Vocabulary instruction for English language learners in grades three through five: A multivocal synthesis
iii TABLE OF CONTENTS
CHAPTER ONE INTRODUCTION 1
CHAPTER TWO LITERATURE REVIEW 8 Interactive Oral Reading 10 Direct and Systematic Instruction 10 Text Talk 11 Anchored Instruction 12 Reading Aloud to Older Students 14 Semantic Relatedness Techniques 15 Personalized Word Learning 19 Mnemonic Strategies 19 Self-Selection of Words 21 Fostering Incidental Word Learning 22 Word Consciousness 24 Teaching Individual Words 26 Identifying Words for Instruction 27 Methods for Teaching Individual Words 28 Computer Assisted Learning 31 Vocabulary Instruction and ELLs 33 Recent Studies 34 Lesson Learned from the Research 36 Taking advantage of student’s first language 37 Basic words 38 Review and Reinforcement 39 Multivocal Synthesis Models 40
CHAPTER THREE METHOD 45 Design 45 Theoretical Lens 49 Researcher Background 49 Setting 50 Participants 51 Professional Work Group Participants 52 Student Work Group Participants 52 Data Collection Procedures 53 Data Analysis 55
CHAPTER FOUR THE VOICES 56 Recommendations for Effective Practices in Vocabulary Instruction 57 Building Background Knowledge 57 Accessing Prior Knowledge 59 Personal connections 59 Cultural connections 61 Native language connections 62 Frontloading 64 Previewing 65 Pairing fiction with non-fiction 66 Realia 67 Videos 68 Experiential Learning 70 Visuals 72 Movement 72 Experiments 74 Field Trips 74 Thematic units and planning 75 Differentiation 78 Planning 80 Basic Words and Concepts 82 Process Words 84 Adaptation of Text 86 Small Group Instruction 87 Interaction 89 Purposeful Pairing or Grouping 91 Cooperative Learning 92 Centers 93 Book Clubs or Literature Circles 94 Social Interaction 95 The Affective Environment 98 Honoring Student Knowledge 99 Community 101 Motivation 104 Motivation by Teachers 104 Motivation by Parents 108 Motivation by Peers 109 Problems Identified in Current Practice 111 Issues in Computer Assisted Instruction 111 Format Issues 113 Issues with Interaction 115 Positive Features 118
v Teacher-Centered Issues 120 Time 121 Issues with Teacher Preparation 122 Student-Centered Issues 125 Difficulties with the Dictionary 125 Loss of Native Language 131
CHAPTER FIVE DISCUSSION 135 Where the Voices Bled Together 135 Building Background Knowledge 136 Accessing Prior Knowledge 136 Frontloading 138 Experiential Learning 138 Differentiation 139 Planning 139 Basic or High Frequency Words 140 Small Group Instruction 142 Interaction 143 Teacher-Centered Issues 143 Conclusion 146 Suggestions for Further Research 146 Limitations of the Study 147 The Landscape: Implications for Practice 149
vi LIST OF TABLES
Table 4.1. Building Background Knowledge 78
Table 4.2. Differentiation 88
Table 4.3. Interaction 97
Table 4.4. Affective Environment 103
Table 4.5. Motivation 110
Table 4.6. Issues in Computer Assisted Instruction 120
Table 4.7. Teacher-Centered Issues 124
Table 4.8. Student-Centered Issues 134
Table 5.5. Data Convergence Table 145
1 CHAPTER ONE Introduction “It would be like me coming to school wearing a bright pink bikini. That would definitely be improper!”
Those words, along with a torrent of giggles, rang in my head one evening as I was preparing for the next day’s lessons. I was water coloring a large mural to be used as a backdrop for an interactively written word wall. I had uttered those lighthearted words earlier during the day, in the midst of a review of proper and improper fractions. I said them knowing that using a humorous description of what the word “improper” means would not only entertain my students, but would help them to remember it. I said them hoping these words would in some way become theirs, as would the mental picture they had painted to hang them on. While I painted the mural, I reflected on the vocabulary instruction for which I was planning. Many voices swirled in my mind. I could hear other silly examples of improper events that had been elicited from my English language learners (ELLs). Students, like David, who had excitedly proclaimed, “Improper would be like throwing a pie in the principal’s face!” As I painted the foreground, I heard the voice of my administrator reinforcing the expectation that key vocabulary words be displayed in the classroom. I heard a teacher trainer reminding me that these words need not only be seen, but read, written, and spoken by students as well. Adding to the background, I heard the voice of the late scholar, Stephen Stahl (2004, p. 59), who wrote that “to expand a child’s vocabulary is to teach that child about the world.” I heard the echoing of this sentiment by a colleague who once told me that words were the best gift I could give my students.
2 With each stroke of my brush, the voices of students, researchers, and educators blended in my thoughts. Having completed my mural, I stood back and looked it at from several angles. I observed how the different hues bled beautifully together in places, yet remained brilliantly separate in others. I admired how all these pigments had come together to create a painting; just as those swirling voices of researchers, students, teachers, and administrators had blended in my thoughts. What landscape of vocabulary instruction would these water colors, these voices paint? By using these voices as my medium to create a living piece of art, could I see where these varying shades, the disparate underlying beliefs of stakeholders bled together? These stakeholders include David, who is just one of the approximately 4.7 million students who meet the criteria for ELL designation in U.S. schools (NCELA, 2006). He is part of a population that has grown approximately 105%, while the general school population has only grown by 12% (National Center for Educational Statistics, 2004). As a researcher and his English as a Second Language (ESL) teacher, I am also a stakeholder. I am one of many educators and members of the academic community for whom the achievement of this dramatically increasing population is of great concern (Brown, 2007). One of the most persistent research findings is that students’ vocabulary knowledge relates strongly to their overall academic success (Lehr, Heibert, & Osborne, 2005). The importance of vocabulary in reading achievement has been recognized for more than fifty years (National Reading Panel, 2000). Specifically, August and Hakuta (1997) found that vocabulary knowledge is the primary determinant of reading
3 comprehension for ELLs. Jimenez (1994, p. 36) identified vocabulary as the “single most encountered obstacle” for ELLs when they are expected to read and learn from texts. However, vast numbers of children from homes in which English is not the dominant language arrive at school with insufficient vocabulary to support literacy acquisition (Moats, 2001). For example, Umbel, Pearson, Fernandez, and Oller (1992) tested the receptive vocabulary of 105 Hispanic first grade students in Miami. The bilingual students whom were designated as middle to high socioeconomic status relative to national norms were divided into groups according to the language spoken in their homes (English and Spanish, or Spanish only). Both groups performed near the mean in Spanish. The group from bilingual homes scored more than one standard deviation higher in English than the Spanish group. However, both groups were significantly below the mean of the norming sample in English, even when the socioeconomic status of the English learners was higher than that of the norming sample. Not only do ELLs enter school knowing fewer words than their English only (EO) speaking classmates, ELLs have been shown to be impaired in depth of word knowledge, even for frequently occurring words (Verhallen & Schoonen, 1993). A study by August, Carlo, Lively, Lippman, McLaughlin, and Snow (1999) examined how the depth of ELL vocabulary knowledge compares to that of native speakers. Students were given two separate tasks that assessed their understanding of common multiple meaning words (i.e., bug, hand, grow), which is an indication of depth of word knowledge. The results indicated a gap in scores for both tasks with ELLs scoring approximately half as well as their EO peers.
4 In summary, previous research indicates that ELLs have limited breadth of vocabulary. In addition, research suggests that ELLs lack depth of vocabulary knowledge. Finally, there is a strong relationship between overall vocabulary knowledge and academic success (Lehr et al., 2006). Despite the importance of vocabulary to comprehension for ELLs, there have been a limited number of studies conducted since 1980 examining the effectiveness of interventions designed to build vocabulary among language minority students learning English (Calderón, August, Slavin, Duran, Madden, & Cheung, 2005). A recent report of the National Literacy Panel on Language-Minority Children and Youth (2006) included only three experimental studies of vocabulary learning (August & Shanahan, 2006). This paucity of research is in contrast to the wealth of research on monolingual English speakers (National Institute of Child Health and Human Development, 2000). The National Reading Panel Report (2000), which identifies vocabulary as a key component in reading, included over 45 experimental intervention studies focusing on this topic. However, the panel made a decision not to include scientific literature available in the development of literacy for students for whom English is not their native language (NRP, 2000). The increasing number of ELLs in schools suggests the need for a research focus on which instructional methods are most effective with students for whom English is not their first language (Lehr et al., 2003). In addition, Yzaguirre (1998) notes that high levels of political passion surrounding this wave of immigration has resulted in a lack of practical discourse aimed toward instructional issues on the topic of how to educate this growing population of students.
5 Ideally, a body of research and discussion focused on vocabulary instruction would inform the practice of those directly engaged in fostering the academic achievement of ELLs on a daily basis. However, Shanahan and Neuman (p. 205, 1997) point out that “views of research as the determining factor in education supposes a hierarchal field in which researchers are at the top of the pyramid, making decisions about how teachers, lower in status, will conduct their work.” Although research can and should influence instructional practice, this is not always the case because research is only one of the many sources of information along with teacher lore, experience, and community values that influence instruction (Shanahan & Neuman, 1997). The voices of the academic community are not always heard by those working in the field. Conversely, the voices of those working in the field are not always heard by the academic community. In his writings on bridging the gap between the academic community and practitioners, Weisbuch (2007, p. 4) asserts “that a strong teacher in the schools knows a great deal more about pedagogy than we (those in academia) do” and stresses the importance of acknowledging the value of input from educators in the field. I further suggest that this input should include students. As an experienced teacher, I have observed students reflecting and articulating a great deal about their own learning. Their observations would also provide information on effective pedagogy. Gersten and Baker (2000) asserted that research on effective instructional approaches for ELLs is appropriate for multivocal synthesis techniques because there is little data documenting the variety of serious perspectives. A multivocal synthesis is a research method that incorporates multiple data sets from various sources, such as professional literature and the perspectives of different types of stakeholders who are
6 knowledgeable about the studied topic (Ogawa & Malen, 1991). The result of a multivocal synthesis is not intended to reach definitive conclusions, but to garner a wide range of perspectives about a phenomenon of interest by examining disparate data sources. Gersten and Baker (2000, p.75) point out that these varying sources “make direct reference to perceptions acquired from a rich mix of informants representing different positions in the system and different perspectives of the phenomenon (e.g., practitioners, academics, participants, observers).” The participation of these stakeholders strengthens the validity of the interpretations that emerge and provides an important link between research and practice (Pressley, 1996). In this study, a multivocal method and its guiding principles were the brushes used to paint a landscape of the phenomenon of vocabulary instruction for ELLs in grades three through five. The disparate data sources which included the perspectives of researchers, ELL students in the upper elementary grades, and those directly responsible for their achievement were the watercolor paints. These participants included researchers, administrators, teacher trainers in the area of ESL, classroom teachers, ESL specialists, and ELLs in grades three through five. The results are an interpretive mural that highlights the overlapping hues of these stakeholders. Data were collected through a review of current literature and the implementation of professional work groups. Professional work groups are a type of focus group. Professional work groups differ from focus groups in that they are comprised of professionals from a given field (Gersten & Baker, 2000). The goals of professional
7 work groups are to identify themes and problems in current practice and to pose recommendations for best practice (Gersten & Baker, 2000). Student work groups were also included so their voices were not lost in the creation of this mural. These work groups are intended to provide an opportunity for the type of discourse described as lacking by Yzaguirre (1998). The following research questions were addressed: 1.) What perspectives do researchers, administrators, teacher trainers, teachers, ESL specialists, and language minority students in grades three through five have regarding vocabulary instruction for ELLs? 2.) In what areas do these multiple perspectives converge?
8 CHAPTER TWO Literature Review “So, what is a word? Well, it depends. It depends on your purpose for asking the question and your view of thought and language.” Baumann, Kame’euii, and Ash, 2003, p.754
The lack of agreement about the nature of vocabulary terms plays havoc with research and theory in vocabulary (Marzano, 2004). Individuals have various types of vocabulary that they use for different purposes such as in social situations, and in academic or occupational settings (Kamil & Heibert, 2004). Kamil and Heibert (2004) point out that it is failure to distinguish among the different kinds of vocabulary that leads to confusion about research findings and instructional implications. Thus, these authors generically define vocabulary as the knowledge of meanings of words. This general description has been adopted for the discussion of vocabulary instruction in the present study. Several issues consistently underlie the discussion in research on vocabulary instruction. First, students must learn a great deal of words and there are too many words to teach through direct instruction (Blachowicz & Fisher, 2000; Graves, 2006, Kamil & Heibert, 2004; Stahl & Nagy, 2006). Accordingly, students need multiple exposures to words (Blachowicz & Fisher, 2000; Graves, 2006; Kamil & Heibert, 2004; Stahl & Nagy, 2006). However, there is no specific formula for which words should be taught (McKeown & Beck, 2003; Kamil & Heibert, 2004). In addition, the role of indirect instruction and incidental vocabulary learning is discussed (Blachowicz & Fisher, 2000; Graves, 2006, Kamil & Heibert, 2004; National Reading Panel; 2000). Lastly, differences
9 in vocabulary learning of two groups of students, ELLs and those potentially at risk, is an instructional issue (Blachowicz & Fisher, 2000; Graves, 2006; Kamil & Heibert, 2004). Considering the complexity of these issues, the National Reading Panel (2000) has pointed out that depending on one single method, whether it is direct or indirect instruction will not result in optimal learning. In addition, there are many dimensions on which vocabulary instruction can be categorized because typically few vocabulary studies distinguish themselves by their differences from, rather than their similarities to other methods (National Reading Panel, 2000). For example, Gersten and Baker (2000) and Slavin and Cheung (2003) indicated that instructional practices that are effective with native English speakers are likely to be effective for ELLs. Many instructional practices described in the existing studies focusing on ELLs build on a number of vocabulary instructional practices that have been effective with English only (EO) learners (August et al., 2005). However, even though vocabulary instruction designed for ELLs can follow the principles that guide practices for native speakers, there are some modifications needed to fully meet the needs of ELLs (Carlo, 2007), and these are discussed in literature specifically regarding the instruction of this population of students. Thus, this review included research in published books and juried journals that focused on vocabulary instruction for elementary school students in general and for ELLs specifically. Seven topics were used to organize this research review: Interactive Oral Reading; Semantic Relatedness Techniques; Personalized Word Learning; Fostering Incidental Word Learning; Teaching Individual Words; Computer Assisted Learning; and Vocabulary Instruction for ELLs.
10 A discussion of earlier studies that employed multivocal synthesis methods was also included. The purpose was not to discuss the results or findings of these studies, but rather to examine the use of multivocal methods in previous work. These studies served as a model, so that I could apply multivocal methodology and its guiding principles to the present study. Interactive Oral Reading Reading storybooks to children has been shown to be an ideal means of introducing them to new words in context (Beck & McKeown, 2001). Much of the research on oral reading was based on the idea that active participation on behalf of the child is necessary for reading aloud to be effective (Adams, 1990; Baker, Schler, & Mackler, 1997; Beck, McKeown, & Kucan, 2002; Biemiller, 2001, 2003; Biemiller & Boote, 2006; Juel & Deffes, 2004; Snow, Tabors, Nicholson, & Kurland, 1995). Graves (2006) identified three main interactive oral reading approaches used with elementary school aged children. Direct and Systematic Instruction (Biemiller, 2001, 2003), Text Talk (Beck, McKeown, & Kucan, 2002), and Anchored Instruction (Juel & Deffes, 2004) are researched and theoretically sound ways of providing special help for elementary grade school children, including ELLs, who enter school with relatively small vocabularies (Graves, 2006). Direct and Systematic Instruction Direct and Systematic Instruction (Biemiller, 2001, 2003) is an interactive oral reading technique that includes direct instruction of words selected from narrative children’s books. It is intended for kindergarten through second grade children. Careful consideration is given to book selection and word selection. Each book is read and reread
11 several times during a five day period. About 24 words from each book are selected for instruction. Words are defined as they are encountered in the text. Words taught are reviewed at the conclusion of a lesson by repeating the sentences in which they appeared and restating the definition. Biemiller and Boote (2006) recently conducted two studies of vocabulary that used this method. Results indicated that a substantial number of word meanings can be learned using repeated oral reading of stories combined with explanations and reviews of words. In addition, the studies showed that children could understand word meanings when tested using context sentences different from the story used for instruction. Overall, it appeared that the levels of initial word knowledge had little effect on word knowledge gained. Text Talk Text Talk (Beck, McKeown, & Kucan, 2002; McKeown & Beck, 2003) is an interactive book reading procedure designed to promote comprehension and language development in kindergarten through second grade. It involves the selection of texts that have an event structure and enough complexity to prompt discussion and higher level thinking. Open ended questioning is used to encourage students to explain, elaborate, and formulate their own questions about a text. Extensive work with sophisticated vocabulary occurs only after the complete book has been read and discussed. The meanings of three or four words are given, with examples of how each word is used. Students are encouraged to generate their own sentences for each word. Each child’s use of the word over time is recorded on a chart.
12 From working with and observing teachers as they implemented Text Talk, Beck et al. (2002) pointed to several concepts that guide the development of effective read aloud experiences. Teacher must have an understanding of the difficulty young children face in gaining meaning from decontextualized language. In addition, teachers need to be aware of how pictures can draw attention away from processing the linguistic content and thus pay attention to the timing and use of pictures. Finally, teachers must take advantage of the sophisticated words found in books and use them as a source for explicit vocabulary instruction. Anchored Instruction Anchored Instruction (Juel & Deffes, 2004) is a vocabulary technique used during read alouds with kindergarten and first grade students. It is motivated by two primary considerations. The first consideration is that students do not know all the words being read aloud and that special attention needs to be paid to words that the students might not know. The second consideration motivating Anchored Instruction is that thoroughly teaching words is multifaceted in that it deals with the context in which the word occurs, the word’s meaning, and some aspects of decoding and spelling. Anchored Instruction (Juel & Deffes, 2004) follows a four step procedure. Books are selected that are at an appropriate level for a read-aloud, yet include some potentially useful words that some students might not know. Eight to 10 words are selected from the text and individual word cards are made for each student. Attention is called to some of the component letters and sounds. The book is read aloud, pausing for instruction as each word is encountered.
13 Juel and Deffes (2004) designed a six-week quasi-experimental study to compare three methods of vocabulary instruction, including Anchored Instruction. The researchers found that Anchored Instruction enabled students to learn the words in the curriculum more effectively than did the other methods, regardless of the students’ general vocabulary knowledge or background characteristics. The findings from this study provided a rationale for promoting the use of multidimensional vocabulary instruction during read alouds. This is because the focus was not only on the meaning of the words in context of the books, but also in the comparison to other words and other contexts. The students were also exposed to the written and spoken forms of the words. The researchers concluded that students benefit from being exposed to multiple sources of information and that this helps them to remember new words they are learning. Research (Beck et al., 2002; Biemiller, 2001, 2003; Juel & Deffes, 2004) indicates common factors for success with using these types of oral reading techniques. Careful attention must be given to the selection of texts and words targeted for instruction. Books that are at an appropriate level, unfamiliar, and interesting to the students should be selected. Words that are useful and unfamiliar should be focused on. Students can benefit from repeated readings of a text and explicit instruction of words. Specific words in a text should be focused on in a systematic way. The ways in which words are focused on in each approach varies. Direct and Systematic Instruction (Biemiller, 2001, 2000) and Anchored Instruction (Juel & Deffes, 2004) techniques involve instruction of words while the text is being read. In addition, several words are selected for instruction (8-24) in both of these techniques. However, Text Talk (Beck et al., 2002) focuses on comprehension, dealing with a limited number
14 of vocabulary words (3-4), only after the complete book has been read and discussed. In addition, Anchored Instruction, unlike the other oral reading techniques, is multifaceted, as this technique deals not only with the context that the word occurs and the meaning, but with spelling and decoding as well. Reading Aloud to Older Students While the techniques described focus primarily on students in the primary grades, there is reason to believe that even after acquiring the ability to read independently, children still benefit from teacher directed read alouds (Cunningham, 2005). Older children can learn the meanings of new words as efficiently from hearing stories read to them as they can from reading the stories themselves (Stahl, Richek, & Vandevier, 1991). In addition, reading to older children can also be a useful way of getting students interested in a book, so that they will continue reading it on their own (Stahl, 1999). Beck et al. (2002) successfully implemented Text Talk in fourth grade classrooms. The instruction was arranged in five day cycles. The vocabulary programs presented sets of words that were organized around themes. Themes were used because the researchers thought it would help students remember the words by giving them ready made connections. They felt that drawing on themes helped in creating coherent activities for the older students. In addition, Brett, Rothlein, and Hurley (1996) conducted a study with fourth grade students. The results of the study indicated that fourth graders can acquire new vocabulary from listening to stories if there is an explanation of new words as students encounter them in the stories. The study provided evidence of the value of reading aloud to children, even if they are independent readers.
15 In general, reading aloud techniques that actively engage students are appropriate for students in all elementary grades. Common features of effective techniques include careful selection of words and texts in conjunction with systematic instruction. The benefits of this type of vocabulary instruction include the acquisition of new words and an understanding of these words in different contexts. Semantic Relatedness Techniques A different form of interactive vocabulary instruction focuses on learning the meaning of specific words where it is important to make connections between and among words and concepts. The major studies in this area focus on techniques that encourage students to see the semantic relatedness of words and concepts being taught (Blachowicz & Fisher, 2000). These techniques include: Semantic Mapping (Heimlich & Pittleman, 1986), Semantic Feature Analysis (Pittleman, Heimlich, Berglund, & French, 1991), Concept Definition Mapping (Schwartz, 1988), Concept Muraling (Farris & Downey, 2004), Vocabulary Networking (Anderson & Roit, 1998), Concept or Content Sorts (Bear & Helman, 2004), and Venn Diagramming. In Semantic Mapping (Heimlich & Pittleman, 1986), the teacher writes a word representing a central concept. The students work in groups to list as many words related to the concept as they can. Next, the words are placed into general categories, and the categories are named. The teacher then discusses with the students the central concept, the other words, the categories, and their interrelationships. Semantic Feature Analysis (Pittleman et al., 1991) uses a grid which requires students to identify and record positive, negative, or possible attributes (Blachowicz & Fisher, 2002). This activity is particularly suited to refining word meanings (Graves,
16 2006). Students are presented with a grid that contains a set of related words on one axis. The other axis had a list of features that each word may or may not have. Students are required to insert plus or minus signs indicating whether or not a particular feature applies to each word. When students are proficient in working with partially completed grids supplied by the teacher they are encouraged to create their own grids. Bos and Anders (1992) provided support for using Semantic Mapping (Heimlich & Pittleman, 1986) and Semantic Feature Analysis (Pittleman et al., 1991). Working with upper elementary students in a series of six experiments, they implemented what they called The Interactive Teaching Project. Students participated in one of three interactive strategies including Semantic Mapping (Heimlich & Pittleman, 1986), Semantic Feature Analysis (Pittleman et al., 1991), and Semantic/Syntactic Analysis (Pittleman et al., 1991). Results indicated that the long term effects for the semantic strategies as compared to the definitional group were substantial. The students greatly increased their content knowledge from the pretest to posttest. This learning was maintained when the students were tested a month later. The researchers concluded that interactive strategies such as Semantic Mapping (Heimlich & Pittleman, 1986) and Semantic Feature Analysis (Pittleman et al., 1991) were superior to traditional instruction, such as the dictionary method. Additionally, they concluded that students learn most effectively when they are actively engaged with the content through discovering the meanings of ideas and comparing new learning to previous learning Some studies employed instructional strategies combining the use of semantic relatedness and written text (Nash & Snowling, 2006; Nelson & Stage, 2004). Nash and