposition (differences and similarities [of equal depth]), or (c) an embedded clause ( tragedies and comedies [that flourished Greek entertainment] ). Among these expanded no un phrases, (b) and (c ) are much more abstract and dense than (a) and would naturally be valued more in the context of historical writing. Therefore if we examined the number of post modifier (b ) and (c ), we might get different results because they were r eally contributing to the academic ness in the writing. Last but not least, the findings regarding the association between academic language features and writing quality also shed light on the issues of writing development. According to the outlines of Ch ristie and Derewianka (2008) that show the de velopmental changes in historical writing adolescent students encounter more and more uncommonsense knowledge that is construed in non congruent grammatical resources. During the late adolescence (9 th 12 th gra de), students need to deploy grammatical resources to express abstraction and generalization. Dense expanded noun phrase nominalization, embedded clause and lexically condense clause structure are all included as features that are characteristic of the ad olescence. Among all these
176 features nominalization may still be the feature that is valued the most in the language of high school history. Issues of Interest Based on the findings in Chapter 4 and discussions in Chapter 5, there are a number of partic ularly interesting and noteworthy issues. These issues emerged in the discussions and may also lead to the implications in my study. 1. Students in this study did not exhibit a proficient mastery of academic language features. 2. Academic language features are complex. Even a single feature has a lot of internal variances in terms of types and functions. 3. N ominalization and lexical density are particularly important features for adolescents 4. Better readers show better command of some academic language features b ut not others, although reading ability has an overall significant effect on the use of these features. 5. Academic language features have a significant impact on teachers' evaluations of writing qualities. Limitations and Implications Limitation s I t is important to recognize that the definition of writing quality used in my study rested on human judgments by a language art teacher and a history teacher. They were both a part of academic communities in secondary schools who knew little about the linguisti c features that interest us. They were skilled readers who had read numerous essays on the same topic and format. They were also trained to reliably using a
177 standardized rubric. Other communities or cultures, such as historians, may have different perspect ives on what constitutes good writing. For instance, historians may value causality in historical discourse that is constructed using nouns and verbs rather than conjunctions. For historians, the use of nominalization may have a more significant impact on writing quality. An additional concern is that the student s in this study were responding to the particular demands of the task at hand. The limited time f or writing essays, the limited length of the essays, and the pre established topic (meaning students were writing based on the same type of textbook portion and lecture) all might have inhibited them from fully demonstrating their knowledge of academic language. There are other limitations in the present study. In order to examine students' academic lang uage proficiency, students' academic writing was analyzed and discussed. However, only one writing sample written in one class period was collected. Although the expository essay that student wrote in the present study was the most representative genre in secondary history classrooms, we would obtain a more comprehensive picture if we could take into consideration more writing tasks and more genres. Meanwhile, when analyzing the writing samples, data analysis protocols that focus on a certain number of acad emic features were utilized. I was unable to analyze all the academic language features in the essays, which was also a limitation in the present study. Overall such potential limitatio ns do not negate the findings in my study. Rather, we might be able t o find more detailed analysis of linguistic features and then highlight the connection between these language features with academic writing. This study
178 progresses an important step toward identifying and evaluating the roles of lexical and grammatical fea tures and their impact on quality writing and reading development in secondary history classrooms. Implications for Practice The analysis of student writing in the academic subject such as history offers teacher and cu rriculum designer insights regarding how to help students develop control of academic language in the context of secondary history learning Building awareness My findings indicate a need for more attention to academic language features for adolescents in content area literacy instruction. The awareness and understanding of the linguistic challenges should be specific to different lexical and grammatical features. In the finding, we can see students' limited uses of academic language features and the scarc ity of nominalization, a key academic language feature. Students must be taught explicitly abou t these features and their functions in academic writing. The relatively low lexical density in students' writing also indicates a need for training students to use more lexical items, arrange the lexical items in a cohesive way, and writing to the academic genre and register. To better prepare our students for content area learning, it is also crucial to stress the importance of all academic language features to all students. Since better readers do not necessarily have better command of academic language features, instructional practice to teach academic language should not be limited to ELLs or struggling readers.
179 Features that must be emphasized Whereas acade mic language features are closely related to reading ability and writing quality, it is important to consider each feature with more specificity. For example, when we examine academic vocabulary, the roles of general academic vocabulary and content specifi c vocabulary are different. Whereas general academic vocabulary does not play a significant role in students' writing quality, the content specific vocabulary may receive more attention from disciplinary specialist and thus play a more important role in th eir evaluation. In terms of expanded noun phrase s, different types of post modifiers play different roles in academic writing as well. A simple noun connection using a preposition does not necessarily make a noun phrase more complex and abstract. An embedd ed clause serves a more important role in a dense construction of experiential meaning and a higher lexical density. The significance of different academic language features varies according to grade levels. In high school setting, most students already have a basic functional knowledge of features including expanded noun phrase s, academic vocabulary, and embedded clause which could be seen in the results of the present study Nominalization seems more challenging but it is particularly important because students start building their knowledge of this feature in high school content area learning. This leads to the discussion regarding the different roles of these academic language features for a certain stage. As Fitzgerald and Shanahan (2000) suggest at the high school level, it is pivotal for students to have the critical features such as complicated syntactical structures and organizational structures within texts. These constitute a necessary step for students to be able to comprehend and create compl ex text in secondary content areas. The mastery of complicated syntactical and organizational
180 structures, in the context of reading and writing in history, relies heavily on the deployment of nominalization in a complex and abstract text with high lexical density. Knowledge of nominalization to construct dense chunk of text must be emphasized in high school years. The acknowledged significance of academic language features and their connection to reading ability and writing quality furthermore allow us to d iscuss what lexical and grammatical features must be taught in the classrooms, given the limited time in content area classrooms and wide range of ways in which language is used in academic settings (B unch, 2006). In the context of nin th grade history clas sroom, students must be taught to use nominalization proficiently to construct an abstract, dense and complex structure in history. They should learn how to pack more lexical items and content words into clauses that are structured into a hierarchical way. Most importantly, when reading in history, they have to learn how historians use these language functions realize meaning in this content area. This need for more awareness and better knowledge of academic language features can be addressed through a func tional approach to language (e.g., Fang & Schleppegrell, 2008), one that focuses not only on grammatical forms but also the functions these forms realize in disciplinary meaning making. The effect of using a functional approach on academic language develop ment and content area learning has been documented by a number of studies (e.g. Schleppegrell & de Oliveira, 2006; Aguirre MuÂ–oz el al, 2006a, 2006b; Hammond, 2006; Spycher, 2007; Huang, 2004; Echevarria et al. 2006). The key idea is to explicitly teach s tudents features that are representative of academic language and specific content area learning. In terms of
181 successful models of using academic language in history. Analysis of history textbooks can show students successful uses of academic language features. Students must also be taught explicitly to fulfill the requirements of an academic essay in content areas such as history. Based on the importance of nominalization, students must be taught to interpret and create this feature in their historical reading and writing. Strategies such as teaching students to convert between interactional register and academic register can help s tudents understand the functions of nominalization. Suggestion s for preparing teachers to teach academic language The need for better instructional practice naturally leads to the issue of teachers' professional development. Due to the complexity and vari ety of academic language features, we cannot expect content area teachers be able to teach these features automatically. As Coffin (2006a) argues, teachers must be provided with a framework for professional development in which a collaborative approach and a partnership between educational linguists, literacy consultants, and history teachers is essential. Changing teachers' attitudes toward academic language is the first crucial step to ensure any long term impact on students' academic language development (Coffin, 2006 a ). Lesley et al. (2007) proposes that preservice teacher s' attitudes and reading behaviors must be addressed in a content area literacy course and preservice teachers must develop improved concepts about academic reading and a willingness to explore multiple texts. In order to promote effective instruction in secondary content area classrooms, teachers should be offered opportunities to re conceptualize their notions of content area literacy practice in school settings (Stevens, 2002). A good knowledge of the relationship between academic language and content area learning could improve
182 teachers' awareness to teaching previously invisible aspects of texts and subsequently develop ing students' control of academic language (Coffin, 2006 a ). Teach ers must understand the specialized demands of academic language and linguistic features specific to disciplinary areas. They should be prepared to directly and explicitly address the specific and highly specialized disciplinary reading demands of content are as and teacher training programs must incorporate the grammatical and lexical features that construct disciplinary knowledge (e.g., de Oliveira, 2010; Schleppegrell & de Oliveira, 2006). Teacher education must take the role of preparing teachers with we ll established theory and application in academic language development. Professional development can build positive reading and writing attitude, successful reading and writing behaviors, and knowledge of academic language features in various academic disc ipline s Implication s for Research First of all, the academic language development of mainstream students in content area classrooms should receive more attention. Whereas most empirical studies included ELLs as their target population in developing acade mic language, the results in the present study suggest, mainstream students, especially students of low reading abilities, exhibit a low use of all academic language features and thus demand more attention in their academic language development. Therefore the research with respect to academic language development should not be limited to ELLs. A ny research studying academic language development must adopt a com prehensive understanding of the complex nature of academic language A narrow conceptualization of academic language as any feature alone such as academic vocabulary limits the effectiveness of academic language instruction. In the current
183 study, academic vocabulary and embedded clause that traditionally receive most of the attention in academic lang uage research cannot explain very well the differences in writing qualities. Any feature alone is not sufficient as complex grammatical structures and discourse patterns together determine the writing quality (Carr, Sexton, & Lagunoff, 2006; Zwiers, 2008; Bruna, et al., 2007). But for different age groups and in a certain subject area some feature may be more important and challenging than other features. Although the present study did not study teachers' attitudes and knowledge directly, t he results may offer some background to understand teachers' attitudes and knowledge. Empirical studies have been conducted to explore teachers' knowledge and expertise in academic language and its influence on classroom instruction (e.g. Bruna et al. 2007; Spycher, 200 7; Bailey et al., 2007). Although the data in this study say little with respect to teachers' attitudes, the results revealed the close connec tion between academic language use and teachers' evaluation of writing quality. These results implied that teacher s hold linguistic expectation on students' academic writing and their accurate use of academic language features which corresponds to the previous studies regarding teachers' expectation on academic writing in content area learning ( de Oliveira, 2011; Agu irre MuÂ–oz el al, 2006a, Zwiers, 2007). The question that remains to be answered is the progress students may make if their teachers specify and elaborate their expectation before the academic writing. Future research can definitely explore further to addr ess this question. T hrough this study, we understand the impact of different uses of academic language features on writing quality. The academic language features in the current study point to the linguistic complexity, abstraction and technicality in aca demic
184 language of secondary schooling (Christie, 2002; Christie & Derewianka, 2008). Academic language development over the years of schooling presents new challenges at each level and in each new subject area For the academic language in ninth grade history classroom, nominalization that construes abstractness and technicality was the most challenging feature. In future work, we must consider to what degree specific intervention of a certain linguistic feature will benefit the adolescents. A co mprehensive linguistic model that can predict an essay's writing quality by examining the use of various academic language features is especially desirable. Last but not least, for future research, it is important to recognize the internal complexity in us ing and understanding the grammatical and lexical features of academic language. Whereas quantitative study of language feature can give us a large scale analysis, it is important to acknowledge the complex nature of each feature. For example, future resea rch will need to examine the relationship between content specific (history) vocabulary and reading ability as well as writing quality. Also as we could see in the finding part, whereas embedded clause is regarded as a significant index of academic languag e, at times students use embedded clauses to realize an interactional style in their writing. Future research should explore for example, the connection between "interactional" embedded clause and the "academic ness" of their essays. Without understanding the intricate nature of every academic language feature, we cannot accurately understand their connection to students' academic language development. Overall, by focusing upon secondary students' academic writing in world h istory classrooms, this study w as designed to lead to further discussion regarding
185 adolescents' academic language development. A dopting S ystemic Linguistic Analysis, the present study explored ninth graders' use of academic language in their historical writing The finding suggests that students used various academic language features differently in their historical writing. Nominalization was the most challenging feature based on its fewest occurrences in the corpus of essays. The results of this study indicate that more proficient read ers used academic language features more successfully These findings also suggest the close association between academic language use and the writing qualities evaluated by classroom teacher. This study contributes to the ongoing discussion regarding con ceptualization and instruction of academic language in secondary school content areas (history, in this particular study). Whereas academic vocabulary and comprehension strategies have received more attention in teaching participants to read and write in c ontent areas, this study raises the awareness of teachers and student to the demanding nature and essential role of academic language features. The knowledge of a variet y of academic language features can also provide a framework for classroom teachers to effectively incorporate grammatical features into reading and writing instruction. The finding reveals the crucial role of nomina lization for adolescent writers For students of different reading abilities, the most conspicuous differences lie in the use of nominalization and lexical density rather than academic vocabulary. It suggests that a common sense of weakness in academic vocabulary for low reading student s may not play a decisive role in academic language development. Furthermore, Students' needs in developing academic writing should be addressed by developing students' lexical richness in the context of reading and writing rather than solely teaching acad emic
186 vocabular y and focusing on the quantity of these academic vocabulary words In addition, in the corpus of the text, students' frequent use of expanded noun phrases is noteworthy. The insignificant connection between writing quality and embedded clause suggests that the use of more embedded clauses does not necessarily result in high er quality academic writing. We must pay more attention to t he functions these embedded clauses serve including their contribution to the experiential meaning and relevancy to the topic flow. T he present study enables educators and researchers to identify what adolescents know and can do with academic language in the current learning context so that instruction or remediation can be designed to meet the ir needs. By examining the relationship between academic language use and reading ability, the present study sheds light on the actual challenges of academic language to students of different reading abilities. The analyses of academic language features also indicate the develop ment of academic writing in the content area of world history. In high quality essays, we can find historical meanings are usually construed in a more abstract way that move beyond personal perception and lived experience toward a more objective and impers onal analysis and infused judgment. Academic language is the language used in schooling for purposes of learning. It evolves along with the knowledge students develop across the years of schooling and in different subj ect areas, becoming more dense and abstract as students advance through different phases of schooling. In secondary school, we expect students to use academic language to read and write successfully in different content areas. To do that, students must mas ter constellations of features that together construct academic texts
187 successfully. And this attention to language must go far beyond vocabulary development and knowledge of word meaning. Students must be taught to make explicit connection between the resources of linguistic features and meaning in the academic context.
188 APPENDIX A BENCHMARK ESSAYS Five Point Essay The Romans and Greeks greatly influenced later civilizations, even the present day. Their accomplishments, ideas and examples were revoluti onary and many of them are incorporated into modern day society. These two great empires were powerful, important parts of history. Their governments were predecessors to some of the most influential and successful governments today, like the United States and many others. Three of their most well known influences on modern culture are their religion, government, and the level of equality they had with their empires. Greek and Roman Gods have influenced an unimaginable number of literacy words. Authors of thousands of different genres of books have taken inspiration from the Greek and Roman Gods. The Greek Gods were the forefathers of the Roman Gods. They are incredibly similar in their powers, attitudes, and human like qualities. The Roman Gods were believ ed to be a sort of copycat of the Greek Gods, but nevertheless were a great influence on the world. Both the Roman and Greek gods were temperamental, based on common necessities of life and or unexplainable events or forces, and had many Gods and Goddesses Thus they backed each other up, and strengthened their influence on future peoples. So the Roman and Greek gods had similar characteristics, formed for different reasons. Both Greece and Rome were broken up into city states (polis). Their forms of gover nments in this way were quite similar. However in some parts of Greece (Sparta) these were at least a partial democracy. All adult males were allowed to vote on issues,
189 thus giving them power over their own fate. In Rome, there was not even partial democra cy. Unlike Greece Rome's Government was less lenient and equal. Woman had many more rights under Greek government (in some parts) than Rome. They had jobs in trade, and in general had a mere equal standing with men than Rome. Their systems of Government an d partial equality of women most likely inspired governments like the United States to look to Greek government for ideas, whereas Rome's government was less influential in that respect because it was patriarchal, like most ancient, even current nations. Slaves were imperative to the survival and well being of both Greece and Rome. They allowed Greek people the time to vote, sculpt, and study science, and philosophy. They also were one third of Rome's population and thus were very influential to their weal th and survival, providing labor for agriculture etc. slaves allowed the creation of the famous Greek philosophy, art, science, government and sculpting which are still used today. Slaves also allowed the Romans to survive and prosper, keeping their civili zation alive, and allowing it to expand its power and influence, thus allowing it to have a greater influence on the present day. Greece and Rome were two major civilizations from the past whose influences were so great that they survived and prospered ev en the present times. Many other civilizations have left their mark on the world through countless means. Each one adds a piece of culture and knowledge to our ever growing supply, and allows us to learn from the past and improve our present. Four Point Essay
190 Greece and Rome have similarities and differences. Those two civilizations had a time period that was close to one another. They are both similar and different in government, culture, and influence on European societies. Government of classical Gree ce and the Roman empire were both different. Rome was a republic meaning that people elected representatives to speak for them in the government. Greece had many different governments. They had a monarchy, in which a king or queen ruled but had limited pow er. They ad an Aristocracy, meaning that the wealthy ruled. They had an oligarchy which only a handful of people ruled. They also had a democracy where the people have the power. Both of these civilizations had tyrants or dictators. Tyrants back then were people that go power by caring about their people and listening to their needs. The culture of those people were similar in some ways but different in others. They were body polytheistic, meaning that they believed in many gods. However, they did not beli eve in all the same gods. Greeks usually built temples and statues to honor the gods. The Romans also did the same. Greece and Rome were also interested it sports but the Romans are a little more violent. Greece created the Olympics and Athletes competed i n sporting events against each other. Romans had the coliseum which had gladiator fights, chariot races, and bloody davels. Both of these civilizations had a great influence on European society. Greeks had an alphabet system that is the basis of many alpha bets today, even English. Also they had many advancements in science and art. Their creation of the Olympics made games that are still performed today. The Romans had an advanced number system called Roman numerals. They also had a language that is the bas is of many languages
191 today. English, Spanish, French, etc, all have roots in the Roman language Latin. Roman law is also the basis of many law systems today. Rome and Greece are both important civilizations. We probably would not be where we are today if we did not have them. They both have shaped human kind. They have similarities and difference s in their government, culture, and influence. They are both similar in how they are great civilizations of our past. Three Point Essay Romans and Greeks are two great people in the past. Their culture and government was so alike, but different, the Romans with their everlasting writing and the Greeks with their marvelous statues. The Romans and the Greeks are so alike and they are differe nt. The Romans gods were originated from the Greeks. All the gods were exactly the same, but they had different names. They still saw Zeus or Jupiter as the god of the heavens and Hera or Junow was still his wife. A difference is they worshiped different gods as better than other. For the Greeks they disliked Ares or Mars because they hated war, but the Romans loved him because they enjoy war. Their government was very different. While the Greeks stayed as a democracy the Romans government changed. They w ent from a Republic to an empire. For the Greeks there was voting among the people. The Romans had people that were elected a position to make the laws and rules. Something in their government that was the same was that only men that were citizens of the E mpire could vote or be in an election for a position.
192 The cultures of both empires influence us today. They both influence people in architect. The Greeks were known for their statues that showed strength and power, but no face expressions. The Romans sho wed statues that were doing an activity and they did have face expressions. One difference was the Romans Latin language was used more than the Greeks. Today it is ued in all the languages and writing. That is why the Romans and Greeks are the same, but v ery different. Even though they share the same gods they worship them different. Also they have a different government system, but the only allowed men to elect. For their influence today they both had statues, but they were shaped different. Also the Roma ns Latin language has lived on when the Greeks has not. Two Point Essay Greeks and Romans are both very similar and both very different than each other. They have similar cultures and different cultures. They have similar political ways and different poli tical ways. The two things that stick out to me are that they both have 2 different founders and 2 different ways of government. The most obvious similarity is that they are both located in the Mediterranean. The Greeks are located in the med. They were f ounded by Alexander Great. This happened after the fall of Persia. (I think) the Greeks also had a very distinctive way of their government. They used a system called the polis. The Romans were also located in the Mediterranean. They were founded by Romul us. Romulus had a brother too named Romulus who helped him. But Romulus was scared that his brother would take his position on the throne so he killed him. That
193 is where Rome got his name, after their founder Romulus. The Romans had a government system tha t was popular basis in the day. They had aristocrats. There were small city states, that were individually ruled by an aristocrat. Then there was an emperor who ruled over the aristocrats. People had to be born in to a family that already had an aristocrat to become one. All in all the Romans & Greeks were similar and different. They were both located in Mediterranean. They had different founders. Rome was founded by Alexander the Great. They both had very different government systems. Rome had aristocrats and Greece had the Polis system. One Point Essay Greeks are very good at architecture and sculpturing, but that was back then. There has been a lot of advances from the time or the Greeks to modern day America like the hand made building and sculptures to the money they used to get and lets not for get the opportunities they had. So now lets talk about that money! Back then you really did not get that much, but when the Helenistic Kings saw all the beauty people made with their own hands they were willi ng to spend money to beatify their cities within their states. Bothe Hellenlstic Kings and rich citizens patronized sculptures, though sands or statue erected in towns. Also in modern day America go its kind or the same. Then there is the opportunities th ey did not really have many. Unlike people in America they had to do the most out of them for each the thing we are doing today. Its
194 easier for Americans now in days because we have so much going on like carts, apartments, churches, schools and so much mor e things that but Greeks did not
195 APPENDIX B ACADEMIC VOCABULARY LIST (COXHEAD, 2000 ) Headwords Other words in the family. abandon abandoned, abandoning, abandonment, abandons, e.g. abstract abstraction, abstractions, abstractly, abstracts, e.g. academ academia, academic academically, academics, academies, e.g. Access accessed, accesses, accessibility, accessible, accessing, inaccessible accommodat accommodated, accommodates, accommodating accommodation accomp an accompanied accompanies, accompaniment, accompanying, unaccompanied Accumulate accumulated, accumulating, accumulation, accumulates Accura accuracy, accurately, inaccuracy, inaccuracies, inaccurate Achiev achievable, achieved, achievement, achievements, achieves, achieving acknowledg acknowledged, acknowledges, acknowledging, acknowledgement, acknowledgements Acquir acquired, acquires, acquiring, acquisition acquisitions Adapt adaptability, adaptable, adaptation, adaptations, adapted, ad apting, adaptive, adapts Adequa adequacy, adequately, inadequacies, inadequacy, inadequate, inadequately adjacent Adjust adjusted, adjusting, adjustment, adjustments, adjusts, readjust, readjusted, readjusting, readjustment, readjustments, readjusts Administrat administrates, administration, administrations, administrative, administratively, administrator, administrators Adult adulthood, adults advocat advocacy, advocated, advocates, advocating Affect affected, affecting, affective, affectively, affects, unaffected aggregat aggregated, aggregates, aggregating, aggregation Aid aided, aiding, aids, unaided albeit allocat allocated, allocates, allocating, allocation, allocations alter alterable, alteration, alterations, altered, altering, alte rnate, alternating, alters, unalterable, unaltered
196 alternative alternatively, alternatives ambigu ambiguities, ambiguity, unambiguous, unambiguously amend amended, amending, amendment, amendments, amends analog analogies, analogous analys analysed, analyser, analysers, analyses, analysing, analysis analyst, analysts, analytic, analytical, analytically annual annually anticipat anticipated, anticipates, anticipating, anticipation, unanticipated apparent apparently append appendix appended, appends, appending, appendices, appendixes appreciat appreciable, appreciably, appreciated, appreciates, appreciating, appreciation, unappreciated approach approachable, approached, approaches, approaching, unapproachable appropria appropriacy, appropriately, appropriateness, inappropriacy, inappropriate, inappropriately approximat approximated approximately, approximates, approximating, approximation, approximations arbitra arbitrariness, arbitrarily area areas aspect aspects assembl assembled, assembles, assemblies, assembling, assembly assess assessable, assessed, assesses, assessing, assessment assessments, reassess, reassessed, reassessing, reassessment, unassessed assign assigned, assigning, assignment, assignments, assigns, re assign, reassigned, reassigning, reassigns, unassigned assist assistance, assistant, assistants, assisted, assisting, assists, unassisted assum assumed, assumes, assuming, assumption, assumptions assur assurance assurances, assured, assuredly, assures, assuring attach attached, attaches, attaching, attachment, attachments, unattached attain attainable, attained attaining, attainment, attainments, attains, unattainable attitude attitudes attribut attributable, attributed, attributes, attributing, attribution
197 author authored, authoring, authors, authorship automat automatic, automated, automates, automating, automatically automation availabl availability, unavailable aware awareness, unaware behalf benefi beneficial, beneficiary, beneficiaries, benefited, benefiting, benefits bias biased, biases, biasing, unbiased bond bonded, bonding, bonds brief brevity, briefed, briefing, briefly, briefs bulk bulky capabl capabilities, capability, incapable capacit capacities, incapacitate, incapacitated catego categories categorisation, categorise, categorised, categorises, categorising, categorizing ceas ceased, ceaseless, ceases ceasing challeng challenged, challenger, challengers, challenges, challenging channel channe lled, channelling, channels chapter chapters chart charted, charting, charts, uncharted chemical chemically, chemicals circumstance circumstances Cite citation, citations, cited, citing, cites cita civil clarif clarification, clarified, clarifies, clarifying, clarity classic classical, classics clause clauses code coded, codes, coding coheren coherence coherently, incoherent, incoherently coincid coincided, coincides, coinciding, coincidence, coincidences, coincident, coincidental collaps collapsed, collapses, collapsible, collapsing
198 colleague colleagues commenc commenced commences, commencement, commencing, recommences, recommenced, recommencing comment commentaries, commentary, commentator, commentators, commented, commenting, comments commission commissioned, commissioner, commissioners, commissioning, commissions commit commitment, commitments, commits, committed, committing commodit commodities communicat communicable, communicated, communicates, communicating communication communications, communicative, communicatively, uncommunicative communit communities compatibl compatibility, incompatibility, incompatible compensat compensated, compensates, compensating, compensation compensations, compensatory compil compilation, compilations, compiled compiles, compiling complement complementary, complemented, complementing, complements complex complexities, complexity component componentry, components compound compounded, compounding, compounds comprehensive comprehensively comprise comprised, comprises, comprising comput computation, computational, computations, computable, computer computed, computerised, computers, computing conceive conceivable, conceivably, conceived conceives, conceiving, inconceivable, inconceivably concentrate concentrated, concentrates, concentrating, concentration concept conception, concepts, conceptual, conceptualisation, conceptualise, conceptualised, conceptualises, conceptualising, conceptually concl ud concluded, concludes, concluding, conclusion conclusions, conclusive, conclusively, inconclusive, inconclusively concurrent concurrently conduct conducted, conducting, conducts confer conference, conferences, conferred, conferring, confers
199 confin confined confines, confining, unconfined confirm confirmation, confirmed, confirming, confirms conflict conflicted, conflicting, conflicts conform conformable, conformability, conformance, conformation, conformed, conforming, conformist, conformists, co nformity conforms, nonconformist, nonconformists, nonconformity, non conformist, non conformists, non conformity consen consensus, consented, consenting, consents consequent consequence, consequences consequently considerable considerably consist consisted, consistency, consistent consistently, consisting, consists, inconsistencies, inconsistency, inconsistent constan constancy, constantly, constants, inconstancy, inconstantly constitu constituencies, constituency, constituent, constituents, constituted, constitutes, constituting, constitution, constitutions, constitutional constitutionally, constitutive, unconstitutional constrain constrained, constraining, constrains, constraint, constraints unconstrained construct constructed, construct ing, construction constructions, constructive, constructs, reconstruct, reconstructed, reconstructing, reconstruction, reconstructs consult consultancy, consultant, consultants, consultation, consultations, consultative, consulted, consults, consulting consum consumed, consumer, consumers, consumes, consuming, consumption contact contactable, contacted, contacting, contacts contemporar contemporaries context contexts, contextual, contextualise, contextualised, contextualising, uncontextualised contract contracted, contracting, contractor, contractors, contracts contradict contradicted, contradicting, contradiction contradictions, contradictory, contradicts contrar contrarily contrast contrasted, contrasting, contrastive, contrasts contribut contributed, contributes, contributing, contribution contributions, contributor, contributors
200 controvers controversies, controversial, controversially, uncontroversial convene convention convenes, convened, convening, conventional, conventionally, con ventions, unconventional converse conversely conver conversion, conversions, converted, convertible, converting, converts convinc convinced convinces, convincing, convincingly, unconvinced cooperat cooperated, cooperates, cooperating, cooperation, cooperative, cooperatively, co operate, co operated, co operates, co operation, co operative, co operatively coordinate coordinated, coordinates, coordinating, coordination coordinator, coordinators, co ordinate, co ordinated, co ordinates, co ordinating co ordination, co ordinator, co ordinators core cores, coring, cored corporat corporates, corporation, corporations correspond corresponded, correspondence, corresponding correspondingly, corresponds coupl coupled, coupling, couples creat created, creates, creating, creation, creations, creative, creatively, creativity, creator, creators, recreate, recreated, recreates, recreating credit credited, crediting, creditor, creditors, credits criteri criterion crucial crucially cultur cultural, culturally, cultured, cultures, uncultured currenc currencies cycl cycled, cycles, cyclic, cyclical, cycling data debat debatable, debated, debates, debating decade decades declin declined, declines, declining deduc deduced, deduces, deducing, deduction deductions defin definable, defined, defines, defining, definition definitions, redefine, redefined, redefines, redefining, undefined definit definitely, definitive, indefinite, indefinitely demonstrat demonstrable, demonstrably, demonstrated, demonstrates, demonstrating, demonstration, demonstrations, demonstrative,
201 demonstratively, demonstrator, demonstrators denot denotation, denotations, denoted, denotes, denoting deny deniable, denial, denials, denied, denies, denying, undeniable deni depress depressed, depresses, depressing, depression deriv derivation, derivations, derivative, derivatives, derived derives, deriving design designed, designer, designers, designing, designs despite detect detectable, detected detecting, detection, detective, detectives, detector, detectors, detects deviat deviated, deviates, deviating, deviation deviations device devices devot devoted devotedly, devotes, devoting, devotion, devotions differentiat differentiated, differentiates, differentiating, differentiation dimension dimensional, dimensions, multidimensional diminish diminished diminishes, diminishing, diminution, undiminished discrete discretely, discretion, discretionary, indi screte, indiscretion discriminat discriminated, discriminates, discriminating, discrimination displac displaced, displacement displaces, displacing display displayed, displaying, displays dispos disposable, disposal, disposed, disposes, disposing distinct distinction distinctions, distinctive, distinctively, distinctly, indistinct, indistinctly distort distorted distorting distortion, distortions, distorts distribut distributed, distributing, distribution, distributional, distributions, distributive, distributor, distributors, redistribute, redistributed, redistributes, redistributing, redistribution divers diversely, diversification, diversified, diversifies, diversify, diversifying, diversity document documentation, documented, docume nting, documents domain domains domestic domestically, domesticate, domesticated, domesticating, domestics dominat dominance, dominant dominated, dominates, dominating, domination
202 draft drafted, drafting, drafts, redraft, redrafted, redrafting, redrafts drama dramas, dramatic dramatically, dramatise, dramatised, dramatising, dramatises, dramatisation, dramatisations, dramatist, dramatists, dramatization, dramatizations, dramatizing duration dynamic dynamically, dynamics econom economic economical, economically, economics, economies, economist, economists, uneconomical edit edited, editing, edition, editions, editor, editorial, editorials, editors, edits element elements eliminat eliminated, eliminates, eliminating, elimination emerg emerged, emergence, emergent, emerges, emerging empha emphasise, emphasised, emphasising, emphatic, emphatically empiric empirically, empiricism enabl enabled, enables, enabling encounter encountered, encountering, encounters energ energetic, energetically, energies enforc enforced, enforcement, enforces, enforcing enhanc enhanced, enhancement, enhances, enhancing enorm enormity, enormously ensur ensured, ensures, ensuring entit entities environment environmental, environmentalist, environmentalists, environmentally, environments equat equated, equates, equating, equation equations equip equipment, equipped, equipping, equips equivalen equivalence erod eroded, erodes, eroding, erosion error erroneous, erroneously, errors establish disestablish, disestablished, disestablishes, disestablishing, disestablishment, established establishes, establishing, establishment, establishments estate estates
203 estimat estimated, estimates, estimating, estimation, estimations, over estima te, overestimate, overestimated, overestimates, overestimating, underestimate, underestimated, underestimates, underestimating ethic ethical ethically, ethics, unethical ethnic ethnicity evaluat evaluated, evaluates, evaluating, evaluation, evaluations, evaluative, re evaluate, re evaluated, re evaluates, re evaluating, re evaluation eventual eventuality, eventually eviden evidenced, evidence evidential, evidently Evolv evolution evolved, evolving, evolves, evolutionary, evolutionist, evolutionists exceed exceeded, exceeding, exceeds exclud excluded excludes, excluding, exclusion, exclusionary, exclusionist, exclusions, exclusive, exclusively exhibit exhibited, exhibiting, exhibition, exhibitions, exhibits expand expanded, expanding, expands, expansion, expansionism, expansive expert expertise, expertly, experts explicit explicitly exploit exploitation, exploited, exploiting, exploits export exported, exporter, exporters, exporting, exports expos exposed, exposes, exposing, exposure, exposures external externalisation, externalise, externalised, externalises, externalising, externality extract extracted, extracting, extraction, extracts facilitat facilitated, facilitates, facilities, facilitating, facilitation, f acilitator, facilitators, facility factor factored, factoring, factors featur featured, features, featuring federa federation, federations fee fees file filed, files, filing final finalise, finalised, finalises, finalising, finality, finally, finals financ financed, finances, financial financially, financier, financiers, financing finite infinite, infinitely
204 flexibl flexibility, inflexible, inflexibility fluctuat fluctuated, fluctuates, fluctuating, fluctuation, fluctuations focus focused, focuses, focusing, refocus, refocused, refocuses, refocusing format formatted, formatting, formats formula formulae, formulas, formulate, formulated, formulating, formulation, formulations, reformulate, reformulated, reformulating, reformulation, reformulations forthcoming found founded founder, founders, founding, unfounded framework frameworks function functional, functionally, functioned, functioning, functions fund funded, funder, funders, funding, funds fundamental fundamentally furthermore gender genders generat generated, generates, generating glob global globally, globalisation, globalization goal goals grad graded, grades, grading grant granted granting, grants guarantee guaranteed, guaranteeing, guarantees guideline guidelines hence hierarch hierarchical, hierarchies highlight highlighted, highlighting, highlights hypothes hypotheses, hypothesise, hypothesised, hypothesises, hypothesising, hypothetical, hypothetically identical identically identif identifiable, identification, identified identifies, identifying, identities, identity, unidentifiable ideolog ideological, ideologically, ideologies ignor ignorance, ignore, ignored, ignores, ignoring illustrat illustrated illustrates, illustrating, illustration, illustrations, illustrative
205 imag imagery, images immigrat immigrant, immigrants, immigrated, immigrates, immigrating, immigration impact impacted, impacting, impacts implement implementation implemented, implementing, implements implicat implicated, implicates, implicating, implication, implications implicit implicitly imply implied, implies implying impli impos imposed imposes, imposing, imposition incentive incentives inciden incident incidentally, incidents inclin inclination inclinations,inclined, inclines, inclining income incomes incorporat incorporated, incorporates, incorporating, incorporation index indexed, indexes, indexing indicat indicated, indicates, indicating, indication, indications, indicative, indicator, indicators individual individualised, individuality, individualism, individualist, individualists, individualistic, individually, individuals induc induced induces, inducing, induction inevitabl inevitability, inevitably infer inference, inferences, inferred, inferring, infers infrastructure infrastructures inherent inherently inhibit inhibited, inhibiting, inhibition, inhibitions, inhibits initial initially initiat initiated, initiates, initiating, initiation, initiations, initiative, initiatives initiator, initiators injur injured, injures, injuries, injuring, injury uninjured innovat innovation innovated, innovates, innovating, innovations, innovative, innovator, innovators input inputs insert inserted, inserting, insertion, inserts
206 insight insightful, insights inspect inspected, inspecting, inspection, inspections, inspector, inspectors, inspects instance instances institut instituted, institutes, instituting, institution, institutional, institutionalise, insti tutionalised, institutionalises, institutionalising, institutionally, institutions instruct instruction, instructed, instructing, instructions, instructive, instructor, instructors, instructs integral integrat integrated, integrates, integrating, integration integrity intelligen intelligence intelligently, unintelligent intens intensely, intenseness, intensification, intensified, intensifies, intensify, intensifying, intension, intensity intensive, intensively interact interacted, interacting, interaction interactions, interactive, interactively, interacts intermediate internal internalise, internalised, internalises, internalising, internally interpret interpretation interpretations, interpretative, interpreted, interpreting interpretive, interprets, misinterpret, misinterpretation, misinterpretations, misinterpreted, misinterpreting, misinterprets, reinterpret, reinterpreted, reinterprets, reinterpreting, reinterpretation, reinterpretations interval intervals interven intervened, intervenes, intervening, intervention, interventions intrinsic intrinsically invest invested, investing, investment investments, investor, investors, invests, reinvest, reinvested, reinvesting, reinvestment, reinvests investigat investigate d, investigates, investigating, investigation investigations, investigative, investigator, investigators invok invoked invokes, invoking involv involved involvement, involves, involving, uninvolved isolat isolated, isolates, isolating, isolation, isolationism issu issued, issues issuing item itemisation, itemise, itemised, itemises, itemising, items
207 job jobs journal journals justif justifiable, justifiably, justification justifications, justified, justifies, justifying, unjustified label labelled, labelling, labels labour laboured, labouring, labours layer layered, layering, layers lectur lectured, lecturer, lecturers, lectures, lecturing legal illegal, illegality, illegally, legality, legally legislat legislated, legislates, legislating, legislation legislative, legislator, legislators, legislature levy levies liberal liberalise liberalism, liberalisation, liberalised, liberalises, liberalising, liberalization, liberate, liberated, liberates, liberation, liberations, liber ating, liberator, liberators, liberally, liberals licen licences, license, licensed, licensing, licenses, unlicensed likewise link linkage, linkages, linked, linking, links locat located, locating, location locations, relocate, relocated, relocates, relocating, relocation logic illogical, illogically, logical, logically, logician, logicians maintain maintained, maintaining, maintains, maintenance major majorities, majority manipulat manipulated, manipulates, manipulating, manipulation, manipulations, manipulative manual manually, manuals margin marginal, marginally, margins matur immature, immaturity, maturation, maturational, matured, matures, maturing, maturity maximis max, maximised, maximises, maximising, maximisation, maximum mechanism mechanisms media mediate mediated, mediates, mediating, mediation medical medically
208 medium mental mentality, mentally method methodical, methodological, methodologies, methodology, methods migrate migrant, migrants, migrated, migrates, migrating, migration, migrations, migratory military minimal minimalisation, minimalise, minimalises, minimalised, minimalising, minimalist, minimalists, minimalistic, minimally minimis minimised minimises, minimising minimum minist ministered, ministering, ministerial, ministries minor minorities minority, minors mode modes modif modification, modifications, modified, modifies, modifying, unmodified monitor monitored, monitoring, monitors, unmonitored motiv motivate, motivated, motivates, motivating, motivation, motivations, motives, unmotivated mutual mutually negat negative negated, negates, negating, negatively, negatives network networked, networking, networks neutral neutralisation, neutralise, neutralised, neutralises, neutralising, neutrality nevertheless nonetheless norm norms notion notions notwithstanding nuclear objectiv objectively, objectivity obtain obtainable, obtained obtaining, obtains, unobtainable obvious obviously occup occupancy, occupant, occupants, occupation, occupational, occupations, occupied, occupier, occupiers, occupies, occupying
209 occur occurred, occurrence, occurrences, occurring, occurs, reoccur, reoccurred, reoccurring, reoccurs odd odds offset offsets, offsetting ongoing option optional, options orient orientate orientated orientates, orientation, orientating, oriented, orienting, orients, reorient, reorientation outcome outcomes output outputs overall overlap overlapped, overlapping, overlaps overseas panel panelled, panelling, panels paradigm paradigms paragraph paragraphing, paragraphs parallel paralleled, parallels, unparalleled parameter parameters participat participant, participants, participated, participates, participating, participation participatory partner partners, partnership partnerships passive passively, passivity perceiv perceived, perceives, perceiving, perception, perceptions percent percentage, percentages period periodic, periodical, periodically, periodicals, periods persist persisted, persistence, persistent persistently, persisting, persists perspective perspectives phase phased, phases, phasing phenomen phenomena, phenomenal philosoph philosopher, philosophers, philosophical, philosophically, philosophies, philosophise, philosophised, philosophises, philosophising physical physically plus pluses
210 polic policies portion portions pose posed poses, posing positive positively potential potentially practitioner practitioners preced preceded, precedence, precedent, precedes, preceding, unprecedented precis imprecise, precisely, precision predict predictability, predictable, predictably, predicted predicting, prediction, predictions, predicts, unpredictability, unpredictable predomina predominance, predominantly predominate, predominated, predominates, predominating preliminar preliminaries presum presumably, presumed, presumes, presuming, presumption, presumptions, presumptuous previous previously primar primarily primacy primacy principal principally principle principled, principles, unprincipled prior priorit priorities, prioritisation, prioritise, prioritised, prioritises, prioritising proceed procedural, procedure, procedures, proceeded, proceeding, proceedings, proceeds process processed, processes, processing professional professionally, professionals, professionalism prohibit prohibited, prohibiting, prohibition, prohibitions, prohibitive, prohibits project projected, projecting, pr ojection, projections, projects promot promoted, promoter, promoters, promotes, promoting, promotion, promotions proportion disproportion, disproportionate, disproportionately, proportional, proportionally, proportionate, proportionately, proportions prospect prospective, prospects
211 protocol protocols psycholog psychological, psychologically, psychologist, psychologists publication publications publish published, publisher, publishers, publishes, publishing, unpublished purchas purchased, purchaser, purchasers, purchases, purchasing pursu pursued, pursues, pursuing, pursuit, pursuits qualitative qualitatively quot quotation, quotations, quoted, quotes, quoting radical radically, radicals random randomly, randomness rang ranged, ranges, ranging ratio ratios rational irrational, rationalisation, rationalisations, rationalise, rationalised, rationalises, rationalising, rationalism, rationality, rationally react reacted, reacts, reacting, reaction reactionaries, reactionary, reactions, reactive, reactivate, reactivation, reactor, reactors recover recoverable, recovered, recovering, recovers, recovery refin refined, refinement, refinements, refines, refining regime regimes region regional, regionally, regions register deregister, deregistered, deregistering, deregisters, deregistration, registered, registering, registers, registration regulat deregulated, deregulates, deregulating, deregulation, regulated, regulates, regulating, regulation,regulations, regulator, regul ators, regulatory, unregulated reinforc reinforced reinforcement, reinforcements, reinforces, reinforcing reject rejected, rejecting, rejection, rejects, rejections relax relaxation, relaxed relaxes, relaxing releas released, releases, releasing relevant irrelevance, irrelevant, relevance reluctan reluctant reluctantly rely reliability, reliable, reliably, reliance reliant, relied, relies, relying, unreliable reli
212 remov removable, removal, removals, removed removes, removing requir required requirement, requirements, requires, requiring research researched, researcher, researchers, researches, researching resid resided, residence, resident residential, residents, resides, residing resolv resolution, resolved, resolves, resolving unresolved resourc resourced, resourceful, resources resourcing, unresourceful, under resourced respond responded, respondent, respondents, responding, responds, response responses, responsive, responsiveness, unresponsive restor restoration, restored, restores, restoring restrain restrained, restraining, restrains, restraint, restraints unrestrained restrict restricted, restricting, restriction, restrictions, restrictive, restrictively, restricts, unrestricted, unrestrictive retain retaine d retaining, retainer, retainers, retains, retention, retentive reveal revealed, revealing, reveals, revelation, revelations revenue revenues revers reversal, reversed, reverses, reversible, reversing, reversals, irreversible revis revised, revises, revising, revision revisions revolution revolutionary, revolutionaries, revolutionise, revolutionised, revolutionises, revolutionising, revolutionist, revolutionists, revolutions rigid rigidities, rigidity, rigidly role roles rout routed, routes, routing scenario scenarios schedul reschedule, rescheduled, reschedules, rescheduling, scheduled, schedules, scheduling, unscheduled schem schematic, schematically, schemed, schemes, scheming scope section sectioned, sectioning, sections sector sectors secur insecure, insecurities, insecurity, secured, securely, secures, securing, securities, security seek seeking, seeks, sought
213 select selected, selecting, selection, selections, selective, selectively, selector, selectors, selects sequenc sequenced, sequences, sequencing, sequential, sequentially series sex sexes, sexism, sexual, sexuality, sexually shift shifted, shifting, shifts significan insignificant, insignificantly, significance, significantly, signified, signifies, signify, si gnifying similar dissimilar, similarities, similarity, similarly simulat simulated, simulates, simulating, simulation site sites so called sole solely somewhat sourc sourced, sources, sourcing specific specifically, specification, specifications, specificity, specifics specif specifiable, specified, specifies, specifying, unspecified spher spheres, spherical, spherically stabl instability, stabilisation, stabilise, stabilised, stabilises, stabilising, stability, unstable statistic statistician, statisticians, statistical, statistically, statistics status straightforward strateg strategic, strategies strategically, strategist, strategists stress stressed, stresses, stressful, stressing, unstressed structure restructure, restructured, restructures, restructuring, structural, structurally, structured, structures, structuring, unstructured styl styled, styles, styling, stylish, stylise, stylised, stylises, stylising submit submission, submissions, submits, sub mitted, submitting subordinat subordinates, subordination subsequent subsequently subsid subsidiary, subsidies, subsidise, subsidised, subsidises, subsidising substitut substituted, substitutes, substituting, substitution
214 successo succession, successions, successive, successively, successors successi sufficien sufficiency, insufficient, insufficiently, sufficiently sum summation, summed, summing, sums supplement supplementary, supplemented, supplementing, supplements survey surveyed, surveying, surveys surviv survival, survived, survives, surviving, survivor, survivors suspend suspended suspending, suspends, suspension sustain sustainable sustainability, sustained, sustaining, sustains, sustenance, unsustainable symbol symbolic, symbolically, symbolise, symbolises, symbolised, symbolising, symbolism, symbols tap taped, tapes, taping target targeted, targeting, targets task tasks team teamed, teaming, teams technical technically technique techniques technolog technological, technologically temporar temporarily tens tension tensely, tenser, tensest, tensions termina terminal, terminals, terminated, terminates, terminating, termination terminations text texts, textual theme themes, thematic, thematically thematic theor theoretical, theoretically, theories, theorist, theorists thereby thesis theses theses topic topical, topics trac traceable, traced, traces, tracing tradition non traditional, traditional traditionalist, traditionally, traditions
215 transfer transferable, transference, transferred, transferring, transfers transform transformation, transformations, transformed, transforming, transforms transit transited, transiting, transition, transitional, transitions, transitory, transits transmi transmis sion, transmissions, transmitted, transmitting, transmits transport transportation, transported, transporter, transporters, transporting, transports trend trends trigger triggered, triggering, triggers ultimate ultimately undergo undergoes, undergoing, undergone, underwent underwent underl underlay, underlies, underlying undert undertaken, undertakes, undertaking, undertook uniform uniformity, uniformly unif unification, unified unifies, unifying unique uniquely, uniqueness utilis utilisation, utilised, utilises, utilising, utiliser, utilisers, utility, utilities valid invalidate, invalidity, validate, validated, validating, validation, validity validly vary invariable, invariably, variability, variable, variables, variably, variance, variant, variants, variation, variations, varied, varies, varying vari vehicle vehicles version versions via violat violated, violates, violating, violation violations virtual virtually visibl visibility, visibly, invisible, invisibility vision visions visual visualise, visualised, visualised, visualising, visualisation, visually volume volumes, vol vol
216 voluntar voluntarily, volunteer, volunteering, volunteered, volunteers welfare whereas whereby widespread
217 LIST OF REFERENCES Achugar, M., & Schleppegrell, M. (2005). Beyond connectors: The construction of cause in history textbooks. Linguistics and Education 16 (3), 298 318. Adams, T. L. (2003). Reading mathematics: More than words can say. The Reading Teach er, 56 (8), 786 795. Adams, M. (2005). Unmasking the myths of structured English immersion. Retrieved December 12, 2009, from http://www.worldlanguage.com/articles/59.htm Adamson, H.D. (1990). ESL students' use of academic skills in content courses. English for Specific Purposes, 9 67 87. Aguirre Munoz, Z., Boscardin, C. K., Jones, B., Park, J., Chinen, M., Shin, H. S., et al. (2006b). Consequences and validity of performance assessment for English language learners: Integrating academi c language and ELL instructional needs into opportunity to learn measures. CSE 678. National Center for Research on Evaluation, Standards, and Student Testing (CRESST). Aguirre Munoz, Z., Parks, J. E., Benner, A., Amabisca, A., & Boscardin, C. K. (2006a). Consequences and validity of performance assessment for English language learners: C onceptualizing & d eveloping teachers' expertise in academic language CSE 700 National Center for Research on Evaluation, Sta ndards, and Student Testing (CRESST). Anstrom, K., DiCerbo, P., Butler, F., Katz, A., Millet, J., & Rivera, C. (2010). A Review of the literature on a cademic English: I mplications for K 12 English l anguage l earners. Arlington, VA: The George Washington University Center for Equity and Excellence in Education. August, D., & Hakuta, K. (1997). Improving schooling for language minority children: A research agenda. National r esearch c ouncil c ommittee on d eveloping a r es earch a genda on the e ducation of l imited English p roficient and b ilingual s tudents. National Academy Press: Washington, D.C. Aukerman, M. (2007). A culpable CALP: R ethinking the conversational/academic language proficiency distinction in early literacy in struction. Reading Teacher, 60 (7), 626 635. Bachman, L. F. (2003). Issues in a ssessing l anguage p roficiency and a cademic a chievement Paper presented at the East Coast Organization of Language Testers Conference (ECOLT), Washington, D.C. Bailey, A. L. (2006). From Lambie to Lambaste: T he conceptualization, operationalization and use of academic English in the assessment of ELL
218 students In K. Rolstad (Ed.), Rethinking school language (pp. 8 44) Mahwah, NJ: LEA. Bailey, A. L. (2007). The language demands of school: Putting academic language to the test New Haven: Yale University Press. Bailey, A. L. & Butler, F. (2003). An evidentiary framework for operationalizing academic language for broad application to K 12 education: A design document. CSE technical report 727. National Center for Research on Evaluation, Standards, and Student Testing (CRESST). Bailey, A. L., & Heritage, H. M. (2008). Formative assessment for literacy, grades K 6: Building reading and academic language skills across the cur riculum Thousand Oaks, CA: Corwin Press. Bailey, A. L., Huang, B. H., Shin, H. W., Farnsworth, T., & Butler, F. A. (2007). Developing academic English language proficiency prototypes for 5th grade reading: Psychometric and linguistic profiles of tasks. CSE technical report 727. National Center for Research on Evaluation, Standards, and Student Testing (CRESST). BartolomÂŽ, L. I. (1998). The misteaching of academic discourses: The politics of language in the classroom. Boulder, Colo.: Westview Press. Beck, I. L. McKeown, M. G., & Kucan, L. (2002). Bringing words to life: Robust vocabulary instruction New York: The Guilford Press. Berman, I., & Biancarosa, G. (2005). Reading to achieve: A governor's guide to adolescent literacy Washington, DC: Natio nal Governors Association Center for Best Practices. Biber, D. (1988). Variation across s peech and w riting. Cambridge: Cambridge University Press. Biber, D. (1995). Dimensions of r egister v ariation: A c ross linguistic c omparison Cambridge: Cambridge Uni versity Press. Blanton, L. L. (1998). Discourse, a rtifacts, and the o zarks: Understanding a cademic li teracy. In V. Zamel & S. Ruth, ( Eds ) Negotiating a cademic l iteracies (pp. 219 236). Mahwah, New Jersey: Lawrence Erlbaum Associates, Inc., publishers. B loor, M., & Bloor, T. (1992). Given and new information in the thematic organization of text: An application to the teaching of academic writing. Occasional Papers in Systemic Linguistics, 6 33 43.
219 Boscardin, C. K., Aguirre Munoz, Z., Chinen, M., Leon, S ., & Shin, H. S. (2003). Consequences and validity of performance assessment for English language learners: Assessing OTL in grade 6 language arts (CRESST technical report). Los Angeles: University of California, National Center for Research on Evaluation, Standards, and Student Testing. Bray, J. H., & Maxwell, S. E. (1982). Analyzing and i nterpreting s ignificant MANOVAs. R eview OF E ducational R esearch 52 (3), 340 367. Bruna, K. R., Vann, R., & Escudero, M. P. (2007). What's language got to do with it?: A case study of academic language instruction in a high school English Learner Science'' class. Journal of English for Academic Purposes 6 (4), 36 54. Bunch, G. (2006). "Academic English" in the 7th grade: Broadening the lens, expanding access. Journal o f English for Academic Purposes 5 (4), 284 301. Bunch, G. Abram, P. L., Lotan, R. A., & ValdÂŽs, G. (2001). Beyond sheltered instruction: Rethinking conditions for academic language development. TESOL Journal, 10 (2 3), 28 33. Butler, F. A., Bailey, A. L ., Stevens, R., Huang, B., & Lord, C. (2004a). Academic English in fifth grade mathematics, science, and social studies textbooks. CSE report 642. Center for the Study of Evaluation (CSE)/National Center for Research on Evaluation, Standards, and Student T esting (CRESST). Butler, F. A., Lord, C., Stevens, R., Borrego, M., & Bailey, A. L. (2004b). An approach to operationalizing academic language for language test development purposes: Evidence from fifth grade science and math. CSE report 626. Center for the Study of Evaluation (CSE), National Center for Research on Evaluation, Standards, and Student Testing (CRESST). CalderÂ—n, M. E. (2007). Teaching reading to English language learners, grades 6 12: A framework for improving achievement in the content a reas Thousand Oaks, CA: Corwin Press. CalderÂ—n, M., August, D., Slavin, R. E., Cheung, A., DurÂ‡n, D. & Madden, N. A. (2005). Bringing words to life in classrooms with English language learners. In Hiebert, A & M. Kamil (eds.) Research and practice on vo cabulary (pp. 115 136). Mahwah, NJ: Lawrence Erlbaum. Campbell, M., & Km iecik, M. (2004). The Greatest literacy challenges facing contemporary high school teachers: Implications for secondary teacher p reparation. Reading Horizons 45 (1), 1 25.
220 Carlo, M. S., August, D., McLaughlin, B., Snow, C. E., Dres sler, C., Lippman, D. N., et al. (2004). Closing the gap: Addressing the vocabulary needs of English language learners in bilingual and mainstream classrooms. Reading Research Quarterly, 39 (2), 188 215. Carlo, M.S., August, D., & Snow, C.E. (2005). Sustained v ocabulary l earning s trategy i nstruction for English l anguage l earners. In E.H. Hiebert and M.L. Kamil (Eds.), Teaching and l earning v ocabulary (pp. 137 153). Mahwah, NJ: Lawrence Erlbaum Associates. Carr, J., Sexton, U., & Lagunoff, R. (2006). Making science accessible to English language learners: A guidebook for teachers San Francisco: West Ed Carrier, K. A., & Tatum, A. W. (2006). Creating sentence walls to help English language learners develop content literacy. Reading Teacher, 60 (3), 285 288. Castro, C. D. (2004). Cohesion and the social construction of meaning in the essays of F ilipino college students writing in L2 E nglish. Asia Pacific Education Review, 5 (2), 215 225. Celce Murcia, M., & Larsen Freeman, D. (1999). The grammar book: An ESL/EFL teacher's course Boston MA : Heinle & Heinle. Chall, J. S. (1996). Stages of reading development (2nd ed.). Fort Worth, TX: Harcourt Brace. Chamot, A., & O'Malley, M.J. (1986). A cognitive academic language learning approach: An ESL content based curriculum Wheaton, MD: National Clearinghouse for Bilingual Education. Chamot, A., & O'Malley, M.J. (1987). The cognitive academic language learning approach: A bridge to the mainst ream. TESOL Quarterly, 21 227 249. Chamot, A. U., & O'Malley, J. M. (1994). The CALLA handbook: Implementing the cognitive academic English learning approach Reading, MA: Addison Wesley Publishing Company. Chamot, A. U., & O'Malley, J. M. (1996). The c ognitive academic language learning approach: A model for linguistically diverse classrooms. Elementary School Journal, 96 (3), 259 273. Christie, F. & Derewianka, B. (2008). School discourse: learning to write across the years of s chooling London: Conti nuum Press
221 Christie, F. (1998). Learning the literacies of primary and secondary schooling. In F. Christie & R. Misson (Eds.), Literacy and schooling (pp. 47 73). London: Routledge. Christie, F. (2002). The development of abstraction in adolescence in s ubject English. In M. Schelppegrell & M. C. Colombi (Eds.), Developing advanced literacy in first and second language: Meaning with power (pp. 45 66). London: Routledge. Coelho, E. (1982). Language across the curriculum. TESL Talk, 13 56 70. Coffin, C. (1997). Constructing and giving value to the past: An investigation into secondary school history. In F. Christie and J.R. Martin (Eds), Genres and institutions: Social processes in the workplace and school (pp. 196 230). London: Cassell. Coffin, C. (200 4). Learning to write history: The role of causality. Written Communication 21 (3), 261 289. Coffin, C. (2006a). Historical discourse: The language of time, cause, and evaluation London: Continuum. Coffin, C (2006 b ). Learning the language of school history: the role of linguistics in mapping the writing demands of the secondary school curriculum. Journal of Curriculum Studies, 38 (4), 413 429. Cohen, J. (1988). Statistical p ower a nalysis for the b ehavioral s ciences (2 nd Ed.). Hillsdale, NJ: Lawrence Erlbaum Associates. Cohen, J., Cohen, P., West, S. G., & Aiken, L. (2003). Applied multiple regression: Correlation analysis for the behavioral sciences (3rd ed.). Mahwah, NJ: Lawrence Erlbaum. C o lombi, M. C. (2002). Ac ademic language development in Latino students' writing in Spanish. In Schleppegrell, M.and Colombi, M. C. (Eds). Developing a dvanced l iteracy in f irst and s econd l anguages ( pp. 67 85 ) Mahwah, NJ: Lawrence Erlbaum Associates. Connor, U. (1995). Examinin g s yntactic v ariation a cross t hree English s peaking n ationalities t hrough a m ultifeature/ m ultidimensional a pproach In Rubin, D. L. (Eds.). Composing social identity in written language (pp. 75 88) Hillsdale, N J: L. Erlbaum Associates. Coxhead, A. (2000). A new academic word list. TESOL Quarterly, 34 (2), 213 238.
222 Crowhust, M. (1990). The development of persuasive/argumentative writing. In R. Beach & S. Hynds (Eds.), Developing discourse practices in adolescence and adulthood (pp. 200 2 23). Norwook, NJ: Ablex. Cruz, M. (2004 ). Can English language learners acquire academic English? English Journal 93 (4), 14 17. Cummins, J. (1981). The role of primary language development in promoting educational success for language minority students In Schooling and language minority students: A theoretical framework (pp. 3 50). Los Angeles: California State University. Cummins, J. (2000a). Language, power, and pedagogy b ilingual children in the crossfire : Bilingual education and bilingualism Clevedon, England: Multilingual Matters. Cummins, J. (2000b). Academic language learning, transformative pedagogy, and information technology: Towards a critical balance. TESOL Quarterly, 34 (3), 537 547. de Oliveira, L. C. (2010). Nouns in history: Pack aging information, expanding explanations, and structuring reasoning. The History Teacher 43 (2), 191 203. de Oliveira, L. C. (2011). Knowing and writing school history: The language of students' expository writing and teachers' expectations Information Age Publishing. Delpit, L. D. (1998). The p olitics of t eaching l iterate d isco urse. In V. Zamel & S. Ruth, Negotiating Academic Literacies (pp. 207 218). Mahwah, New Jersey: Lawrence Erlbaum Associates, Inc., publishers. Diaz Rico, L.T., & Weed, K.Z. (200 2). The cross cultural, language, and academic development handbook: A complete K 12 reference guide (2nd ed.). Boston: Ally & Bacon. Donlevy, J. (2007). A new year's resolution for 2007: Increase understanding of adolescent literacy to help close the ac hievement gap. International Journal of Instructional Media, 34 ( 1), 1 4. Draper, R. J. (2002). Every teacher a literacy teacher? An examination of the literacy related messages in secondary methods textbooks. Journal of Literacy Research, 34 357 384. D raper, R. J. (2008). Redefining content area literacy teacher education: Finding my voice through collaboration. Harvard Educational Review, 78 (1), 60 83.
223 Echevarria, J., Short, D., & Powers, K. (2006). School reform and standards based education: A mode l for English language learners. Journal of Educational Research 99 (4), 195 210. Edelsky, C. (1991) With l iteracy and j ustice for a ll: R ethinking the s ocial in l anguage and e ducation London: Falmer Press. Eggins, S. (2004). An introduction to systemic functional linguistics (2 nd ed.) London: Continuum International Publishing Group Ltd Fang, Z. (2005). Thematic analysis as a new tool for genre assessment in early literacy research. National Reading Conference Yearbook, 54, 168 181. Fang, Z. (2008a). Going beyond the Fab Five: Helping students cope with the unique linguistic challenges of expository reading in intermediate grades. Journal of Adolescent and Adult Literacy, 51 (6), 476 487. Fang, Z. (2008b). Nouns in s cho ol c hildren's s cience w riting: A f unctional l inguistics p erspective. In Proceedings of ISFC 35: Voices around the w orld (pp. 25 30). Retrieved June 5 th 2009, from http://minerva.ling.mq.edu.au/isfc/. Fang, Z. (2008c). Technicality and r easoning in s cience. In Z. Fang & M. J. Schleppegrell (Ed.), Reading in secondary content areas : A language based pedagogy ( PP. 18 38 ) Ann Arbor, MI: University of Michigan Press. Fang, Z., & Schleppegrell, M. J. (20 08 ). Reading in secondary content areas: A language based pedagogy Ann Arbor, MI: University of Michigan Press. Fang, Z., Schleppegrell, M. J., & Cox, B. E. (2006). Understanding the language demands of schooling: Nouns in academic registers. Journal of Literacy Research, 38 (3), 247 273. Fang Z., & Wang Z. (2011). Beyond rubrics: Using functional language analysis to evaluate student writing. Australian Journal of Language and Literacy 34(2), 147 165. Field, A. (2005). Discovering statistics using SPSS London: Sage. Fitzgerald, J., & Shanahan, T. (2 000). Reading and writing relations and their development. Educational Psychologist, 35 39 50. Flynt, E., & Brozo, W. (2008). Developing academic language: Got words? Reading Teacher 61 (6), 500 502. Freeman, Y., & Freeman, D. (2003). Struggling English language learners: Keys for academic success. TESOL Journal, 12 (3), 5 10.
224 Fries, P. H. (2002). The flow of information in a written Eng lish text. In P.H. Fries, et al ., (Eds.), Relations and Functions Within and Around Language (pp. 117 155). London/New York: Gee, J. P. (1996) Social l inguistics and l iteracies: Ideology in d iscourses (2 nd Ed.) London: Falmer Press. Gee, J. P. (2004a). Language in the science classroom: Academic social languages at the heart of the school based literacy, In Saul, E. W., Crossing borders in literacy and science instruction (pp. 13 32 ). International Reading Association. Gee, J. P. (2004b). Situated language and learning: A critique of traditional schooling New York: Routledge. Gee, J. P. (2008). Social linguistics and literacies: ideology in discourses London: Routledge. Genesee, F., Lindholm Leary, K., Saunders, W., & Christian, D. (2005). English language learners in U.S. schools: An o verview of r esearch f indings. Journal of Education for Students Pl aced at Risk, 10 ( 4), 363 386. Gibbons, J., & Lascar, E. (1998). Operationalising academic language proficiency in bilingualism research. Journal of Multilingual and Multicultural Development, 19 (1), 40 50. Gibbons, P. (2003). "Mediating l anguage l earni ng: Teacher i nteractions with ESL s tudents in a c ontent b ased c lassroom." TESOL, 37(2), 247 273. Graff, G. (1999). The academic language gap. Clearing House, 72 (3), 140 142. Graves, D. H. (1983). Writing: t eachers and children at work Potsmouth, N.H.: Heinemann. Guerrero, M. D. (2004). Acquiring academic English in one year. Urban Education, 39 ( 2), 172 199. GutiÂŽrrez K. D. (1995). Unpacking academic discourse. Discourse Processes, 19 21 37. GutiÂŽrrez, K. D. (2008). Developing a sociocritical literacy in the third space. Reading Research Quarterly 43 (2), 148 164. Halliday, M.A.K. (1978). Language as social semiotic Baltimore: University Park Press.
225 Halliday, M.A.K. (1985). Spoken and written language. Oxford: Ox ford University Press. Halliday, M.A.K. (2008). Complementarities in Language. Beijing. The Commercial Press. Halliday, M.A.K. & Hasan. R. (1989) Language, context and t ext: Aspects of language in a social semiotic p erspective Oxford: OUP. Halliday, M. A. K., & Martin, J. R. (1993). Writing science: Literacy and discursive power Pittsburgh: University of Pittsburgh Press. Halliday, M. A. K. & Matthiessen, C. (2004). An introduction to functional grammar (3 rd Ed.), London: Hodder Arnold. Halliday, M. A. K., & Webster, J. (2004). The language of science London: Continuum. Hammond, J. (2006). High challenge, high support: Integrating language and content instruction for diverse learners in an English literature classroom. Journal of Englis h for Academic Purposes 5 (4), 269 283. Harwell, M. (1988). Univariate and Multivariate Tests: ANOVA versus MANOVA. Educational Research Quarterly 12 (3), 20 28. Hasan, R. (1996). Literacy, everyday talk and society. In R. Hasan & G. Williams (eds.), Literacy in society (pp. 377 424). London: Longman. Heath, S. B. (1983). Ways with words New York: Cambridge University Press. Hinkel, E. (2002). Second language writers' text: Linguistic and rhetorical features. Mahwah, NJ: Lawrence Erlbaum Associates. Hirose, K. (2003). Comparing L1 and L2 organizational patterns in the argumentative writing of Japanese EFL students. Journal of Second Language Writing 12 (2), 181 209. Huang, J. (2004). Socialising ESL students into the discourse of school science through academic writing. Language & Education: An International Journal, 18 (2), 97 123. Hunt, K. W. (1965). Grammatical structures written at three grade levels. Champaign, IL: National Council of Teachers of English. Jacob, E., RotternBerg, L, Patrick S., & Wheeler, E. (1996). Cooperative learning: Context and opportunity for acquiring academic English. TESOL Quarterly, 30 (2), 253 280.
226 Jacobs, V. A. (2008). Adolescent literacy: Putting the crisis in context. Harvard Educational Review, 78 (1), 7 39. Johansson, V. (2008). Lexical diversity and lexical density in speech and writing. Lund University, Department of Linguistics and Phonetics. Retrieved from http://www.lunduniversity.lu.se/o.o.i.s?id=24732&postid=1623113 Jordan, R. R. (1997). English for academic purposes: A guide and resource book for teachers Cambridge, U.K ; New Yor k, NY: Cambridge University Press. Just, M. A., & Carpenter, P. A. (1992). A capacity theory of comprehension. Psychological Review, 99, 122 149. Kamberelis, G. (1999). Genre d evelopment and Learning: Children w riting s tories, s cience r eports, and p oems Research in the T ea c hing of English 33, 403 459. Kamil, M.L. (2004). Vocabulary and comprehension instruction: Summary and implications of the National Reading Panel findings. In P. McCardle & V. Chhabra (Eds.), The voice of evidence in reading researc h (pp. 213 234). Baltimore, MD: Paul H. Brookes. Kidd, R. (1996). Teaching academic language functions at the secondary level. Canadian Modern Language Review, 52 (2), 285 307. Knight, S. L., & Wiseman, D. L. (2006). Lessons learned from a research synth esis on the effects of teachers' professional development on culturally diverse students. In K. Tellez, & H. C. Waxman (Eds.), Preparing quality educators for English language learners: Research, policy, and practice (pp. 71 98). Mahwah, NJ: Lawrence Erlba um Associates (Bks). Krashen, S. (1997). Does free voluntary reading lead to academic language? Journal of Intensive English Studies, 11 1 18. Lea, M., & Street, B. (2006). The academic literacies model: Theory and applications. Theory Into Practice 45 (4), 368 377. Lemke, J. (1990). Talking science: Language, learning, and values New York: Ablex. Lesley, M., Watson, P., & Elliot, S. (2007). "School" reading and multiple texts: Examining the metacognitive development of secondary level preservice t eachers. Journal of Adolescent & Adult Literacy, 51 (2), 150 162. Liang, X., & Mohan, B. (2003). Dilemmas of cooperative learning and academic proficiency in two languages. Journal of English for Academic Purposes 2 (1), 35.
227 Lukin, E. (2008). Reading Lit erary Texts: Beyond Personal Responses. In Schleppegrell & Fang (Eds). Reading in s econdary c ontent a reas : A l anguage b ased p edagogy (pp. 84 103) Ann Arbor, MI: The University of Michigan Press MacSwan, J. (2000) The threshold hypothesis, semilingualism, and other contributions to a de fi cit view of linguistic minorities. Hispanic Journal of Behavioral Sciences, 22 (1), 3 45. Makalela, L. (2004). Differential error types in second language students' written and sp oken texts. Written Communication 21 (4), 368 385. Martin, J. R. (1989). Factual writing. Oxford, England: Oxford University Press. Martin, J. R. (2002). Writing history: Construing time and value in discourses of the past. In Schleppegrell M. J. & Colo mbi M. C. Developing a dvanced Literacy in f irst and s econd Languages: M eaning with p ower ( pp. 87 118 ) Mahwah, NJ, Lawrence Erlbaum Assoc. MartÂ’nez, I. A. (2001). Impersonality in the research article as revealed by analysis of the transitivity structure. E nglish for Specific Purposes, 20, 227 247. Martin Jones, M., & Romaine, S. (1986). Semilingualism: A half baked theory of communicative c ompetence. Applied Linguistics 7 (1), 26 38. Martin, J. (1997). Analyzing genre: Functional parameters. In F. Christie & J. Martin (Eds.), Genre and institutions: Social processes in the workplace and school (pp. 3 39). London: Continuum. Mays, L. (2008). The cultural divide of discourse: Understanding how English language learners' p rimary d iscourse influences acquisition of literacy. Reading Teacher, 61 (5), 415 418. McNam ara, D. S., Crossley, S. A., & McCarthy, P. M. (2010). Linguistic f eatures of w riting q uality. Written Communication. 27 (1), 57 86. Menyuk, P. (1988). Language development: Knowledge and use. Glenview, IL: Scott, Foresman and Company. Mohan, B., & Beckett, G. H. (2003). A functional approach to research on content base d language learning: Recasts in causal explanations. The Modern Language Journal, 87 421 432. Mohan, B., & Slater, T. (2006). Examining the theory/practice relation in a high school science register: A functional linguistic perspective. Journal of English for Academic Purposes, 5 302 316.
228 Moje, E. B., Overby, M., Tysvaer, N., & Morris, K. (2008). The complex world of adolescent literacy: Myths, motivations, and mysteries. Harvard Educational Review, 78 (1), 107 154. Montes, F. (2002). Enhancing c ontent areas through a cognitive academic language learning based collaborative in south Texas. Bilingual Research Journal, 26 (3), 697. Myhill, D. (2008). Becoming a designer: trajectories of linguistic development. In Beard, R. Myhill, D. Nystrand, M. an d Riley. J. (eds): T he SAGE H andbook of W riting D evelopment (pp. 401 14) London: SAGE. Nagy, W., & Anderson, R. C. (1984). How many words are there in printed school English? Reading Research Quarterly, 19 304 330. Neugebauer, S. R. (2008). Double the work: Challenges and solutions to acquiring language and academic literacy for adolescent English language learners/the language demands of school: Putting academic English to the test. Harvard Educational Review, 78 (1), 252 263. O'Halloran, K. L. (2005), Mathematical discourse: Language, s ymbolism and visual i mages London and New York: Continuum. Paxton, R. J. (1999). A deafening silence: History textbooks and the students who read them Review of Educational Research, 69 (3), 31 5 339. Pearson, P.D., Hiebert, E.H., & Kamil, M.L. (2007). Vocabulary assessment: What we know and what we need to learn. Reading Research Quarterly, 42 282 296. Perera, K. (1990). Grammatical differentiation between speech and writing in children aged 8 12. In Carter, R. (ed.) Knowledge about language and the curriculum ( pp. 216 33 ). London: Hodder & Stoughton Persky, H. R., Daane, M. C., & Jin, Y. (2003). The nation's report card: Writing 2002 U.S. Department of Education, National Center for Education Statistics. Rampton, B., Roberts, C. Leung, C. & Harris, R. (2002). Methodology in the analysis of classroom discourse. Applied Linguistics 23 (3), 373 392. RAND Reading Study Group. (2002). Reading for understanding: Toward an R&D p rogram in reading comprehension Santa Monica, CA: RAND. Raynor, A., & Pollatsek, K. (1994). The psychology of reading Cliffs, NJ: Prentice Hall.
229 Ryan, J. (1997). Struggling with academic language: Tertiary literacy. Literacy and Numeracy Studies, 7 (2), 55 64. Scarcella, R. (2003). Academic English: A Conceptual Framework. The University of California l inguistic minority research institute t echnical Report 2003 1. Scarcella, R. (2008). Academic language: Clarifying terms. AccELLerate! T he Quarterl y Newsletter of the National Clearinghouse for English Language Acquisition (NCELA), 1 (1), 5 6. Schleppegrell, M. J. (1996). Conjunctions in spoken English and ESL writing. Applied Linguistics, 17 (3), 271 285. Schleppegrell, M. J. (2001). Linguistic f eatures of the language of schooling. Linguistics and Education, 12 (4), 431 459. Schleppegrell, M. J (2003). Grammar for writing: Academic language and the ELD Standards Retrieved from http://www.lmri.ucsb.edu/publications/finalgrantreports/schleppegrell.pdf Schleppegrell, M. J (2004). The language of schooling: A functional linguistics perspective Mahwah, NJ: Erlbaum. Schleppegrell, M. J. (2005). Helping content area teachers work with academic language: Promoting English language learners' literacy in history University of California Linguistic Minority Research Institute. Schleppegrell, M. (2009). Language in academic subject areas and classroom instr uction: What is academic language and how can we teach it? Retrieved April 2010 from http://www7.nationalacademies.org/cfe/Paper_ Mary_Schleppegrell.pdf Schleppegrell, M. J., & Achugar, M. (2003). Learning language and learning history: A functional linguistics approach. TESOL Journal, 12 (2), 21 27. Schleppegrell, M. J., Achugar, M., & Oteiza, T. (2004). The grammar of history: Enhancing content based instruction throug h a functional focus on language. TESOL Quarterly: A Journal for Teachers of English to Speakers of Other Languages and of Standard English as a Second Dialect, 38 (1), 67 93. Schleppegrell, M. & Colombi. M.C. (1997). Text organization by bilingual writers Written Communication.14 (4) 481 503. Schleppegrell M. J., & Colombi, M. C (2002). Developing advanced literacy in first and second languages: Meaning with power Mahwah, NJ: Lawrence Erlbaum.
230 Schleppegrell, M. J., & de Oliveira, L. C. (2006). An integrated language and content approach for history teachers. Journal of English for Academic Purposes 5 (4). 254 268. Schott, C. (2007). Improving adolescent literacy: Content area strategies at work. Journal of Adolescent & Adult Literacy, 51 (1), 86 87 Shanahan, T., & Shanahan, C. (2008). Teaching disciplinary literacy to adolescents: Rethinking content area literacy. Harvard Educational Review, 78 (1), 40 59. Short, D. (1994). Expanding middle school horizons: Integrating language, culture, and soc ial studies. TESOL Quarterly, 28 581 608. Short, D., & Echevarria, J. (2004). Teacher skills to support English language learners. Educational Leadership 62 (4), 8 13. Short, D., & Fitzsimmons, S. (2006). Double the work: Challenges and s olutions to a cquiring l anguage and a cademic l iteracy for a dolescent English l anguage l earner Retrieved Oct. 28, 2008; http://www.all4ed.org/files/DoubleWork.pdf Snow, C. (2008). What is the vocabulary of science? In A. S. Rosebery, & B. Warren (Eds.), Teaching science to English language learners: Building on students' strengths (pp. 71 83). National Science Teachers Association. Snow, M.A. (2005). A model of ac ademic literacy for integrated language and content instruction. In E. Hinkel (Ed.), Handbook of Research in Second Language Teaching and Learning (pp. 693 712). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Solomon, J., & Rhodes, N. (1996). Assessing ac ademic language: Results of a survey. TESOL Journal, 5 (4), 5 8. Spanos, G., Rhodes, N., Dale, T., & Crandall, J. (1988). Linguistic features of mathematical problem solving: Insights and applications. In R.R. Cocking & J.P. Mestre (Eds.), Linguistic and cultural influences on mathematics learning (pp. 221 240). Hillsdale, New Jersey: Lawrence Erlbaum. Spigler, K. (2005). Preparing for FCAT reading: Grade 9 New York: Amsco School Publications, Inc. Spycher, P. (2007). Academic writing of adolescent En glish learners: Learning to use "although". Journal of Second Language Writing 16 (4), 238 254.
231 Stevens, J. P. (1996). Applied multivariate statistics for the social sciences (3rd ed.). Mahwah, NJ: Erlbaum. Stevens, L. (2002). Making the road by walking : The transition from content area literacy to adolescent literacy. Reading Research and Instruction 41 (3), 267 278. Stevens, R. A., Butler, F. A., & Castellon Wellington, M. (2000). Academic language and content assessment: Measuring the progress of English language learners (ELLs) (CSE Technical Report No. 552) Los Angeles: University of California, National Center for Research on Evaluation, Standards, Student Testing. Szpara, M. Y., & Ahmad, I. (2007). Supporting English language learners in soci al studies class: Results from a study of high school teachers. Social Studies, 98 (5), 189 195. Tierney, R. J., & Shanahan, T. (1991). Research on the reading writing relationship: Interactions, transactions, and outcomes. In R. Barr, M. L. Kamil, P. Mosenthal, & P. D. Pearson (Eds.), Handbook of reading research vol2 ( pp. 246 280 ) New York: Longman. United States Government Accountability Office (2006). No c hild l eft b ehind a ct: Assistance from e ducation c ould h elp States b etter m easure p rogress of s tudents with l imited English p roficiency (GAO Publication No. GAO 06 815). Washington, DC. Unsworth, L. (1999). Developing critical understanding of specialized language of school science and history texts: a functional grammatical perspective. Journal o f Adolescent & Adult Literacy. 42 (7). 508 521. Unsworth, L. (2001). Teaching multiliteracies across the curriculum: changing contexts of text and image in classroom practice Buckingham, England : Open University. UsÂ— Juan, E. (2006). The c ompensatory n ature of d iscipline r elated k nowledge and English Language p roficiency in r eading English for a cademic p urposes. Modern Language Journal, 90 (2), 210 227. ValdÂŽs, G. (2004). Between support and marginalisation: The development of academic language in lingu istic minority children. International Journal of Bilingual Education & Bilingualism, 7 (2), 102 132. ValdÂŽs, G., & Figueroa, R. A. (1994). Bilingualism and t esting: A special case of bias. New Jersey: Alex Publishing Corporation. Villalva, K. (2006). Hid den literacies and inquiry approaches of bilingual high school writers. Written Communication 23 (1), 91 129.
232 Vygotsky, L. S., & Kozulin, A. (1986). Thought and language Cambridge, Mass: MIT Press. Walker, N., & Bean, T. (2004). Using multiple texts in content area classrooms. Journal of Content Area Reading 3 (1), 23 35. Warschauer, M., Grant, D., Real, G., & Rousseau, M. (2004). Promoting academic literacy with technology: successful laptop programs in K 12 sc hools. System 32 (4), 525 537. Wiley, T. G. (1996). Literacy and language diversity in the United States. Washington, DC: Center for Applied Linguistics and Delta Systems. Wineberg, S. (2001) Historical t hinking and o ther u nnatural a cts: C harting the f uture of t eaching the p ast Philadelphia: Temple University Press Wong Fillmore, L., & Snow, C. E. (2002). What teachers need to know about language. ERIC Clearinghouse on Languages and Linguistics Special Report. US Department of Education Office of Ed ucational Research and Improvement. Wright, E. L., & Kuehn, P. (1998). The effects of academic language instruction on college bound at risk secondary students. Journal of Educational Opportunity, 17 (1), 9 22. Wright, N. (2007). Building literacy communi ties of practice across subject disciplines in secondary schools. Language & Education: An International Journal, 21 (5), 420 433. Zamel, W., & Spack, R. (2008). Negotiating academic literacies: Teaching and learning across languages and cultures. NY: Rout ledge. Zwiers, J. (2004). The third language of academic English. Educational Leadership 62 (4), 60 63. Zwiers, J. (2006). Integrating academic language, thinking, and content: Learning scaffolds for non native speakers in the middle grades. Journal of English for Academic Purposes 5 (4), 317 332. Zwiers, J. (2007). Teacher practices and perspectives for developing academic language. International Journal of Applied Linguistics 17 (1), 93 116. Zwiers, J. (2008). Building academic language: Essential practices for content classrooms San Francisco, CA: Jossey Bass.
233 BIOGRAPHICAL SKETCH Ting Shen was born in 1979 in Hunan, China. She is currently an assistant professor in the department of foreign language and culture at the University of Mount Union. She received her Ph.D. from the University of Florida in the December of 2011. She recei ved the Bachelor of Arts in Chinese Language and Literature from Peking University in 2001 and the Master of Philosophy in Chinese Literature from Chinese University of Hong Kong in 2003. She was awarded Alumni Fellowship (2005 2009) from the University of Florida. She served as a Teaching Assistant in the School of Teaching and Learning at the University of Florida from 2005 to 2010.