Second language oral production in classroom contexts

¶ … SECOND LANGUAGE ORAL PRODUCTION IN HIGHSCHOOL WITHIN THE CONTEXT OF CLIL

This study is motivated by theoretical and pedagogical interests: to inform instructional design intended to integrate language and content and to explore how form and meaning intersect in SLA (second language acquisition). Both interests draw on an extensive body of research that encompasses theory and practice underlying three different yet related frameworks and lines of inquiry: content-based language teaching, form-focused instruction and attention and awareness in SLA (second language acquisition). All three of these areas are linked by a concern with the intersection of form and meaning in second language classrooms. Content-based language instruction was originally inspired as an alternative to traditional approaches to language teaching that favored form over meaning. Form-focused instruction brought language form to the foreground when meaning-focused, content-based approaches relegated the learning of language form to an incidental role. Research in attention and awareness has explored a focus on form and meaning as internal learner processes. The research questions guiding the present study were motivated by an interest in these areas.

Benefits of Second Language Acquisition

Bilingual and immersion programs are often highly content-driven. The purpose of these programs is to learn the academic subject matter while simultaneously learning the language. Examples include French immersion programs in Canada and Content and Language Integrated Learning (CLIL) across Europe. French immersion programs began in the 1960s. In these programs, non-French speaking elementary and secondary school students study all or part of their academic content courses in French, a language which shares official status with English but remains a minority language in many parts of Canada (see overview by Swain & Johnson, 1997). Typically, immersion classes are taught by the same teacher who is responsible for both language and content instruction. In Europe, CLIL has been widely adopted since the 1990's. While English, French and German are the most common target languages, over 30 target languages across the European Community are included. In CLIL, the mandate is to focus not only on language and content but also on developing skills for intercultural communication and internationalization.

These models are more moderate than either immersion or theme-based programs in terms of the degree to which they are language or content driven. They are common in universities and colleges and usually designed to help adults learn the language necessary to study specific subject matter. In sheltered programs, one teacher teaches the same content as in the mainstream program but employs strategies to accommodate learners' proficiency levels so they can work through the content. In adjunct classes, two teachers are typically involved: a content specialist teaches the subject matter and a language specialist in a separate class teaches the language that connects with the content (Burger et al., 65). Some sheltered programs, such as those at the University of Ottawa in Canada, employ a comprehension-based approach. The goal of this approach is to develop comprehension as a meaning-base and help learners with the academic language they need for the content in the program. Although the comprehension-based curriculum does not exclude a role for grammar teaching, it is situated as 'emerging' from the content, unplanned and incidental.

A primary goal of much research in the field of second language acquisition (SLA) has been to develop an understanding of how form and meaning connect in the process of second language (L2) learning: how do they intersect, what makes it happen and why does it sometimes fail to happen? An understanding of these processes is crucial in helping us design instruction that will facilitate form-meaning connections in the language classroom.

An example of empirical research that has been conducted in content-based programs for adults is situated in the post-secondary sheltered and adjunct models of bilingual education at the University of Ottawa in Canada. This bilingual university offers degree programs in both English and French. Burger et al. provide an overview of the development of this program and summarize research conducted since its early inception in 1981. Program evaluations over several years have consistently reported positive results for both language gains and content mastery by learners in the sheltered and adjunct programs. In these evaluations, language learning has been assessed with standardized global proficiency tools and content outcomes have been assessed with standardized academic achievement tests. With regard to French language proficiency, research has shown that, when compared with students enrolled in French language schools in Ontario, French immersion students achieve close to native-like levels of reading comprehension, although their fluency and written productive skills are far from native-like, a finding that will be discussed in more detail later. Tests of academic achievement have shown no difference between immersion students and non-immersion students. On the question of English language proficiency, the research has shown that within a year of English language instruction being introduced into the French immersion program, the reading skills of early French immersion students are at the same level as those of non-French immersion students at the same grade level.

Benefits of Multilinguism

In line with the assumption that some form of attention is required for language acquisition, researchers have explored questions regarding the conditions and processes of instructed SLA that might contribute to the noticing of form in meaning-based instruction. In particular, two research areas or domains have been highly influential in subsequent FFI research: interaction and output. An example of research concerned with interaction is the work of Long who examined how interaction in the classroom contributes to establishing accurate form-meaning connections. He proposed that when learners interact and communication breaks down, they are forced to negotiate the language they use to make meaning. This process is thought to enhance the comprehensibility of input each learner receives, and to provide them with an opportunity to notice linguistic form, specifically those forms they need to communicate clearly. In addition, Long argues that when the learner is engaged in interaction and paying attention during the conversation, "the chances that the learner will detect the changes, understand them, and incorporate them is likely to be higher than when both form and meaning are opaque" (Long, 453). Interest in conversational interaction between L2 learners has inspired numerous studies exploring attention to form and meaning.

Interest in the role of output has also contributed to the research agenda investigating form-meaning connections in SLA. Krashen, in line with an input-driven perspective of language learning, argued that output was a product of learning and served only to provide more comprehensible input. Swain (1985) agreed with the essential role of comprehensible input but suggested that learner output, rather than existing simply as the product of learning, also contributed to the process. She drew on research in French immersion, a content-driven, input rich context, which showed that teachers did not elicit extensive talk from the students and did not push them to produce grammatically accurate language when they did so. Swain suggested that this may be a contributing factor to the lower than expected outcomes in grammatical competence in production. She noted that when negotiating meaning, learners can make themselves understood with language that is both grammatically incorrect and socially inappropriate and she argued that learners need to be 'pushed' to use more accurate language to convey meaning. Articulated as the output hypothesis, Swain (1995) maintained that the act of producing output can contribute to learning in three ways: (a) it provides opportunity for learners to notice gaps between what they hear and read and what they are able to produce; (b) it allows language users to test hypotheses about how language works while they try to produce their own meaning; and (c) it provides the opportunity for learners to reflect on language metalingustically and to use the language to shape their thoughts. Drawing on this hypothesis, Swain and her colleagues have investigated how output can encourage learners to become more aware of their use of language form in communicative and content-based contexts such as French immersion (Kowal & Swain, 284). The link between output and noticing of specific linguistic forms has also been investigated in adult ESL (English as a Second Language) classrooms (Izumi & Bigelow, 278) and continues to be a focus of SLA research investigating a focus on form situated in highly meaning-focused contexts such as content-based language teaching.

Benefits of CLIL (Content and Language Integrated Learning).

CLIL (Content and Language Integrated Learning). is premised on the belief that language and content are inseparable in SLA, and that language is "a system that relates what is being talked about (content) and the means used to talk about it (expression)" (Mohan, 1). As a pedagogical framework, CLIL has been widely adopted as an alternative to traditional models of teaching that separated language and content. These models promoted teaching of language as the subject of classroom instruction and have been criticized as consisting of "piecemeal, bottom-up approaches" (Stryker & Leaver, 4). In adult and school-based education, it "aims at eliminating the artificial separation between language instruction and subject matter classes" (Brinton., p. 2) and offers a two for one approach (Lightbown & Spada, 206). Theoretically, CLIL draws on research that situates the integration of language and content as the relationship between form and meaning. An understanding of the theory and practice related to the content-based classroom is essential to the present study. In this section of the chapter, I outline the underlying theory and rationale commonly cited as a basis for CLIL, review empirical research that has evaluated CLIL in the classroom, and outline various approaches designed to integrate language and content.

CLIL is an umbrella term that captures a wide range of classroom models that include attention to content and language. CLIL is premised on the belief that language and content are inseparable in SLA, and that language is "a system that relates what is being talked about (content) and the means used to talk about it (expression)" (Mohan, p. 1). As a pedagogical framework, CLIL has been widely adopted as an alternative to traditional models of teaching that separated language and content. These models promoted teaching of language as the subject of classroom instruction and have been criticized as consisting of "piecemeal, bottom-up approaches" (Stryker & Leaver, 6). In adult and school-based education, it "aims at eliminating the artificial separation between language instruction and subject matter classes" (Brinton. 2) and offers a two for one approach. Theoretically, CLIL draws on research that situates the integration of language and content as the relationship between form and meaning. An understanding of the theory and practice related to the content-based classroom is essential to the present study. In this section of the chapter, I outline the underlying theory and rationale commonly cited as a basis for CLIL, review empirical research that has evaluated CLIL in the classroom, and outline various approaches designed to integrate language and content

Selinker pointed out that context may affect L2 learners' inter-language (IL) development and performance. The term content area literacy has recently come to be associated with emergent term

In the professional and research literature, support for the integration of language and content draws largely on research situated in SLA and related disciplines. For example, research in cognitive psychology exploring the link between depth of processing and memory has found that input that is semantically rich encourages greater depth of processing and facilitates recall of learning (Anderson & Reder, 403), and research in learning theory suggests that interesting content engages learners and encourages the development of an associative network (Tobias, 94). As a framework for instruction, however, CLIL has been most strongly influenced from within SLA by the work of Stephen Krashen who made theoretical claims that promoted an exclusive focus on content and meaning, and an incidental role for attention to language structure. Krashen viewed language acquisition as a process in which learners acquired structure by focusing on meaning rather than form. He argued that learners focus on the meaning and not on the form of the message and proposed that language acquisition occurs when learners are provided with comprehensible input: meaningful input that is provided via language that is just slightly beyond the learners' linguistic proficiency, thus making it challenging to the learner. According to Krashen, the best type of input is "so interesting and relevant that the acquirer may even 'forget' that the message is encoded in a foreign language" (Krashen, 66). In line with this perspective, the effective classroom was considered one in which teachers focused primarily on content with the assumption that learning of language form would follow. Examples of this include the Natural Method; the task-based approach of Prabhu's Communicational Teaching Project and some Content and Language Integrated Learning (CLIL) classrooms in which a 'language bath' stimulates language acquisition without direct instruction (Dalton-Puffer, 207).

CLIL was also influenced by the development of communicative language teaching (CLT), an approach which emerged as an alternative to language instruction in which learners learned much about the structure of language but failed to develop proficiency as language users. CLT advocated a strong role for meaning in the classroom and is premised on the notion that the goal of language learning is the development of communicative competence (Canale & Swain, 47). Consistent with this goal, CLT involves learning activities intended to prepare the learner to "communicate meanings effectively" (Littlewood, 16). Unlike Krashen's perspective, however, a role for attention to form is acknowledged: "Communication cannot take place in the absence of structure, or grammar, a set of shared assumptions about how language works, along with a willingness of participants to cooperate in the negotiation of meaning" (Savignon, 640). In a review of CLT, Spada affirms a balance of form and meaning as consistent with the underlying philosophy of CLT. This symbiotic relationship between form and meaning, as language and content, is the cornerstone of much content-based language teaching.

CLIL has also been informed and encouraged by research and pedagogy concerned with subject matter teaching in schools. In this context, Cummins drew attention to the relationship between the learning of content knowledge and learners' language proficiency. Citing research that has documented the academic challenges faced by minority language children who appear to be proficient linguistically yet unable to achieve academically at grade appropriate levels, Cummins suggested a model of proficiency that included both interpersonal communicative skills rooted in the social context and academic language skills required for the classroom. He argued that in order to help children learn the academic language, teaching must consider how content and language intersect in the classroom and plan scaffolding tasks for learners. Along these lines, Cummins proposed a framework that integrates content and language learning by considering the degree of contextual support and cognitive demand inherent in the tasks. A greater degree of context in a task makes it easier for learners to access the content with less cognitive demand while less context forces learners to stretch their linguistic resources. For example, literacy is cognitively demanding because it requires a more precise command of language without added support in order for content to be understood. Appropriately, literacy becomes a primary focus in content-based programs where documents are often used as a source of content. In this framework, content and language are integrated when tasks are planned to scaffold learners from a reliance on context and low cognitive demand, to a greater reliance on language to communicate and understand content. Cummins' theoretical and empirical research has provided support for the integration of content and language while underscoring the complexity of this relationship.

Despite general agreement on the theoretical soundness of CBLT (Content-based language teaching), the question of how to integrate language and content in the classroom has been approached from diverse perspectives and a number of proposals have emerged. One of the earliest is Mohan's model of knowledge frameworks. This model draws on a functional linguistics approach that focuses on the functions of language rather than the structural components. This model views content knowledge as framed by key concepts which are shared across topics and themes. It provides a framework that guides teachers and learners through processes in which they draw on the concepts to identify information specific to the topic content. In this way, learners develop the language to understand and communicate the content. In addition to the support afforded by the framework, graphic organizers are used to guide learners through the content.

Snow, Met, and Genesee propose a different approach to integrating content and language. They recommend that teachers begin by analyzing learners' academic and communicative needs in order to identify relationships between language and content. These relationships are framed within a perspective of 'language serving content': language that is essential to the content is 'content-obligatory' and language that naturally occurs in the content is 'content-compatible'. This activity helps the teacher decide what language outcomes to address. In this approach, language and content are ordinarily taught by different teachers who collaborate to develop language objectives to be addressed alongside content objectives. It could also be taught by one teacher who adopts a dual role.

Another perspective on the integration of language and content is reflected in The Cognitive Academic Language Learning Approach (CALLA) (Chamot & O'Malley, 94). This approach draws together content learning, language learning and explicit instruction in learning strategies. The aim of this approach is to help learners become better at learning so they can learn language and content knowledge. The content determines both the language and the strategies needed and lessons are structured to scaffold learners through the content.

Classroom tasks have also been adopted as a means of integrating language and content. Pedagogic tasks in the SLA literature have been commonly defined in terms of goal focused activities in which meaning is primary (Ellis, 353). A prominent example of this approach is Prabhu's School-based project in Bangalore, India. Prabhu adopted tasks as a means of creating communicative need in the classroom and providing a framework for activities that help learners infer meaning and solve problems using language. He argued that "meaning-focused activity ensures that any attention to form is (1) contingent to dealing with meaning and (2) self-initiated (i.e., not planned, predicted, or controlled by the teacher)" (Prabhu, p. 76). Outside of the academic context, tasks have provided a framework for professional and vocational content-based language teaching for adults. In this context, tasks have been used to provide access to the 'expert language' of occupations and professions and in this way help bridge the gap between the learners' proficiency levels and more complex language that is key to specific disciplines (Bogaert, Van Gorp, Bultynck, Lanssens, & Depauw, 128).

More recently, other researchers have suggested that the relationship between content and language is more complex than one of conduits that provide access. Barwell argues that this conduit model conceptually separates language and content by suggesting that language can be learned first and then followed by content. He also argues that this model situates language and content as static and fixed rather than variable, and obscures the social dimensions and relationships that are shaped by education. Similarly, others view content and language integrated learning as a process in which learners are socialized into communities of practice. In this model, the relationship between content and language is constructed by the participants rather than existing as a predefined body of knowledge (Davison, 237).

In addition to the range of perspectives, CBLT can be implemented via a myriad of teaching models situated in a range of contexts including schools, colleges and universities and other programs of adult education. Although diverse, the various models share a number of common features that distinguish them from other models of language teaching and learning. Firstly, with regard to syllabus, content-based programs are organized according to the content knowledge, whether this is a topic of interest, academic subject matter or vocation specific knowledge. Units are defined as content and language is typically derived from the content. In traditional language programs, the course may be organized as units of grammatical features, while in the communicative classroom, the course may be planned with specific grammatical features in mind and the content of topics and themes is chosen to provide a context in which the grammar is taught. Secondly, material and task selection take on a particular role. They are selected by the teacher in content-based classrooms as authentic representations of text and tasks that the learner would interact with outside the classroom. The material provides the source of the subject knowledge, whether it is an academic text or a workplace document. Strategies are employed to allow learners to work with the authentic material that is typically not simplified. In communicative language classrooms that are not driven by content, the teacher may choose to simplify the text in order to make it comprehensible. Finally, content-based programs are usually geared to the professional or personal interests of a group of learners. A core language class premised on a communicative framework also includes topics of interest but ordinarily these are derived from an initial analysis of learner needs and interests that the teacher tries to accommodate. Content-based programs are designed with a specific learner group in mind.

These common features distinguish content-based programs from other language instruction. However, classroom practice varies in terms of the degree to which it is driven by content or language. Rather than being absolute, this variation can be captured as a continuum that reflects 'strong' and 'weak' versions with regard to the role of content. At one end of the continuum are those programs that are distinctly content-driven: immersion and foreign language medium instruction. Their purpose is to teach content in the second language. Content determines the course goals and content knowledge is what is evaluated. Ordinarily, in such a case, the class is taught by a teacher who is primarily an expert in the content and the learners in these programs see the teacher as a source of content knowledge. Both teachers and learners are distinctly focused on content as their common purpose. At the other end of the continuum are programs that use content to teach the language. In such contexts, language determines the course goals and language proficiency is evaluated. The teacher is ordinarily a language expert with or without knowledge of the content. In such programs, content is provided via material and tasks, rather than the teacher as source, and the learners are typically more aware that they are there to learn language with content as a bonus.

Diverse terminology is used to identify pedagogic models that combine attention to language and content. Some of the more common models can be captured within four primary strands and described as follows:

Thematic programs represent a highly language driven model of CBLT. These programs use content to establish a context for language learning and teaching. The content gives shape and purpose to language learning. They can be found in a variety of settings, including university language programs and community-based language programs helping newcomers with settlement (Gianelli, 142). Language skills are spiraled throughout the program as learners work with the content.

Research in programs identified as content-based has been carried out in school-based immersion programs, university credit programs and adult education. With regard to adult education programs, the research literature has consisted for the most part of program descriptions intended to present models of best practices (Kasper, 230; Stoller, 261;Wesche & Skehan, 207). It has been largely descriptive, showcasing specific programs and exploring how integrating content and language influences program design, methodology, materials development and teacher preparation (Gaffield-Vile, 114). Research reporting outcomes has generally been designed as program evaluation and has included reports of learner satisfaction and gains in global language proficiency as indicators of program success and testimonials in support of the content-based model (Graham & Beardsley, 227).

Theorizing and research underpinning FFI (Form Focused Instruction) has been premised largely on the belief that conscious processing plays a role in SLA (second language acquisition). In the research literature concerned with FFI, the role of consciousness is most widely discussed in terms of the concept of noticing: the premise that a linguistic form must be 'noticed' in order to become part of a learners' language. This concept is widely referenced to Schmidt (1990, 1995) who articulated it as the noticing hypothesis. This hypothesis initially grew out of data drawn from the personal language diaries of Schmidt while engaged in learning Portuguese (Schmidt & Frota, 322). Analysis of Schmidt's diary data showed evidence of a connection between his noticing of linguistic form, and its emergence in production.

Drawing on this data and research in psychology and cognitive science, Schmidt proposed that "conscious processing is a necessary condition for one step in the language learning process" (Schmidt, 131). In discussing the role of consciousness, he distinguishes between consciousness as awareness, as intention and as knowledge: awareness is perception; intention describes action over which the learner has control; and knowledge is demonstrated when a learner can talk about the rules of a language. Noticing, in this model, takes place at the level of conscious awareness and can be commonly reported and reflected upon by the individual. Schmidt acknowledges that awareness can also include perception at an unconscious level but argues that conscious awareness, as noticing, is essential for language input to be processed and become part of the learner's language.

On the question of what influences noticing, Schmidt suggests that a number of factors contribute to noticing of form. These include the frequency and the perceptual salience of a particular feature: the more frequently the item appears in the input or the more salient it is, the more likely that it will be noticed. The skill level of individuals in attending to both form and meaning may also influence the likelihood that a particular form is noticed; and the demands of the task may influence processing and the focus of the learner. Factors such as these have been manipulated in instruction that is designed to effect noticing by introducing a focus on form in the classroom. While the noticing hypothesis has been criticized on the basis that it is relevant only to the development of met linguistic knowledge (Truscott, 135), it has been widely adopted as the premise for many studies in the instructed SLA literature (Norris & Ortega, 528).

The term focus on form was popularized by Long who used it to describe instruction that attempts to "overtly draw students' attention to linguistic elements as they arise incidentally in lessons whose overriding focus is on meaning" (Long,, 47). He used this term to contrast it with focus on forms, which he intended to describe approaches in which teaching is organized around a pre-planned syllabus of grammatical forms. For Long, the defining features of this alternative approach were a primary focus on meaning, and attention to form that was unplanned, arising during instruction. Other researchers, however, have since used the term form-focused instruction to describe a broader application of a focus on form. Ellis applies this term to include attempts to draw attention to form within any teaching approach, including programs that are organized around linguistic units. Spada situates FFI within meaning-based instruction and provides a definition appropriate to content-based language instruction: "any pedagogical effort which is used to draw the learners' attention to language form either implicitly or explicitly" and "occurs within meaning-based approaches to L2 instruction" (Spada, p.73). In the present study, the term FFI will be used to indicate instruction designed to draw attention to form in content-based classrooms.

Studies concerned with FFI have operational zed a focus on form via a range of pedagogical options that share a concern with conscious processing of form in a meaning-based context but capture this in a variety of ways. In attempting to capture the range of options in an organized framework, researchers have adopted diverse perspectives and terminology. In this regard, the literature on FFI includes terms such as proactive vs. reactive (Doughty & Williams, 261); planned vs. incidental; and planned vs. spontaneous (Williams, 2005). A focus on form can be planned and proactive when a teacher chooses to identify, prior to instruction, language forms that will be challenging for the learner and/or essential to the communicative activities and plans instruction to draw attention to these forms. In content-based programs, the pre-selected forms are often those that emerge from the content and intersect with the meaning. Long's original construct of incidental focus on form refers to instruction that draws attention to form as it arises in lessons that are meaning-focused. This is consistent with spontaneous and reactive FFI.

A number of reviews outline various teaching options formulated within FFI and offer taxonomies to illustrate how they reflect particular perspectives (Doughty & Williams, 261). Doughty and Williams organize various pedagogical options according to the potential to obstruct the flow of communication in the classroom. This approach is appropriate to the present study because a primary concern is the degree to which a focus on language form detracts from a focus on the meaning of content. It can be examined in terms of implicit or explicit instruction. Explicit instruction is defined in this study as "instruction aimed at inducing learners to thinking consciously about some sort of rule" (Ellis, p. 24), and implicit instruction can be defined as "providing learners with input data containing the target structure and/or with opportunities to produce the target structure, in both cases without any awareness of what the target structure is. Thus, in this case, learners are expected to process input/output for message content" (Ellis, 24). Some attempt to draw attention to form implicitly and not intrude on meaning while others are very explicit and make learners consciously aware that they are expected to pay attention to form. They include specific types of written and verbal input, task design, and different ways of providing feedback on learner errors. They are organized according to the degree to which they explicitly draw attention to form and have the potential to intrude on attention to meaning.

Content and Language Integrated Learning (CLIL) is an umbrella term used to describe a wide range of teaching models in which second languages are taught via subject matter other than language itself, for example, mathematics, social studies, psychology and other subject areas. CLIL is premised on the belief that language and content are inseparable in communication and that learning is enhanced by integrating the two in instruction. This premise has provided the framework for instructional models in a range of language education settings for both adults and children. One of the most widely known examples is Canadian French immersion programs situated in elementary and secondary schools. Other CLIL programs are situated in colleges and universities, continuing education programs for adults, and workplaces. They combine language and content by drawing on themes, academic subject matter, vocational skills, and professional knowledge.

State of the art until now related to CLIL.

As an instructional framework, Content and Language Integrated Learning has become widely accepted as effective pedagogy and draws support from both classroom practice and research. Teachers and program developers describe how content in language instruction motivates learners, provides meaningful learning, builds on past experiences, and supports the development of academic language proficiency (Cummins, 23). Numerous evaluations in foreign language programs have reported increased learner satisfaction, improvements in language proficiency, and growth in content knowledge. Similarly, empirical research in a number of L2 immersion programs has documented significant increases for learners in both language proficiency and academic achievement.

Given these successful results, it is not surprising that CLIL programs have grown in popularity and have been adopted in a variety of contexts, all premised on the benefits of the integration of language and content. Yet, instruction varies widely with regard to the focus given to content or language as teachers attempt to achieve a balance between the two. A range of frameworks and approaches has emerged: Mohan's Knowledge Framework involves identifying key concepts and structures that comprise content knowledge; Snow, Met, and Genesee have proposed that teachers begin by analyzing learners' academic and communicative needs and address 'content-obligatory' language objectives;, and the Cognitive Academic Language Learning Approach (CALLA) promotes explicit instruction in strategies in order to help English language learners in schools learn language and access academic content (Chamot & O'Malley, 207). Thus, questions remain as to how best to actualize the pedagogical links between content and language.

In addition, despite the many positive results from the Canadian French immersion programs, research has shown that the productive abilities of learners in the areas of grammatical and sociolinguistic knowledge were less developed than expected after many years of rich content-based instruction via subject-matter teaching. Subsequent classroom observations revealed that instruction did not effectively integrate language forms with content, leading to the conclusion that "not all content teaching is necessarily good language teaching" (Swain, 1988, p. 68). Efforts to better combine the two have led to considerable research in French immersion programs. Some of this research falls into the broader category of form-focused instruction (FFI), an approach to achieving the integration of content and language by drawing attention to language form in meaning-oriented language teaching such as content-based programs.

A basic premise underlying FFI is the claim that attention to language form is necessary for language acquisition (Schmidt, 322). In the classroom, it is implemented via a range of pedagogical options from the very implicit, in which learners are exposed to language form indirectly, to very explicit instruction that includes meta-linguistic explanations of grammar. The common feature is that both are intended to integrate attention to form with attention to meaning. Research investigating the overall effectiveness of FFI has generally been positive and it is widely accepted that instruction that connects form and meaning in instructed SLA is more effective than instruction that focuses exclusively on meaning. However, findings have not been consistent and a number of more specific questions related to FFI have merited attention. These include: what types of FFI are most effective; which linguistic forms benefit most from FFI; and to what degree should a focus on form be integrated into meaning-focused instruction (Spada & Lightbown, 207).

Similar to French immersion programs, research in adult CLIL programs has revealed that a focus on language form has not been widely adopted and there is concern that explicit attention to language will detract from attention to content in such programs .This is consistent with the claim that learners experience difficulty attending to both the structure of language and the meaning of content at the same time (VanPatten, 301).

To date, FFI research in adult CLIL contexts has been limited. FFI research situated in content-based language programs has most commonly been carried out with children in French immersion programs (Lyster, 37). Studies which have been carried out in CLIL programs for adults have generally not investigated the effects of FFI on learning outcomes. Therefore, questions remain as to the effect of FFI on both language and content learning in content-based programs with adults.

The role of Oral Practice

Oral practice is very important in language learning. There are many interventions and classroom strategies adopted to improve students' oral production. In a more recent study, two cohorts of students, one enrolled in an English language adjunct program and the other in a French language adjunct were involved in a two-year study designed specifically to measure the development of oral proficiency. As a comprehension-based program, the main focus of instruction is to help learners access the content of the lectures and texts used in the subject matter courses. Grammar is described as incidental, dealt with as it arises in the lesson and always in context. Language outcomes were assessed by the language teacher using language measures closely aligned with the content. Subject mastery was assessed by the subject matter teacher using tests and exams that were a standard part of the mainstream course. The results showed that students in this program met or exceeded language proficiency gains and subject mastery when compared to students in the mainstream programs. In school-based programs for children, extensive research has been carried out in Canadian French immersion programs. Evaluation in French immersion programs has addressed questions of French language proficiency development, academic achievement, and the effect of French immersion on English language development (Lyster, 66).

Importance of ICT related to Oral Production

Although reading researchers have demonstrated the benefits of effective literacy instruction in the content areas, too often young adolescents miss out on such instruction. In fact, secondary grades (6-12) are filled with adolescents whose reading ability is below grade level. Specifically, according to national testing, approximately eight million students do not read at grade level (Biancarosa & Snow 16). Further results from the 2009 NAEP Reading Report indicate that 75% of young adolescents entering the ninth grade were reading below grade level in 2001, the IRA and the NMSA issued a joint position statement supporting young adolescents' literary learning. It stated that elementary students' had achieved grade level literary skills in the lower grades; however, the level of student literary achievement declined during the middle school and high school years. As a result, adolescents' low levels of literary achievement limited their academic success in school.

When young adolescent arrive at the middle grades, increasing demands are made on their literary skills; they are expected to read and comprehend greater amounts of knowledge from informational or expository text. In many middle schools, reading instruction is no longer provided for students. The purpose of reading changes to acquiring knowledge through content area learning, and students cannot master the complex vocabulary and syntax contained in the content area textbooks. Daniels and Zemelman assert that textbooks, which dominate the secondary-level content area classes, were hard to read and were not written for students. Due to the inordinate amount of content information presented in textbooks, students must be taught comprehension-fostering strategies, monitoring procedures, and critical thinking skills, which help to develop adolescents' vocabulary development, self-monitoring skills, and comprehension strategies (Flood, 931).

2. Hypotheses

Through this researcher the researcher aims to investigate the effectiveness and effect of FFI on both language and content learning in a specific CLIL program, one designed to increase the oral production of high School in second language acquisition. The study is concerned with three aspects of FFI instruction in content-based programs. Firstly, as such programs are designed to achieve gains in language proficiency, it is essential to establish the effectiveness of FFI in this context. The first question is then: What effect does FFI have on language learning in a content-based language program for adults? Secondly, because content learning is often a shared goal in such programs, the second research question asks: What effect does FFI have on content learning in a content-based language program for high School Students? The third research question investigates the potential conflict between attention to language and attention to content. This question asks: What effect does FFI have on learner awareness of language and content in this context?

The research questions guiding the present study are concerned with form-meaning connections framed by the relationship between content and language and situated in classroom instruction that integrates FFI with content-based teaching. The literature review shows that early research in classrooms designed to integrate a focus on content with a focus on language has been positive: Canadian French immersion classrooms showed that students improved their language proficiency and continued to achieve academically, while studying content in an additional language. Research in European CLIL programs has also reported favorable results for both language learning and academic achievement (e.g., Van de Craen, Mondt, Allain, & Gao, 78) and studies situated in university programs with adults have shown gains in both language proficiency and content mastery (e.g., Burger & Chretien, 84). The role of meaning was primary in these classrooms and much of the instruction was guided by the assumption that language acquisition is a by-product of content mastery. Yet, when researchers took a closer look at how language proficiency has been defined and measured in some of these contexts, evidence emerged that not all areas of proficiency were in fact well developed (e.g., Harley, Cummins, Swain, & Allen, 25; Harley & Swain, 311; Swain, 253). This raised the question as to why meaningful input via content-based instruction may not enable learners to develop sufficiently high levels of L2 knowledge and ability in all aspects of their L2 proficiency.

While the value of meaningful input is widely accepted in SLA practice and research, an extensive body of research has investigated the role that conscious attention to form may play in language learning, specifically with regard to developing linguistic accuracy in meaning-based classrooms. Research investigating the effectiveness of FFI has generally shown that a focus on form and meaning is more effective than a focus on meaning alone. However, the impact of numerous variables has also emerged, including how a focus on form is implemented, when it is introduced, whether the effectiveness of FFI depends on the type of language feature taught, and what type of knowledge results from FFI. Questions such as these continue to set the agenda for research in FFI (see Ellis, 236, Norris & Ortega, 528; Spada & Lightbown, 207; Spada & Tomita, 308; Williams, 692). Most significantly, for the present study, context has emerged as an important variable that can influence the effectiveness of different options in FFI. An example of this is the effectiveness of recasts which have been found to be effective in laboratory contexts where the feedback is more salient than in content-based classrooms in which learners may not notice recasts as corrective feedback (e.g., Lyster & Ranta, 66; Nicholas, Lightbown, & Spada, 758). Another example is the use of tasks that require learners to talk explicitly about the form: some researchers argue that such tasks are appropriate in foreign language programs in which learners are already inclined to focus on form (Fotos & Ellis, 628).

The present study will bring context to the foreground. A great deal of the research in the FFI literature has been carried out in communicative foreign and second language classrooms with a 'weak' role for content, driven by themes or topics. When situated in content-based classrooms with a 'strong' focus on content, it has primarily been carried out in academic contexts such as school-based immersion and university programs. There is no empirical research, to my knowledge, which has investigated the effectiveness of a focus on form in a classroom with adults studying vocational language for employment. This context presents a unique opportunity to investigate how learners respond to a focus on form when engaged in what may be considered a high-stakes educational endeavor, i.e., learning language and content for their livelihood. While the present study does not intend to contrast the vocational context with the academic context, evidence of effective FFI in this additional context would be a valuable addition to the existing research literature on CLIL. The current body of research suggests that the context will demand specific options but will show that a focus on form and meaning is more effective than a focus on meaning alone. The first hypothesis of the present study is:

H1: Introducing a focus on form in a meaning-focused content-based language program will result in greater gains in grammatical knowledge, as measured by written grammar correction and cloze tasks and picture-elicitation oral production tasks, than instruction focused on meaning alone.

An aspect of the content-based classroom that distinguishes it from other communicative classrooms is the concern with content learning, either on the part of the learner or on the part of both the learner and the program. Early research in immersion classrooms asked: does a focus on form have a negative impact on content learning? Studies in immersion classrooms have shown that, when compared to non-immersion students, academic achievement is not negatively affected. Other research, however, has taken a more fine-grained approach and has investigated how specific options affected learner attention to content. Studies such as these have had mixed results with some research showing that comprehension of content was negatively affected by what is widely considered an implicit option in FFI, input enhancement (Lee, 118), and other research showing that the same option appeared to enhance content learning (Wong, 45). Only one experimental study, to the best of my knowledge, has explored this question in a content-based classroom with adult learners: Grim found that explicit FFI did not negatively affect content outcomes. Further research in CLIL is needed. Drawing on the existing research, the second hypothesis that will be investigated in this study is:

H2: Introducing a focus on form in a meaning-focused content-based language program will have no effect on content learning, as measured by discrete point content knowledge tests, when compared with instruction focused on meaning alone.

Related to both the learning of content and language is the question of attention and awareness. The essential role of conscious awareness has been widely accepted in SLA. Yet more research has been carried out premised on this belief than attempting to investigate the construct itself. Some researchers have explored how internal learner processes related to attention and awareness can be captured and measured and, furthermore, if they can be linked to the process of learning (see overviews by Egi, 264; Ellis, 305; Mackey & Gass, 430). There is evidence that awareness of form, as reported by the learner, correlates to gains in language learning (e.g., Mackey, 2006; Palmiera, 1995; Rosa & O'Neill, 1999). This body of research, however, is limited, and has not included measures intended to capture an awareness of content as well as language, as in the present study. Awareness of content is important in the present study because of the concern that a focus on language detracts from a focus on content. Along these lines, the present study will investigate the following two hypotheses:

H3: Learners will be able to report awareness of both content and language in instruction that includes both a focus on grammatical form and content meaning.

H4: Learners reporting higher degrees of language awareness will show stronger outcomes on the language measures than learners who report a lower degree of awareness.

The present study has been designed to investigate each of these hypotheses in a content-based program in a high School

3. Research Instruments/Methodology

This classroom-based study will use a quasi-experimental, pre-test, immediate, and delayed post-test design. The research questions will be investigated by comparing the effects of two different instructional treatments on language and content learning outcomes in this particular teaching context. A learner awareness protocol will be developed to investigate learner awareness of language and content in their instruction.

3.1 Schedule

The study will take place in two intact language classes over 27 weeks of a 40-week program. Both classes will be designed for learners in the same range of proficiency levels, CLB 3 to 5, low- to mid-intermediate, but will be offered at two different times so that learners may select a class based on their availability. Both classes will met for 5 hours per week, all day Saturday or 2 evenings per week, and both classes will be taught by the same teacher.

Before conducting the study the program coordinator, the teacher and the learners in both groups will be asked for consent to participate in the study. After gaining consent the learners will be enrolled in both groups and will be asked to complete a questionnaire about their educational backgrounds. Pre-tests to assess language and content knowledge will be completed over a period of three weeks.

The two classes will be randomly assigned as either the group receiving FFI, hereafter referred to as the FF group, or receiving only meaning-focused instruction with no FFI, hereafter referred to as the MF group. The Saturday class will be selected as the FF group and the evening class was selected as the MF group. The differentiated instruction will took place over 10 weeks between the pre-tests and the post-tests. The teacher will be given instructions and material to continue providing meaning-focused instruction to the MF group and differentiated instruction to the FF group during these 10 weeks. During the period between the post-tests and the delayed post-tests, approximately 13 weeks, he/she will be explicitly asked not to provide any focus on form to either group. In order to monitor the instructional treatment, all the lessons for both groups from pre-test to post-test will be audio recorded and the data coded to identify a focus on language form or content.

3.2 Data Collection Instruments

In order to assess the effect of FFI on content and language learning, quantitative data will be gathered using content and language outcome measures. Content knowledge will be measured via a pre-test. These tests will consist of true/false, multiple-choice, and short-answer questions.

Language outcomes will be measured with four tasks including two paper and pencil tasks and two oral tasks:

an error correction (EC) task assessing knowledge of both target linguistic features a cloze task assessing knowledge of both target features two oral production tasks (OPT), each measuring the ability to use one of the two target features

A variety of measures will be developed in order to provide a fuller assessment of the learners' grammatical knowledge and ability under varying conditions. The measures include selected response tasks, limited response tasks, and extended production tasks. Those tasks that call for responses that are more limited require learners to draw on knowledge of grammar form. The extended production tasks draw on knowledge of form and an understanding of grammar in use (see Purpura, 2004 for an overview). These task types also differ in terms of the degree to which they draw on more spontaneous or more analyzed language. For example, in this study, the oral production tasks will be designed to assess learners' ability to use the L2 in spontaneous language production while the written tasks will be designed to measure learners' conscious knowledge of the L2..

3.2.1 Error correction task

The first linguistic measure that the current study will use is the error correction (EC) task. EC tasks are considered good measures of analyzed language knowledge (e.g., Ellis, Sheen, Murakami, & Takashima, 2008; Sheen, 2007; Wong, 2003). The EC task in this study will focus the learner's attention explicitly on form. The task will be administered in the classroom by the researcher and learners will be given specific time to complete it. The same version of the task will be used for the pre-test, post-test, and delayed post-test to allow a direct comparison of the learners' performance over time.

3.2.2 Cloze task

The second written task that the researcher will use is a selective deletion cloze task. The cloze has a long-standing history as a measure of language development, and has been investigated as a valid measure of both global proficiency (Lapkin & Swain, 1977; Oller, 1973) and grammatical competence (Oller & Inal, 1971) with positive results. Selective deletion cloze tasks, in which a specific grammatical feature is deleted, have been used in a number of empirical studies (e.g., Harley, 1989; Hinkel, 1997; Leeman et al., 1995; Rodgers, 2006; Salaberry, 1998).

3.2.3 Oral production measure

The oral production tasks will be designed to elicit more spontaneous use of the linguistic forms. Each of the two oral production tasks will be designed to elicit one of the linguistic target forms in a task-natural context familiar to childcare providers. Both tasks will consist of picture elicitation tasks and will be completed one-on-one with the researcher and digitally audio recorded. The researcher will guide the learner through the task by prompting them to move to the obligatory context until all the contexts are completed.

3.2.4 Content Knowledge Measures

Empirical studies investigating the effect of a focus on language form in content-based language instruction commonly use two types of means to measure content outcomes: measuring comprehension of content (e.g., Leow et al., 2003), and drawing on learner recall of content (e.g., Lee, 2007; Wong, 2003). There are challenges with both types and disagreement as to which is most effective as a measure of content learning. The content measures in this study will use elements of both types as they will require learners to demonstrate an understanding of course content and to recall discrete items addressed during instruction. The primary concern in developing the tests is to mitigate the disruption to the program by providing a type of test that the teacher customarily use in the course to review content with the learners.