THE EFFECTS OF TEXT ILLUSTRATIONS ON YOUNG CHILDREN?S VOCABULARY ACQUISITION AND CONSTRUCTION OF MEANING DURING STORYBOOK READ ALOUDS Except where references are made to the work of others, the work described in this dissertation is my own or was done in collaboration with my advisory committee. This dissertation does not include any proprietary or classified information. ____________________ Ann Wilkinson Willett Certificate of Approval: ______________________________ Edna G. Brabham Associate Professor Curriculum and Teaching ______________________________ Steven B. Silvern, Chair Professor Curriculum and Teaching ______________________________ Margaret E. Ross Associate Professor Educational Foundations, Leadership and Technology ______________________________ Joe F. Pittman Interim Dean Graduate School THE EFFECTS OF TEXT ILUSTRATIONS ON YOUNG CHILDREN?S VOCABULARY ACQUISITION AND CONSTRUCTION OF MEANING DURING STORYBOOK READ ALOUDS Ann Wilkinson Willett A Dissertation Submitted to the Graduate Faculty of Auburn University in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy Auburn, Alabama December 15, 2006 iii THE EFFECTS OF TEXT ILUSTRATIONS ON YOUNG CHILDREN?S VOCABULARY ACQUISITION AND CONSTRUCTION OF MEANING DURING STORYBOOK READ ALOUDS Ann Wilkinson Willett Permission is granted to Auburn University to make copies of this dissertation at its Discretion, upon request if individuals or institutions and at their expense. The author reserves all publication rights. __________________________ Signature of Author __________________________ Date of Graduation iv VITA Ann Marie (Wilkinson) Willett was born November 1, 1974, in Mobile, Alabama. Ann is the youngest of five children born to James and Sally (Long) Wilkinson. Ann?s siblings are James Wilkinson, Patrick Wilkinson, David Wilkinson, and Ashley (Wilkinson) Benford. Ann married Cecil Ray Willett on June 19, 1999. They have a daughter, Anna Grace, born on March 1, 2005. Ann attended the University of South Alabama and graduated with a Bachelor?s of Science degree in Early Childhood Education in December of 1997. She received a Master of Arts degree in Early Childhood Education from the University of South Alabama in August of 2000. Ann entered graduate school at Auburn University in May of 2001 to pursue a Doctorate of Philosophy degree in Early Childhood Education, with a specialization in Reading. Ann worked as a kindergarten teacher for three years with Mobile County Public Schools, a kindergarten teacher and a Reading Teacher for Phenix City Public Schools for two years, and has been employed with Auburn City Schools for the past four years as a second and third grade teacher. v DISSERTATION ABSTRACT THE EFFECTS OF TEXT ILUSTRATIONS ON YOUNG CHILDREN?S VOCABULARY ACQUISITION AND CONSTRUCTION OF MEANING DURING STORYBOOK READ ALOUDS Ann Wilkinson Willett Doctor of Philosophy, December 15, 2006 (M.A., University of South Alabama, 2000) (B.S., University of South Alabama, 1997) 72 Typed Pages Directed by Steven B. Silvern The effectiveness of illustrations for increasing students? vocabulary knowledge and construction of meaning during storybook read alouds had been studied extensively. Although decades of research have suggested that illustrations may improve story recall, there were also contradictory findings that indicated pictures inhibited young children?s construction of meaning. This study examined the effects of withholding the presentation of illustrations until the text was discussed so that young children could construct their own meaning from the text, rather than relying on pictures to gain meaning. The method of temporarily withholding pictures provided a scaffold and encouraged student talk before, during, and after the story to support students? ability to construct meaning from decontextualized language. This study investigated two hypotheses for how the delayed vi presentation of illustrations influenced students? learning during read alouds. One hypothesis explored the theory that the removal of pictures enabled the student to focus on the linguistic content of the text. The second theory examined how the removal of pictures until each portion of the text was discussed acted as a scaffold for younger students. Sixty-four first grade students participated in a read aloud of Jan Brett?s The Mitten. Students in the picture-withheld treatment group did not view the illustrations until the corresponding portion of the text was discussed; whereas, the text and treatment group had pictures presented as the story was read aloud. A statistical analysis of means indicated there was not a significant difference in vocabulary and comprehension performance when pictures were withheld. However, students identified as less- proficient readers from the pre-test performed better when the presentation of pictures was delayed until the text had been discussed, which suggested that the practice of withholding pictures may have served as a scaffold for building meaning while requiring the student to focus on the linguistic content of the text. vii ACKNOWLEDGEMENTS The author would like to dedicate this dissertation to her parents, Jim and Sally Wilkinson, who taught her she could achieve anything if she was willing to work for it. Dr. Silvern is gratefully acknowledged for his expertise, patience, and commitment to the author?s work. The author also wishes to express her gratitude to Dr. Brabham and Dr. Ross for their knowledge, insight, and time. A special thanks is extended to the first grade teachers at Dean Road Elementary who allowed the author to interrupt their school day in order to conduct research with their students. Finally, an expression of appreciation must be given to the author?s husband, Ray, for his support and encouragement. viii Style manual or journal used The Publication of the American Psychological Association, 5 th Edition Computer Software used Microsoft TM Word ix TABLE OF CONTENTS LIST OF TABLES??????????????????????????..?xi INTRODUCTION???????????????????????????. 1 Research On Reading Aloud????????????????????.. 1 Linguistic and Cognitive Benefits of Storybook Read Alouds???????... 3 Literature For Literacy Growth???????????????????.. 6 Divergent Practices in Storybook Read Alouds?????????????.. 8 The Development of the Text Talk Method.??????????????. 12 The Construction of Meaning From Print and Pictures.?????????? 16 Pictures As a Visual Context In Narrative Prose????????????.. 19 Purpose of the Study???????????????????????..29 METHOD??????????????????????????????..33 Subjects????????????????????????????..33 Materials????????????????????????????33 Pretests/Posttests.????????????????????????...36 Procedures???????????????????????????..37 RESULTS????????????????????????????????41 DISCUSSION??????????????????????????????...43 Limitations??????????????????????????.?47 x Implications??????????????????????????...48 Suggestions For Further Study??????????????????.?.50 REFERENCES?????????????????????????????..53 Appendix A: Posttest Rubric for The Mitten- Vocabulary??.??????????..59 Appendix B: Posttest Rubric for The Mitten- Comprehension?????.?????....60 Appendix C: IRB Application Cover Page...??????????????????61 xi LIST OF TABLES Table 1. Means and Standard Deviations 9in parentheses) for Vocabulary and Comprehension Scores Reported by Listening Comprehension Ability??...?41 1 INTRODUCTION Reading books with pictures has long been considered effective for promoting emergent readers? early literacy development (Adams, 1990; Trelease, 1989); however, removing accompanying pictures from the immediate reading context may be just as advantageous. Beck and McKeown (2001) discovered through naturalistic classroom observations of storybook read alouds with young children that withholding the presentation of pictures until after the text had been discussed enhanced the child?s ability to construct meaning from the decontextualized language of the text. The purpose of this study with first graders was to experimentally examine how delaying the presentation of text illustrations during a storybook read aloud would affect comprehension and vocabulary performance when compared with a similar read aloud style that presented pictures as the story was read. Research On Reading Aloud Before enrolling in formal schooling, many children have participated in a variety of literacy-building experiences. These early experiences with literacy, provided by adults, socialize children into literary practices (Pellegrini & Galda, 2003). Once young children enter formal schooling, they may already have amassed an assortment of literacy skills through interactions with literate adults. These early interactions between an adult and child, typically in the form of read-aloud experiences, have provided the child with exposure to the conditions and conventions of literacy (Allison & Watson, 1994). 2 Read alouds have been used for decades by early childhood educators to expose young children to the pleasures of reading. The classroom read aloud evolved into a more organized instructional practice to build young children?s linguistic and cognitive competence as research on early literacy development defined how literacy knowledge was acquired (Barrentine, 1996: Holdaway, 1979; Allison & Watson, 1994). The read aloud, which was once a practice of merely reading the author?s words, became a setting for social interaction and discussion to reflect on and extend the text of a storybook reading (Martinez & Teale, 2001). The potency of the read-aloud has been derived from the process of cultural transmission of knowledge (Pellegrini & Galda, 2003). Joint book reading between the adult, the expert, and the child, a novice, introduced children to reading conventions and provided the child with many opportunities to apply these novel skills. The expert aided the novice in discovering literacy conventions that were culturally specific and could not otherwise be learned. Culturally specific knowledge included modes of discourse for creating dialogue from text (Martinez & Teale, 1993). Martinez and Teale (1993) defined these modes of discourse as ?ways of taking? (p. 178) from texts. ?Ways of taking? from texts were modeled and taught by teachers through read alouds. The read aloud provided opportunities to model and scaffold discourse processing strategies, how to respond to text, and how to construct meaning from texts (Reese & Cox, 1999; Martinez & Teale, 1993). These models were internalized by the child through consistent interactions with the text from storybook read alouds (Martinez & Teale, 1993). Drawing on Vygotsky?s (1978) theories of language and cognitive 3 development, researchers such as Martinez and Teale (1993) point out that the literate adult had a significant influence on the read aloud because the interactions initiated by the adult were then internalized by the child. Learning originated in these interactions and discussions. The interactions and discussions of both the adult and child, as well as the nature of the text, constructed the context of the read aloud. By asking questions about the language and ideas presented in the text, children learn how to think about the process of building meaning from stories. Children began to utilize terms and strategies that demonstrated their developing metalinguistic and metacognitive awareness (Pellegrini & Galda, 2003). The social interactions and text discussions provided the child with a medium for constructing meaning from the language and ideas presented by the author in a storybook (Ferreiro & Teberosky, 1979; Martinez & Teale, 2001). Linguistic and Cognitive Benefits of Storybook Read Alouds Decades of research have demonstrated both the emotional and academic benefits of reading aloud to children (Adams, 1990; Morrow, 2001; Trelease, 1989). Joint book reading between parent and child established a literate environment, which fosters a love for books and contributes to future success in reading. By 1985, the Commission on Reading had recognized the importance of reading aloud and declared that it was the most beneficial activity parents and caregivers could do to establish a foundation of knowledge in preschoolers that would breed later success in learning to read (Anderson, Hiebert, Scott, & Wilkinson, 1985). As the benefits of parent and child?s shared reading interactions were identified in a growing body of read-aloud research (Rosenhouse, 4 Feitelson, Kita, & Goldstein, 1997), the activity of reading aloud became integrated into the early childhood curriculum as an instructional tool for promoting literacy development instead of as an activity that was simply enjoyed by both teacher and children. A generation of research has acknowledged that reading aloud has specific linguistic and cognitive benefits for young children. The effects of young children?s exposure to storybook read alouds promoted four facets of literacy development (Meyer & Wardrop, 1994). Two of these benefits were gained through indirect means: oral language development and acquisition of knowledge of school-like registers. The final two advantages of young children?s experience with read alouds were acquired directly. Participating in read alouds improved children?s concepts of print and increased children?s motivation to read (Meyer & Wardrop, 1994). A fundamental benefit gained from exposure to storybook read alouds was the experience with decontextualized language (Beck & McKeown, 2003). Decontextualized language referred to words presented alone without physical context. Before entering school, children have been easily able to discuss the context of the world around them, things they could point to and see. Young children learned early how to manipulate language in order to communicate within highly contextualized conversations. Ordinary conversations enabled a great deal of information to be omitted because of the context surrounding the conversation between two face-to-face speakers (Olson, 1977). Conversely, it was difficult for young children to create meaning from ideas presented alone in the words of the book without physical context. However, the ability to 5 manipulate decontextualized language was crucial to understanding and learning from the text (Beck & McKeown, 2003). Snow (1983) explained that the interactions between the adult and child in the read aloud provided a scaffold for the language by elaborating concepts and questioning important ideas. The amount of scaffolding provided decreased as the child?s facility with decontextualized language increased, until the child was able to create meaning from the text without adult assistance (Beck & McKeown, 2003; Meyer & Wardrop, 1994; Snow, 1983). An indirect affect of participating in storybook read alouds was an increase in oral language, which, in turn, improves one?s reading ability. Quality literature can introduce children to rich, descriptive language that is more complex than the language of ordinary conversations (Meyer & Wardrop, 1994); therefore, increasing the opportunity for children to encounter new words. Incidental learning of new word meanings occurs when a child is exposed to the words of the storybook. Elley (1989) found that repeated readings with 7- and 8-year olds increased incidental word learning from storybooks by 15 percent. Elley also found that gains were increased when the term was explicitly discussed during reading. Meyer and Wardrop (1994) reported that Nicholson endeavored to replicate the results of Elley?s original study using just one reading of the storybook. Nicholson?s data revealed that gains were greater for highly skilled readers (10%) than for poor readers (2%). Given the volume of read alouds students participate in throughout the school year, the potential for increasing young children?s oral vocabulary through incidental learning is significant. Research has explored the aspects and behaviors unique to the read-aloud that 6 make it an effective instructional practice (Martinez & Teale, 1993). These studies indicated the student?s participation in analytic talk is the most crucial component of the read aloud (Allison & Watson, 1994; Brabham & Lynch-Brown, 2002; Meyer & Wardrop, 1994; Reese & Cox, 1999; Martinez & Teale, 1993). Analytic talk helped young children make sense out of the decontextualized language of the text by talking about the ideas presented in the story. Dickinson and Smith (1994) discovered through an examination of teacher read-aloud patterns that an approach that includes analytic talk about major story ideas was the most beneficial approach for increasing vocabulary and comprehension performance. Specifically, discussing story concepts as they appeared in the text and giving students time to reflect, rather than delaying questions until the story is completed, yielded the greatest effect on student learning from storybooks. Literature For Literacy Growth Reading aloud to children has been declared an essential activity for building the knowledge and skills needed to excel in reading (Adams, 1990; Galda & Cullinan, 2000). However, Teale and Yokota (1985) voiced a concern that read alouds have been misused and overused. They maintained that simply reading to children is not enough to produce successful readers. Adults must consider what is being read and how the reading is being done in order for the read aloud to become an effective activity for literacy learning (Adams, 1990; Galda & Cullinan, 2000; Teale & Yokota, 2000; Trelease, 1989). The selection of an appropriate text was critical for an effective read-aloud (Galda & Cullinan, 2000; Teale & Yokota, 2000). With the abundance of quality literature available, it was pointless to choose a text of second-rate quality because the instructional 7 benefits of the read aloud would not be achieved with an inferior text. Many of the picture books chosen for read alouds were narratives. Quality narratives were defined by their literary elements (Galda & Cullinan, 2000; Teale & Yokota, 2000). The depiction of characters was an essential element of the picture book. Characters should have been multi-dimensional and actively participated in the story. The setting of the story should have been revealed swiftly through the text and detailed in the illustrations. Children should have been able to easily understand the plot of the story. The plot should have evolved in a chronological order with an identifiable climax and a satisfying resolution. The overall theme of the story should reflect an idea that is relevant to a child?s life. The language of the picture book was an essential ingredient of the read aloud. The language should have been imaginative and rich. The text should have provided a sophisticated vocabulary that will foster discussion and enable the child to increase their oral vocabulary (Beck & McKeown, 2001). Words should have been expressed in a way that was interesting and created excitement and anticipation (Galda & Cullinan, 2000). The melody of the words should have created a rhythm that allowed the language of the text to sound natural (Galda & Cullinan, 2000). Picture books selected as instructional tools for promoting vocabulary acquisition and comprehension development during read alouds should have been picture- independent texts. Picture-independent texts do not rely on illustrations to convey the author?s message. Instead, meaning is inherent in the linguistic content of the text. In order for literacy growth to be achieved, comprehension of the text should depend on 8 one?s ability to build meaning from the ideas and cues presented in the words of the text, not from relying on the pictures as an aid. Simons and Elster (1990) defined the goal of reading as the ability to read and understand a text independent of the pictures surrounding the words. Therefore, the books most advantageous for literacy growth enabled the student to focus on linguistic content rather than visual content. The linguistic content presented in the stories should communicate concepts that are cognitively challenging. Books chosen for read alouds should not be too easy or too difficult for the audience. Read alouds should expand upon children?s background experiences, extend the listeners? oral vocabulary, and be written slightly above the audience?s developmental level. The stories presented to the children should have an event structure that is conceptually demanding in order to provide the children with the opportunity to build meaning (Beck & McKeown, 2001, Trelease, 1989). Challenging texts depicting connected story content yield the most opportunities for increasing young children?s vocabulary acquisition and comprehension performance. Divergent Practices in Storybook Read Alouds Many literacy experts accept that the sharing of quality literature through read alouds served as the primary means of promoting young children?s early vocabulary and comprehension development (Adams, 1990; Anderson, Hiebert, Scott, & Wilkinson, 1985; Galda & Cullinan, 2000; Morrow, 2000). However, research has also indicated that reading aloud is not always an effective practice (Meyer & Wardrop, 1994; Protheroe, 1993; Teale & Yokota, 2000). Meyer and Wardrop (1994) reported a negative correlation between time spent reading aloud in kindergarten and subsequent reading achievement. 9 The negative correlation between time spent reading aloud and reading achievement may be a result of children?s participation in read alouds performed without instructional purpose (Hoffman, Roser, & Battle, 1993; Lickteig & Russell, 1993; Morrow & Brittain, 2003). Research (Hoffman et al., 1993; Lickteig & Russell, 1993; Morrow & Brittain, 2003) indicated that the majority of teachers read aloud to students at least once a day from trade books not related to a unit of study. Teachers once allowed very little discussion before or after the reading, whereas educators are now more likely to engage students in more interactive discussions (Hoffman, Roser, & Battle, 1993; Lickteig & Russell, 1993; Morrow & Brittain, 2003). However, reading aloud is still usually used for reasons of enjoyment and exposure rather than specific instructional purposes. The research (Hoffman et al., 1993; Lickteig & Russell, 1993; Morrow & Brittain, 2003) provided evidence that the read-aloud is not viewed as a teaching component in most teachers? instructional programs. Therefore, the educational values associated with reading aloud to children have not been realized in a large number of the classrooms represented in these studies. Vygotsky?s (1978) theory suggested that the most powerful element of the read- aloud is the adult reader because the model of reading provided by the adult is internalized by the child (Martinez & Teale, 2001). Therefore, the teachers? read-aloud style may have had a significant impact on students? literacy development. Differences in teachers? read-aloud styles had an important influence on how children construct meaning from the text (Martinez & Teale, 1993). Martinez and Teale (2001, p. 127) explained that ??the text from which students learn during storybook time is not merely 10 the book itself but also consists of what the participants say about the book.? If this expanded text encompassed both the storybook and the discourse, children had different experiences with the same books because the text presented by the teacher in one classroom will not be the same as the text read by another teacher. Because reading was a social process in which students internalize the text and the discourse, teachers? read- aloud style had direct impact on students learning (Martinez & Teale, 2001; Vygotsky, 1978). Many dimensions characterize each teacher?s read-loud style. Martinez and Teale (1993) conducted an observation of six teachers? storybook reading styles using the same text for each reading. Three significant aspects of the teachers? read-aloud styles emerged from the data: a) the focus of the teacher talk, b) type of information emphasized, and c) the teacher?s use of instructional strategies. Perhaps the most critical element of teachers? read-aloud style was the type of information emphasized by the teacher (Martinez & Teale, 2001). The types of information discussed by teachers during storybook reading fit into six categories (Martinez & Teale, 2001): textually explicit, pictorially explicit, background, inferential, and personal association. The nature of information emphasized by the teacher influenced how students responded and interacted with the text (Reese & Cox, 1999). Information presented by the teacher was especially significant because students must actively make connections between each text segment as the story was read in order to fill in the ?gaps? created by the author. Students needed to be able to access the appropriate background and conceptual knowledge in order to make logical inferences and construct meaning 11 from the text, and teachers supported these processes by providing models, assistance, and needed background knowledge. Martinez and Teale (1993) also examined the aspect of teachers? use of instructional strategies to determine if students? interactions were influenced by different roles in the read aloud. The instructional strategies used by the teacher to guide the discussion of the story had a direct effect on the classroom text created from the storybook. Seven different types of instructional strategies have been identified as tools for enhancing the read-aloud (Martinez & Teale, 2001): a) eliciting, b) inviting, c) informing, d) reviewing, e) recapitulating, f) eliciting reading, and g) reacting to text. Research (Allison & Watson, 1994; Brabham & Lynch-Brown, 2002; Meyer & Wardrop, 1994; Reese & Cox, 1999; Martinez & Teale, 1993) has noted the appearance of naturally occurring differences in the read-aloud styles of teachers. Syntheses of the research (Allison & Watson, 1994; Brabham & Lynch-Brown, 2002; Meyer & Wardrop, 1994; Reese & Cox, 1999; Martinez & Teale, 1993) revealed several prominent reading styles exhibited by teachers during storybook read alouds. The diversity in styles was reduced to two main elements of the read aloud: the demand level and the placement of comments during the story. The reading styles each were defined by when teachers and students should discuss the text, what should be discussed, and who should do the talking (Dickinson & Smith, 1994). An analysis of the research did not indicate that there is one best method of reading aloud to students (Brabham & Lynch-Brown, 2002; Dickinson & Smith, 1994; 12 Reese & Cox, 1999). However, research (Brabham & Lynch-Brown, 2002; Dickinson & Smith, 1994; Reese & Cox, 1999) did reveal that read-aloud methods incorporating analytic talk were the most advantageous for vocabulary growth and the construction of meaning from storybook read alouds. This implied that educators should consider instructional purposes when selecting a method for reading aloud to students. Educators must include analytic discourse throughout the read aloud if the instructional purpose is to facilitate students? construction of meaning from the text (Allison & Watson, 1994; Brabham & Lynch-Brown, 2002; Meyer & Wardrop, 1994; Reese & Cox, 1999; Martinez & Teale, 1993). Although teachers exhibited a range of styles in their delivery of storybook readings, the style of each teacher usually remained consistent no matter the content of the storybook read aloud (Dickinson & Smith, 1994; Reese & Cox, 1999; Martinez & Teale, 2001). The consistency of a teacher?s read-aloud style was significant because students will internalize the features of their teacher?s style, which will in turn influence how students interact with future texts (Martinez & Teale, 2001). Therefore, teachers with different reading styles provided their students with different literacy experiences during the students? process of literary development. The Development of the Text Talk Method Beck and McKeown (2001) developed the Text Talk Method based on the discrepancy between their knowledge of effective read-aloud practices and the read-aloud practices they had actually observed in elementary classrooms. Beck and McKeown (2001) observed teachers in both kindergarten and first-grade classrooms to develop their 13 own researchers? perspective of how specific read-aloud features promote early literacy growth. Four teachers were observed in two elementary schools. Every teacher was observed two- to four-times. Beck and McKeown?s observations were consistent with other findings in the literature showing that teacher read alouds typically did not incorporate the strategies known to be effective for early literacy growth. The strategies observed by Beck and McKeown were characteristic of low-demand read-aloud methods that did not include analytic talk about the text. Two types of interactions were prevalent throughout each observation. One type of interaction was the practice of clarifying content and vocabulary through closed questions. The other dominant interaction was including students in the developing story by asking questions about the content that was just read. These questions required answers of just two or three words to clarify the content. These types of interactions do not require students to connect or reflect on story content. Children were not prompted to make sense of the decontextualized language presented in the text. Read alouds were beneficial experiences for young children when the read aloud employed research-based ?best practices,? but too often the read aloud ended with the same outcome. Children recalled details from the illustrations and very little information from the text itself (Protheroe, 1993). Read alouds were most effective for promoting the construction of meaning when young children were given the opportunity to explore decontextualized language (Beck & McKeown, 2001). The value of the read aloud for literacy growth was derived from the talk that delved into the ideas presented in the story. ?Customarily children are only invited to briefly add words to a teacher?s response or to 14 echo language from the text? (Beck & McKeown, 2003, pg. 174). Research suggested that analytic talk that occurs throughout the reading was the most beneficial practice for increasing children?s understanding of the concepts presented in the text (Allison & Watson, 1994; Brabham & Lynch-Brown, 2002; Reese & Cox, 1999). Building meaning from the words presented in the text that represent concepts beyond the here and now was difficult for young children. However, building meaning from decontextualized language was necessary for early literacy growth. Meaning was constructed in the sharing of and reflection on ideas introduced in the language of the text. Research (Hoffman et al., 1993; Lickteig & Russell, 1993; Morrow & Brittain, 2003) indicated that most classroom read alouds fail to exploit the full potential of the storybook when children were not invited to interact with the text. Beck and McKeown?s goal for the Text Talk Method was to create a read-aloud experience that incorporated the most effective strategies for constructing meaning from decontextualized language. Their objective was to increase both comprehension and language development for primary-age students. The Text Talk Method began with the selection of books that provided students with quality experiences with decontextualized language. The interactions employed for the Text Talk Method were derived from Beck and McKeown?s (2001) earlier work, which was published in the book Questioning the Author. Questioning the Author describes a literature-based approach designed for use with adolescents. The sequence of the interactions for Text Talk Method was initiated with an introduction to the story to evoke interest and elicit relevant background knowledge. Throughout the reading, questions were interspersed and welcomed by the 15 reader to stimulate students to reflect on the ideas as they were presented in the text. The read aloud was concluded with a discussion of a particular aspect of the story?a character or an idea presented in the text. Throughout Beck and McKeown?s classroom observations, the researchers became aware of and alarmed by the students? tendency to ignore the linguistic content of the text and rely on the pictures for transmission of the story content. Although the linguistic content should have been the primary source for constructing meaning, it was the pictures that the children attended to most often. When developing the questions for the Text Talk stories, Beck and McKeown were mindful of how students would use the picture to formulate answers. To solve problems presented by students? reliance on pictures, Beck and McKeown chose to withhold illustrations in two specific situations: when pictures mirrored the words of the text and when the pictures conflicted with the ideas being presented in the words of the text. Early in the observations of the Text Talk Method, the researchers noted that students found the format of withholding pictures confusing. Students were unable to answer questions without first viewing the pictures. The negative aspect of with relying on the illustrations to construct meaning from the story was that to become a successful reader one must be able to derive meaning from the words. Thus, Beck and McKeown (2001) suggested that when the objective was to promote the processing of the linguistic content to create meaning teachers should be judicious about the use of pictures during storybook read alouds. Results of Beck and McKeown?s (2003) qualitative study indicated that withholding pictures until after the story concepts had been discussed through analytic 16 discourse increased young children?s vocabulary acquisition and comprehension performance. Students revealed a tendency to match their answer to the type of question asked. For example, a close-ended question would elicit a close-ended response. Conversely, an open-ended question prompted an open-ended response the majority of the time. When given the opportunity to express their own ideas, students were able to generate more complex responses to open-ended questions based on ideas they had constructed themselves from listening to and talking about the text. The Construction of Meaning From Print and Pictures Learning to read was an arduous task for young children whose vocabulary and world knowledge was limited due to inexperience. Emergent readers often developed a reliance on pictures to create meaning from the story ideas. Pictures made written text more comprehensible because children did not have to try to decipher the decontextualized language that made written text difficult for the children to understand (Brookshire, Scharff, & Moses, 2002). However, the construction of meaning often broke down when young children confused the relationship between print and picture, assuming the message was inherent in both mediums or in the pictures alone. Ferreiro and Teberosky?s (1979) Literacy Before Schooling described an extended research study undertaken by the authors to examine young children?s acquisition of literacy. The 108 children participating in the study ranged in age from four- to six-years- old. The researchers? experiments revealed that children have definite ideas, hypotheses, and theories about the forms and functions of texts. One question posed by the authors explored the conceptions children have about the relationship between print and pictures 17 and how children interpret these relationships. Children?s early representations of graphic images of writing and drawing were undifferentiated. As children were exposed to print and a variety of early reading experiences, their conceptions of the relationship between print and picture evolved. The findings gathered by the researchers from the sentence-reading task prompted them to propose that if reading is a process, then there must be different levels in the process and different forms of reading. Ferreiro and Teberosky (1979) defined four developmental levels of reading progression: a) picture and print are not differentiated, b) the print is differentiated from the picture, c) initial consideration of graphic properties of print emerges, and d) children search for a one-to-one correspondence between graphic and sound segments. Children who gave an advanced response on the word-reading task were more likely to give an advanced response on the sentence reading task. The developmental progression outlined by Ferreiro and Teberosky (1979) explained how children conceptualize the relationship between print and pictures. Young children initially were unable to differentiate the function of the print from that of the picture. As the child progresses, he theorizes that the print is a label or sentence explaining the picture, but the child does not consider the graphic properties of the text. The third level of progression interpreted the child?s conceptualization of written language as a representation of the picture, but consideration is given to the graphic elements of the text. Finally, the child progressed to a search for one-to-one 18 correspondence between the sound segments and the graphic segments of the text. Ferreiro and Teberosky (1979) found that at all levels the picture alludes to the meaning of the text. The findings from Ferreiro and Teberosky?s (1979) research had significant implications for storybook read alouds. Young children?s conceptualizations of the relationship between print and picture determined how they constructed meaning from the read aloud. Ferreiro and Teberosky?s theory explained how young children would use the illustrations to predict the meaning of the text, while ignoring the linguistic content of the page. The practice of relying on pictures for building meaning created a problem when the pictures did not reflect the message presented in the linguistic content. Simons and Elster (1990) further expanded the findings of Ferreiro and Teberosky?s (1979) study of children?s perceptions of the relationship between print and picture. Simons and Elster (1990) explained that children?s first experiences with print are embedded in the context of their physical environment, known as environmental print. When young children, who have been successful at reading environmental print, were confronted with reading words in isolation, they often experienced confusion. These children had difficulty separating the purpose of the pictures from the function of the print. Children often identified the picture as the source of the story or in both the print and picture, rather than in the text alone (Simons & Elster, 1990). Participation in read alouds and other early literacy experiences helped the emergent reader identify the text as the primary source of the story through repeated exposure to the texts and participation in analytic discussions about the text (Adams, 1990; Beck & McKeown, 2001). 19 Pictures As a Visual Context in Narrative Prose Pictures are a predominant feature in children?s literature, beginning reading instruction, and textbooks. Pictures were included in texts to perform various roles for a range of genres. The function of pictures was determined by the category of text the illustration accompanies. Carney and Levin (2002) suggested five functions of print in texts: decorational, representational, organizational, interpretational, and transformational. For the purpose of this study, the focus was directed to decorational and representational pictures that are the predominant style used along with narrative prose. Decorational pictures served as an embellishment to adorn the page, having little, if any, relationship to the text content. However, representational pictures reflected a segment of the text or the entire content of the story. The type of illustration most frequently used to accompany text was the representational picture. Pictures were used in texts for specific purposes. Illustrations can convey the story setting. Characters can be defined and developed through pictures. An illustration may extend the author?s plot. Illustrations may also provide the reader with a viewpoint different then the one presented in the text. Pictures may aid the text by contributing to its coherence or reinforcing what was stated in the text. Fang (1996) has identified six benefits that pictures provide when they accompany written text, they motivate the reader, promote creativity, provide a mental scaffold, foster aesthetic appreciation, and promote language and literacy. The imaginative illustrations are what often enticed a young reader to delve into the pages of a book. Research confirmed that children prefer books with bold, colorful 20 pictures to books without color or no pictures at all (Brookshire et al., 2002). Pictures function to attract the reader, evoking emotions and motivating the child to read on. Research proposes that pictures may also serve as an aid for young children learning to read by increasing their recollection of story content (Carney & Levin, 2002; Lesgold, DeGood, & Levin, 1977; Pressley, 1977). However, an opposing body of research (Beck & McKeown, 2001; Samuels, 1970) argued that pictures do not enhance learning, but in fact may hinder young children?s developing literacy skills. Pressley (1977) examined the assumption that imagery increases children?s learning of orally presented text. The extensive review of research studied how imagery affects learning at various developmental levels. Pressley (1977) cited a 1975 study by Shimron that employed first- and fourth-grade students in one of four experimental conditions: a) imagery condition- subjects were told to create a mental image of the story content; b) picture condition- imposed pictures presented with the story; c) unstructured pictures- subjects viewed pictures in which elements of the story were randomly placed; d) control condition- subjects listened to the story with no accompanying illustrations or instructions to induce images. The picture and unstructured picture conditions proved to be effective for first- and fourth-grade subjects as both performed better than the imagery and control conditions. The performance of fourth-grade students in the imagery condition exceeded the control group. In contrast, first-grade subjects in the imagery condition did not outperform their counterparts in the control condition. Shimron?s research demonstrated that imposed pictures facilitate young children?s learning from narrative text. However, it is only as children grow older that they were able to 21 effectively construct their own images in order to learn and recall story content. Young children relied primarily on the pictures placed within the text reading as an image of the orally presented content. However, older children were able to generate their own image of the concepts to gain meaning from the ideas presented orally. Guttmann, Levin, and Pressley (1977) tested the story recall of kindergarten, second-, and third-grade children when listening to a story read aloud and accompanied by complete pictures, partial pictures, or no pictures. The picture condition presented complete illustrations of relevant story content. Students in the partial pictures condition received explicit instructions to construct mental images. One-half of the students not shown pictures during the reading were told to generate mental images of the text. These students were in the imagery condition. Students in the control group were not exposed to pictures nor were they instructed to create mental images. Recall was measured through short answer questions. All subjects exposed to pictures during the read aloud recalled more content than those students not exposed to pictures. Kindergarten and second-grade students in the complete picture condition out performed students in the partial pictures, imagery, and control conditions. However, at the second-grade level, some students were able to effectively utilize partial pictures in order to generate images of the story. Third- graders in the complete pictures, partial pictures, and imagery conditions accurately recalled more content than those subjects in the no picture condition. Ruch and Levin (1979) conducted an experiment with first-grade children to try to explain why partial pictures had not been proven to effectively facilitate story recall in young children. The researchers proposed that the young subjects had in fact generated a 22 relevant image during the story but were later unable to retrieve the image when asked to recall story content. The findings from the experiment prompted Ruch and Levin (1979) to conclude that young children do not benefit from partial pictures unless the pictures are reinstated at the time of testing. Partial pictorial cues shown with their corresponding question at the time of testing produced greater recall than non-pictorial cues. The major finding of this study implied that partial pictures do facilitate young children?s recall of story content when given the proper eliciting condition. Research (Guttmann et al., 1977) suggested that young children?s processing of discourse surrounding the text could be facilitated by illustrations serving as a memory aid. In a study by Lesgold, DeGood, & Levin (1977), the authors propose that if pictures can facilitate discourse processing for young children, then illustrations should also be a valuable aid with both simple and complex stories, whether the text is long or short. To test this theory of illustration effect, the researchers presented first graders with passages of varying complexity and length. After listening to a recorded story, subjects in the experimental condition were given appropriate cutouts relevant to the story and instructed to place the cutouts on the background scene accordingly. Experimental and control subjects were matched with a partner. Control subjects colored pictures for the same amount of time it took their partner to place the story cutouts. Subjects were then given the title of the story and asked to recall all that he/she could remember. In a second replication of the experiment, subjects were also asked questions regarding the story. Children in the illustration condition exhibited better cued recall of story content for both long and short passages, whether the passage had one location or two. The young 23 children were easily able to produce a picture relevant to the story using the provided cutouts. The results of the free recall task revealed an interaction of condition with length. The young subjects demonstrated a better recall for shorter passages. This research finding was not completely unexpected as Lesgold, DeGood, and Levin (1977) noted that young children have difficulty effectively organizing images that can be easily retrieved later. The research (Lesgold et al., 1977; Pressley, 1997; Ruch & Levin, 1979) revealed a clear developmental progression in young children?s ability to apply mental imagery strategies to building meaning with narrative prose. Ruch and Levin (1979) proposed a developmental progression that began with children younger than the age of six who are unable to construct relevant mental images that facilitate recall. At this age, children relied on the imposed picture as the primary understanding of the story content. Between the ages of six and seven, children were able to construct relevant images, but experienced difficulty when trying to later retrieve those images to recall story content. After the age of eight, children developed the ability to store and retrieve images when instructed to do so. In this final stage of development, skilled readers no longer needed to utilize the imposed pictures of the text because they are adept at manipulating their own images of the written content. Young children?s comprehension of narrative text was facilitated by a strategy that presented text-relevant illustrations in conjunction with the narrative prose. As children grew older, they were better able to improve their 24 memory of story content by constructing their own mental images. Lesgold, DeGood, & Levin (1977, p. 360) asserted that ?.... older children can often do mentally what younger children need pictures for.? Pictures offered a preconceived image of written context for children who were unable to generate or manipulate their own relevant image. A growing body of literature (Levie & Lentz, 1982; Pressley, 1977) proclaimed illustrations to be particularly advantageous for low ability readers and readers from disadvantaged backgrounds. Levie and Lentz?s (1982) extensive review of research on illustrations and learning revealed a slight trend showing that pictures were more beneficial for poor readers than for good readers. On average, illustrations increased recall performance by 44% for poor readers while producing only a 23% increase for good readers. Only one study reviewed by Levie and Lentz (1982) demonstrated a significant interaction between reading ability and the presence/absence of pictures. A 1977 study by Wardle reported that illustrations helped below average readers, but had no effect on above average readers. Levie and Lentz suggested that Wardle?s results indicated that poor readers might be more dependent on pictures than good readers. A 1978 study by Waldner placed kindergarten, first-, and second-grade students with an experimenter who read aloud a story presented with pictures that were either text redundant or conflicted with the text. Poor readers in all three grade levels of Waldner?s study used more picture specific information than good readers. Poor readers also recalled more illustrated content than text information. Rusted and Coltheart (1979) observed poor readers moving their eyes from text to picture, checking picture details with information presented in the text. Good readers 25 were not observed attending to the illustrations. Pressley (1977) also asserted that the imagery strategy was more effective for some populations than for others. He noted that pictures significantly improved the comprehension performance of disadvantaged black children. The significant improvement reduced the performance difference between whites and blacks on story comprehension and inference tasks. Levie and Lentz?s (1982) review of the effects of text illustrations prompted them to concede that illustrations can facilitate learning. However, research did not clarify just how pictures aid the learning of narrative prose. A large body of research explained that illustrations function as a learning aid by improving text comprehension and/or improving retention of story elements (Lesgold et al., 1977; Levie & Lentz, 1982; Pressley, 1997; Ruch & Levin, 1979). Most traditional research designs have not been effective in differentiating between the effects illustrations have on retention versus comprehension (Levie & Lentz, 1982). Several interpretations have been offered to explain how pictures aid the recall of narrative prose. A quick assumption would be to assume that pictures aid recall through the repetition of redundant story information. However, Levie & Lentz (1982) asserted that the facilitation goes beyond simple repetition of story content. Levie and Lentz (1982) described a study that compared a group exposed to a narrative sentence twice with a group exposed to the sentence one time along with a corresponding illustration. The subjects exposed to information repeated in the pictures performed significantly better at recall tasks than subjects who listened to the same sentence read aloud twice. Levie and Lentz (1982) cited the research as further evidence that improved recall was 26 due to more than repetition or rehearsal of content. Ruch and Levin (1979) demonstrated that pictures provided better facilitation of recall of story content then rehearsal strategies. Guttmann, Levin, and Pressley (1977) showed that employing an imagery strategy could induce deeper understanding of verbal material. Additional theories have been proposed to explain how illustrations facilitate the comprehension of text. Levie and Lentz (1982) suggested that the greater facilitation may be a result of a deeper semantic analysis, a source for clarifying content, providing an imaginal memory trace, or the dual coding of verbal and visual material. Carney and Levin (2002) explained that the concentrated feature of illustrations could help the reader focus his/her attention. The illustrations directed the readers? attention to critical components of the text (Levie & Lentz, 1982). Ruch and Levin (1979) suggested that the illustrations did not direct the readers? attention to restricted story content, but instead to predominant themes of the story. Researchers (Carney & Levin, 1979) concluded that illustrations, when used to focus attention, facilitated greater story recall. Research (Carney & Levin, 2002; Guttmann, Levin, & Pressley, 1977; Lesgold et al., 1977; Pressley, 1977) did support the notion that children who were told a story presented with relevant illustrations recalled more than children who were not shown pictures. Carney and Levin (2002) suggested that text-related pictures provided readers with a powerful memory aid that improved story recall. The pictures served as a concrete representation of story content that allowed the reader to clarify written text. However, 27 research (Carney & Levin, 2002; Levie & Lentz, 1982) only supported this notion when the pictures were text redundant; that is, the content of the picture overlaped with the written content. An extensive review of research conducted by Filippatou and Pumfrey (1996) examined how pictures affected reading comprehension. This comprehensive research revealed that results vary according to student characteristics. Filippatou and Pumfrey (1996) explained that even when given task-appropriate pictures, students? ability to gain meaning from the picture depended upon the student?s ability to utilize certain information processing skills. Young children and low-ability readers had difficulty effectively generating pictures and/or effectively storing and organizing these images for later retrieval. Pictures have also been found to distract poor readers? attention away from the written content. A synthesis of the research from this study indicated that poor readers utilized pictures more often than good readers, even with material that was deemed more difficult. Filippatou and Pumfrey (1996) offered the explanation that pictures are typically selected because of the pictures? prominent cues, making it the easiest stimulus for generating a response. Filippatou and Pumfrey (1996) labeled the practice of accessing pictures first the ?practice of least effort.? Even with the prevalent use of pictures as a learning aid, little was known about the actual characteristics of illustrations that facilitate learning, specifically why and how pictures enhanced the learning of written content. Research has not explained what pictures provide that may or may not improve learning. It has not been determined whether or not pictures should be provided when the purpose of instruction is to construct 28 meaning from the linguistic content of storybooks. Beck and McKeown (2001) proposed withholding the presentation of pictures until the accompanying text has been discussed so that children have the opportunity to construct their own meaning. Research has not yet explored what effect withholding pictures during read alouds would have on young children?s vocabulary acquisition and comprehension performance. Two hypotheses have been proposed to explain why delaying the presentation of pictures would be beneficial for young children?s comprehension of the story content. One hypothesis has been that the absence of pictures allows the child to focus attention on the linguistic content; therefore, relying on the language of the text as the primary source for making meaning. Another hypothesis that has supported the postponement of illustrations is that the practice of withholding pictures serves as a scaffold for young children?s construction of meaning. Some researchers (Beck & McKeown, 2001; Brookshire et al., 2002; Simons & Elster, 1985) disputed the effectiveness of illustrations to direct students' attention. A picture does not always express what words do. Illustrations do not direct attention to the message of the author the way that syntax, stress, and pitch convey the author?s intent. These researchers asserted that illustrations actually served to direct attention away from the written text (Brookshire et al., 2002; Simons & Elster, 1985). For young children just learning how to read or for low-ability readers, pictures functioned as a distraction (Levie & Lentz, 1982). When a reader must refer to a picture for implied information in a picture-dependent text, the readers? attention became divided between text and picture in order to visually process both components (Brookshire et al., 2002). Deciphering the text 29 became increasingly more difficult with the illustrations serving as a distraction. Therefore, the removal of pictures during the presentation of the written text may have enabled the child to focus on the linguistic content, increasing their comprehension of major story ideas. Withholding pictures during storybook read-alouds until the text has been discussed may have also served as a scaffold for children learning to build meaning from storybooks. Scaffolding worked by gradually transferring the responsibility of learning from the adult to the child, or the expert to the novice. Teachers used the talk surrounding the text as a scaffold for understanding the decontextualized language of the text. The scaffold was guided by teachers? sensitivity to the student?s zone of proximal development. Teachers provided support by explaining or elaborating on difficult concepts, questioning significant concepts, and explaining the meaning of unknown words. The illustrations became contextualized because of the language that was generated through teacher-student discussions about the text. When the pictures were presented to the students after relevant discussion, the pictures served as a review of text content. Both the analytic talk and the follow-up presentation of pictures as a review may have caused the increase in young children?s ability to make sense out of the decontextualized language of storybooks. Purpose of the Study The challenge of understanding how the presentation of pictures during read- alouds may prevent children from constructing their own meaning of the story was addressed in the study presented here. Viewing illustrations prior to hearing the text read 30 may have inhibited the child?s ability to process the language of the text (Beck & McKeown, 2001; Protheroe, 1993). During a storybook reading that presented the illustrations and text simultaneously, children tended to focus on the visual aspects of the illustration, ignoring the linguistic content (Beck & McKeown, 2001). Children appeared to rely on the visual content alone to construct meaning. If the illustration conflicted with the text, it was likely the children misinterpreted the story. Withholding the pictures as the text was read allowed the listener to create their own mental image and construct meaning based on the linguistic content (Tovani, 2000). It is critical that young children have opportunities to create their own meaning from the text content if they are going to develop proficient reading and listening comprehension skills (Beck & McKeown, 2001; Protheroe, 1993). The role of illustrations in improving reading comprehension has been studied extensively. Decades of research on this subject have focused on the use of pictures at the word and sentence level, as well as in a variety of texts at various levels. However, the debate over whether pictures enhance or impede literacy development in young children has remained unanswered. Much of the research studying the effect of illustrations on comprehension utilized stories written expressly for clinical use. The stories adopted for the studies did not exemplify the characteristics of quality literature traditionally used in read alouds. The stories created for clinical use were brief in length and did not offer the conceptually challenging events present in high-quality tradebooks. Another limitation of this body of research was that it did not measure literary understanding in terms of critical analysis of story ideas or the ability to utilize context clues to increase word 31 knowledge. Instead, the research was limited to traditional story elements, such as: plot, setting, characters, and theme (Sipe, 2000). With the pervasive use of illustrations in beginning reading instruction, research was needed that explored how the visual content (pictures) of the story helped or hindered young children?s construction of meaning. This study investigated two potential hypotheses for how the pictures impact the construction of meaning during read alouds. The first hypothesis explored how delaying the presentation of pictures until the text has been discussed served as a scaffold for young students. The second hypothesis examined how removing pictures from the read aloud until after discussion enabled students to focus their attention on the text to build meaning from the linguistic content rather than the visual content. This study was designed to empirically determine if there were significant differences in learning outcomes for first graders when the appearance of pictures accompanied text content or was delayed until after the text had been discussed during read alouds. An adaptation of the Text Talk Method developed by Beck & McKeown (2001) was chosen for this study. These researchers argued that presenting stories to children without the presentation of pictures prior to discussion of the story content throughout the read aloud allowed the listeners to create their own meaning. This unconventional method of temporarily withholding pictures served as a scaffold and encouraged student talk before, during, and after the story to support students? ability to construct meaning from decontextualized language. The goal of this study was to extend Beck and McKeown?s (2001) observations and hypotheses about how illustrations hinder young 32 children?s vocabulary acquisition and construction of meaning from decontextualized language. The objective was to incorporate specific aspects of Beck and McKeown?s (2001) Text Talk Method into a controlled, experimental study to provide experimental evidence that delaying the presentation of pictures while reading enhances children?s construction of meaning from decontextualized language. Purposes of the study included: a) manipulating one read-aloud method while controlling for the texts selected and read in order to examine how illustrations affect children?s vocabulary acquisition and construction of meaning and b) employing an experimental design with the possibility to provide empirical evidence to support the hypotheses derived from Beck & McKeown?s (2001) qualitative research. This study was undertaken to provide educators with empirical evidence to support a better understanding of the effect of pictures on vocabulary acquisition and comprehension performance of young children during storybook read alouds. An additional purpose of this research was to present educators with additional information on research-based methods for increasing the educational value of the daily classroom read-aloud. 33 METHOD Subjects The subjects drawn for this study were from seven first-grade classrooms at a public elementary school in Southeastern Alabama. The students selected for the study were chosen based on parental permission to participate in the study. Two groups of students were then selected for each of the treatment groups. From the total of 64 subjects, 14 students were African American, 48 were Caucasian, and two were of other ethnic groups. The sample included 33 boys and 31 girls. The ages of children participating in the study ranged from 6.1 years to 8.11 years. The mean age of the group was 7.3 years. The elementary school where the study was conducted served grades one through five. The school?s average percentile score on the SAT10 was 70. Approximately 423 students attended the school. The school employed twenty-three classroom teachers. Thirty-one percent of the students enrolled received support through the Child Nutrition program. Materials The narrative texts chosen for this study all met the specific criteria outlined by Beck and McKeown (2001) in their description of the Text Talk Method. Texts chosen for the read aloud must be intellectually challenging for the child in order to stimulate the construction of meaning. Books that communicate meaning through the linguistic content 34 the text independently of the pictures were selected for the study rather than texts that relied on illustrations to project meaning. The final standard for selection was that the text must transmit an event structure that would allow children to build meaning through the connected content of the story. The first book chosen for the study was The Mitten (1989) by Jan Brett. Jan Brett has been honored throughout her career for her contributions to children?s literature. The Mitten was chosen for the subtle ideas presented in the linguistic content of the text. Beck and McKeown (2001) used The Mitten as a read aloud for their study of the Text Talk Method. The classic children?s book Harry the Dirty Dog (1984) was the second story selected for the read-aloud. Children have enjoyed the story, written by Gene Zion, for decades. Harry the Dirty Dog introduced children to a story plot that was easily recognizable. The plot introduced events that challenged children to explore and discuss their ideas. Harry the Dirty Dog was also selected by the researcher because the storybook was incorporated into Beck and McKeown?s (2001) original study. Scripts were developed for both the treatment groups to produce readings that could be precisely replicated. The scripts were developed based upon the format of the Text Talk Method outlined by Beck & McKeown (2001). Each reading was initiated with an introduction to the story to activate relevant background knowledge, excite interest, and to set purposes for reading. Introduction to Harry the Dirty Dog: Harry was a dog who liked everything, except getting a bath. He buries his family's scrubbing brush in the backyard and runs away from home before they can get him into the tub. Harry gets dirty playing around town. By the time Harry decides to go back home his family 35 doesn?t recognize him. How will Harry ever convince his family he is really their dog? Introduction to The Mitten: The Mitten is about a little boy named Nicki. Nicki is given a new pair of mittens made especially for him by his grandmother. Sadly, he loses one of his new gloves almost as soon as it is given to him. Soon, many of the woodland animals make their home in the glove. Will Nicki ever find his special glove? I am going to read you this story so that you will learn the mystery of Nicki?s lost glove. After the individual book introduction, students were given directions and purposes for listening. Students participating in the text and picture group were given the following instructions before listening to each read-aloud: I will read all the words on each page without showing you the pictures. As I read, I want you to picture the story in your mind. During the story, I will stop every few pages to ask you about what is happening in the story and to explain the meaning of words. After we talk about the story, I will show you the pictures on the pages I have read up to that point. You may ask questions whenever you like. After the story is over, we will talk about the characters and events in the story. Students participating in the picture-withheld group were given the following instructions before each story was read aloud to them: I will read all the words on each page while showing you the pictures. During the story, I will stop every few pages to ask you about what is happening in the story and to explain the meaning of words. You may ask questions whenever you like. After the story is over, we will talk about the characters and events in the story. As the story was read, scripted questions were interspersed throughout the reading at predetermined points to encourage the connection of story concepts and the expression of ideas. For the picture-withheld group, pictures were withheld until a portion of the text had been read and/or discussed. Once the reading of the story was completed, scripted 36 ?wrap-up? questions were utilized to challenge the children to reflect on a specific idea or event presented in the story. The script for each read-aloud session required that the actual reading and time spent reading be recorded. As the researcher progressed through the script, the introduction, read aloud, discussion, responses, and assessments were checked off. All readings were conducted by the researcher to ensure that each read aloud was precisely replicated. Pretests/Posttests Qualitative Reading Inventory-3. The study was launched by administering The Qualitative Reading Inventory-3 (Leslie & Caldwell, 2000) as a pre-test to all participants to establish the child?s level of listening comprehension. The Qualitative Reading Inventory-3 (QRI-3), an informal reading inventory, was administered individually. The QRI-3 provided the student?s approximate reading level or listening comprehension level. To identify listening comprehension level, a short passage was read aloud to the student. After the passage was completed, each student was asked six questions, explicit and inferential, corresponding to the story. The student?s listening comprehension level was determined when a child received a minimum score of instructional (four or five correct responses) on the comprehension test. Vocabulary and Comprehension Posttest. The researcher developed the posttest used to measure each student?s vocabulary understanding and comprehension of the read- aloud. The comprehension items on the posttest were constructed based on the QRI-3 model (see Appendices A & B). Several of the questions included on the posttest were 37 replications of original questions used in Beck and McKeown?s (2001) qualitative investigation of the Text Talk Method. The instrument was drafted to measure the student?s ability to respond to both explicit and inferential questions corresponding to the storybook read aloud. The instrument was given to an expert judge at Auburn University to establish the content validity of the evaluation instrument. The purpose of the expert judge was to determine if the instrument would yield scores that would enable the researcher to make valid inferences about each student?s ability to construct meaning from the story. Procedures School district personnel were contacted in order to determine if the elementary school would be accessible for the research investigation. Once permission was granted, the school?s principal was contacted in order to provide her with the research proposal and to designate classrooms for participation in the study. After a research site was approved, the Auburn University Human Subjects Review Committee reviewed the proposal for ethical treatment of human subjects and approved the procedure (see Appendix C). Once the review board at Auburn University granted approval, a letter of consent was sent home with each student identified as an eligible participant for the study. Of the 105 forms distributed, 70 forms were returned granting permission for the child to participate in the research. All students returning a completed letter of consent were selected for participation in the study. Students within each classroom were randomly assigned to either the text and picture treatment or picture withheld treatment. To achieve random assignment, the 38 researcher used two colored marbles. Green represented the text and picture group. Red represented the picture-withheld group. As each child was called to the pre-test assessment, a marble was drawn from the examiner?s pocket. The color of the marble drawn indicated the group assignment for that particular student. Four to seven students were randomly assigned to each sub-group within their assigned classroom. There were a total of fourteen sub-groups to minimize the number of participants in each read aloud and to lessen the wait time between individual posttests. An undergraduate student enrolled in elementary education courses was selected to participate in the study. The undergraduate student served the role of examiner by administering listening comprehension pretests to the subjects. The examiner was chosen for the study because of her background in education and familiarity with the administration of the QRI-3 model. The undergraduate student received two hours of training prior to the launch of the study. The training session, conducted by the researcher, consisted of administering the pretests. The administration of the assessments included: reading the scripts for the assessments, recording on the answer document, and scoring the answer document. The examiner practiced the execution of the examination script under the supervision of the researcher. Once the examiner delivered a satisfactory performance of the scripts, the pretest phase of the research commenced. The pretests were administered to determine each student?s level of listening comprehension. Leveled passages and assessment questions were read to the subjects in a scripted format. The examiner explained to the student that she would read a passage out 39 loud and ask questions related to the story. The child was asked to respond to both explicit and inferential questions. The responses were recorded verbatim onto the answer document by the examiner. An answer key was used by the researcher to score each item on the test. The researcher was able to determine the child?s approximate level of listening comprehension based on the correct number of responses given by the student. Scripts were used by the researcher to dictate procedures for both the text and picture treatment and picture-withheld treatment in order to produce precise read alouds. Two storybook read alouds were performed. Two read-aloud sessions were conducted because multiple read alouds permitted the students to adjust to the format of withholding pictures and to become increasingly more aware of the linguistic content. Harry the Dirty Dog was first read aloud to both text and picture treatment and picture-withheld treatment conditions in order to familiarize the subjects with the read-aloud format. No data was collected after this first reading. The Mitten was read aloud by the researcher during the second session to both the text and picture treatment and picture-withheld treatment groups. Each story was read word for word as written by the author. The researcher explained word meanings, interjected comments, and encouraged the connection of ideas as dictated by the script. Questions posed by the students were answered throughout each reading. Following the completion of each read aloud, the researcher encouraged the discussion of story concepts, events, or characters as instructed by the script. An individual vocabulary and comprehension posttest immediately followed the read-aloud session. The researcher read aloud each scripted question to the subject, 40 allowing appropriate time to respond. Student responses were recorded verbatim onto the answer document. A rubric was used to score each answer document. Students received one point for each question answered with a partially correct response, two points for a complete response, and zero points for incorrect responses for a possible total of twenty- six points. The researcher scored each rubric after the testing process was completed. Twenty percent of the posttests were also scored by an independent coder to establish reliability in scoring. Reliability between the two scores was eighty-six percent. 41 RESULTS An independent samples t-test by group for pretest scores of listening comprehension was employed to determine if there was a significant difference in the listening comprehension ability of the two treatment groups. A comparison of the means indicated that students in the picture-withheld group (M= .8125, SD= .821) were similar to the text and picture group (M= .8750, SD= .707). There was not a significant difference in reading ability between the two groups (t = .745, df = 62). Students were divided into two ability groups based on the mean score on the listening comprehension pre-test. Subjects grouped as proficient readers had a mean score at or above .835; whereas subjects determined to be less proficient readers earned a mean score below .835. A group by ability multivariate analysis of variance on listening comprehension and vocabulary was used to analyze posttest scores. Refer to Table 1 for a description of scores grouped by condition and ability. The analysis revealed a significant difference on the two variables by reading ability (F [2, 59 ] = 4.124, p < .025, ? 2 = .123) but not by treatment group (F [2, 59] = .123). The data indicated a significant treatment group by reading ability interaction, (F [2, 59] = 3.425, p < .04, ? 2 = .104). When illustrations were present, proficient readers exhibited better vocabulary acquisition (7.055) than poor readers did (3.071). When pictures were withheld, proficient readers earned a mean vocabulary score of 5.681; whereas less proficient readers exhibited increased vocabulary acquisition with a mean score of 5.600. 42 Comprehension measures revealed a similar pattern. Good readers earned a mean score of 6.111 when pictures were presented and 5.636 when pictures were withheld. On the other hand, less proficient readers had a mean score of 4.214 when exposed to illustrations and 5.40 when illustrations were not presented. These results indicated that the effect of withholding illustrations during a storybook read-aloud on vocabulary and comprehension performance depended on the reader?s initial listening comprehension ability. However, the overall small effect size (? 2 =.104) suggested that these results should be considered within the larger body of research and interpreted cautiously. Table 1. Means and Standard Deviations (in parentheses) for Vocabulary and Comprehension Scores Reported by Listening Comprehension Ability. Pictures and Text Pictures Withheld Total Vocabulary Low Ability 3.071 (2.302) n = 14 5.600 (3.373) n = 10 4.125 (3.011) n=24 High Ability 7.055 (3.472) n = 18 5.681 (2.589) n = 22 6.300 (3.056) n=40 Total 5.3125 (3.587) n= 32 5.656 (2.801) n=32 Comprehension Low Ability 4.214 (1.424) n = 14 5.400 (2.951) n = 10 4.708 (2.216) n=24 High Ability 6.111 (1.906) n = 18 5.636 (1.620) n = 22 5.850 (1.747) n=40 Total 5.281 (1.938) n=32 5.562 (2.078) n=32 43 DISCUSSION Results of the statistical analysis of means indicated that there was not a statistically significant difference in vocabulary or comprehension performance when pictures were or were not withheld during a storybook read aloud. However, the results also indicated that the presence of illustrations throughout the read aloud did not present an advantage as findings from previous research has shown (Brookshire et al., 2002; Lesgold et al., 1977; Levie & Lentz, 1982; Pressley, 1997; Ruch & Levin, 1979). In this section, findings of the present study will be examined in light of results from existing research on the effect of pictures on vocabulary learning and listening comprehension for read alouds. In addition, implications of the present research and suggestions for further study will be proposed. Although the examination of the method of withholding pictures during a storybook read aloud in a controlled, experimental study did not yield significant results for either treatment group, the findings from the present study do offer some curious points to ponder. Results from prior research studies can be used to argue that the presence of illustrations greatly enhanced young children?s comprehension performance, while text-alone conditions resulted in a significantly lower comprehension performance for younger subjects (Lesgold et al., 1977; Levie & Lentz, 1982; Pressley, 1997; Ruch & Levin, 1979). However, in the present study, subjects in the picture condition did not out 44 perform subjects in the no-picture condition. The means for the two treatment groups were not statistically different. Several explanations for these results are possible. The oral presentation of the author?s message and the analytic talk surrounding the text, serving as a scaffold, may have facilitated the construction of meaning from decontextualized language for the readers identified as demonstrating less proficiency in the pre-test. Less proficient students performed better when the presentation of pictures was delayed until after discussion of the text. This result, though unexpected, is not surprising. Less proficient readers are often plagued with distractions while trying to build meaning from texts. It is conceivable that the withholding of illustrations enabled the less proficient readers to focus on the linguistic content. However, the small effect size for this finding requires that the results may have limited generalizibility considering the larger body of existing research (Lesgold et al., 1977; Levie & Lentz, 1982; Pressley, 1997; Ruch & Levin, 1979). An analysis of the proposed hypothesis that temporarily withholding pictures during read alouds would increase young students? ability to make sense out of the decontextualized language of storybooks appears to be true for less proficient readers, but not for readers identified as highly-skilled. It may be that the scaffold of delaying the presentation of pictures was beneficial for less proficient readers because it enabled them to focus all of their faculties on the linguistic content. However, it may be assumed that the negative effect of withholding pictures on proficient readers? vocabulary and comprehension performance was due to the proficient readers familiarity with the format of the expected text with the simultaneous presentation of pictures and information in 45 print. The format of withholding pictures until after discussion may have interfered with proficient reader?s attention to the linguistic content, as the unfamiliar context was distracting. Nevertheless, the results of the present study suggest future research be conducted with similar populations to try to determine if the finding that withholding pictures during read alouds enhances less proficient readers? ability to construct meaning from decontextualized language may be replicated. Furthermore, the results of this study may be explained also by the fact that it is customary in many classrooms for the teacher to solicit hasty, abbreviated responses to texts during storybook read alouds. Often, students are only encouraged to simply echo the language presented in the text. Rarely are children called upon to respond to and interact with the text through the sharing of ideas throughout the reading (Beck & McKeown, 2001). Martinez and Teale (1993) explained that the text surrounding the story from which the students learn is more than just the words written in the book. The actual text is a combination of the words conveying the author?s message and the ideas the students share about the book. Reading is a social process in which a young child actively engages in the construction of meaning through interaction with the text. The present study suggested that the presence of analytic talk and interaction in both treatments might have resulted in similar means when comprehension performance was measured. An additional explanation of the results in this study that contradict past findings may be related to the specific characteristics of the illustrations presented in the storybook. The high-quality illustrations in The Mitten depicted rich, colorful scenes that 46 clearly reflected story content. The pictures were neither conflicting nor merely decorative. The text-redundant pictures did not convey any information not presented in the written text. Perhaps the subjects were able to easily coordinate the ideas presented in the story; therefore, the students did not rely on the pictures as an external memory aid (Lesgold et al., 1977). One cannot underestimate the power of words. A possible reason for the similar comprehension performance between treatment groups may be the impact of the author?s words as delivered by a fluent reader during the storybook reading. A well-written story, such as The Mitten, has the ability to capture the readers? attention with its rich vocabulary and potent language. The pictures cannot express feelings, character motivations, and judgments as effectively as the author?s words; nor can pictures deliver the range of emotions as skillfully as an eloquent reader. Pictures are applauded for the quality of their artwork. Illustrations functioning as a motivation for the child, although effective, is only a secondary concern when pictures are created (Lesgold et al., 1977). As noted by Lesgold, DeGood, and Levin ?....there has been little effort to direct the child?s attention to aspects of the picture the way that syntax, stress, and pitch direct attention to aspects of the auditory message(1977, p. 360)?. The present study was unable to corroborate the findings produced by Beck and McKeown?s (2001) observations of the Text Talk Method. Beck and McKeown?s (2001) observations were extended over the period of one school year. The researchers employed four teachers, recording approximately twenty-five read-aloud lessons per teacher. Four lessons were used to examine the effectiveness of the Text Talk Method. 47 Utilizing a number of classroom teachers over an extensive period of time while collecting large amounts of data was beyond the scope of this study. It would be expected that an exact replication of Beck and McKeown?s (2001) study of the Text Talk Method using a quantitative experimental design would produce significant results for enhancing young children?s construction of meaning from narrative texts. Limitations A methodological limitation to this study was in the one time, one test design. The format of withholding pictures during a storybook read aloud was unusual, if not bewildering, to the students. Students were eager to see the pictures as the story was being told. Exposing the students to the format of withholding pictures during three or four read-alouds would have familiarized the students with the format and enabled them to focus more on the linguistic content of the story. It would also have been advantageous for the researcher to collect comprehension data following each read-aloud. Brabham and Lynch-Brown (2002) explained that multiple readings maximize the comprehension gains so that the difference in effects produced from the read-aloud method is easily perceivable. Certainly an additional limitation to the present study would be the absence of an older group of participants. The presence of such a group would enable the researcher to compare the comprehension performance of two developmentally different age groups when withholding illustrations during storybook read-alouds. A group of nine year-olds whose cognitive abilities are more developed as compared to of first graders may have been more adept at processing the linguistic content of the story to construct meaning. It 48 would have been insightful to see how an absence of illustrations combined with the presence of analytic talk during a read aloud impacts learning at various ages. An additional limitation of the present study to note may have been the selection of a well-known, much-loved story to be used as the narrative text in the experiment. During the book introductions for each group, nearly every student expressed familiarity with the selection. Prior to launching the study, the researcher requested that each first grade teacher not use The Mitten in his or her winter curriculum. Regardless of the first grade teachers? compliance, due to the popularity of the book, it is highly likely that students were exposed to the story prior to first grade. However, it should be noted that many of Jan Brett?s books share a similar illustration style and the picture adorning the front cover of the book may simply have reminded the students of another Jan Brett story. Therefore, it is possible that due to confusion over similar illustrations, the student expressing familiarity with the book had indeed actually not heard the story. Additionally, due to random assignment, any bias from earlier exposure to the text would have likely been offset by equal disbursement among groups (Williamson & Silvern, 1983). Implications The conclusions derived from this study imply that withholding pictures during a storybook read aloud can be particularly advantageous for less proficient readers. Research (Levie & Lentz, 1982) has demonstrated that less proficient readers are highly dependent on pictures for constructing or clarifying meaning. Withholding pictures while presenting the oral reading of the text frees the less proficient reader from dividing 49 his/her attention between the illustrations and the linguistic content. Without the distraction of illustrations, the less proficient reader is able to use words rather than a picture to derive meaning from narrative prose. Thus, in the classroom setting, when the read-aloud is embedded in reading instruction for low-ability readers, teachers should be cautious with the pervasive use of illustrations as a medium for constructing meaning. Research has not acknowledged one best method for reading aloud to children; however, it does acknowledge the best methods for increasing vocabulary growth and comprehension performance. A review of research (Allison & Watson, 1994; Brabham & Lynch-Brown, 2002; Reese & Cox, 1999) indicated educators should adopt read-aloud styles that incorporate and encourage analytic talk before, during, and after the reading in order to facilitate children?s construction of meaning. A style that increases interactions with decontextualized language improves young children?s comprehension performance (Beck & McKeown, 2003) The purpose of this research was not to denounce the value of illustrations. As noted earlier in this text, it is often the rich, colorful pictures that lure a reader to the text. Pictures have aesthetic and affective impacts on the reader, enhancing the experience of reading for pleasure (Levie & Lentz, 1982). Simply stated, pictures are enjoyable. However, when the function of the read aloud is to instruct, not to entertain, one cannot ignore the educational significance of this opportunity. Read alouds are one of the most beneficial classroom practices for literacy development (Adams, 1990; Lickteig & Russell, 1993; Morrow, 1991; Trelease, 1989). Read alouds present young children with the opportunity to build meaning from abstract ideas presented in storybooks (Beck & 50 McKeown, 2003). If our definition of being literate is to be able to build meaning from written words, then we must teach our children to construct meaning from text and not from pictures. Research (Allison & Watson, 1994; Brabham & Lynch-Brown, 2002; Reese & Cox, 1999) confirmed that the optimal environment for enhancing literacy growth is the analytic talk and shared interactions between teachers and students surrounding the text of the storybook read aloud. Beck and McKeown (2001) were inspired to develop the Text Talk Method after observing just how rarely educators used effective strategies when reading aloud. Teachers fail to exploit the proven benefits of the read aloud when they do not invite students to reflect and respond to the text, as well as to share ideas. All too often students are limited to brief replies or to echo the words of the text. ?We need to activate children?s potential to master decontextualized language, in order to meet the increasingly complex demands that they will encounter throughout their school years.? (Beck & McKeown, 2003, p. 174) Suggestions For Further Study Every year educators are inundated with new theories for developing literacy in young children. Educators must rely on knowledge and expertise to delve through the abundant methodologies to determine which methods truly exemplify the best practices for teaching reading. Identifying the best practices for developing early literacy is not always an easy task with the many contradictions that research presents. For example, the extent to which illustrations impact reading is arguably undecided (Beck & McKeown, 2001; Lesgold et al., 1977; Levie & Lentz, 1982; Pressley, 1997; Protheroe, 1993; Ruch & Levin, 1979). 51 Further research is needed to extend Beck and McKeown?s qualitative findings that withholding pictures during storybook read alouds enhances children?s ability to construct meaning. Researchers utilizing an empirical experimental design should incorporate multiple readings over an extended time period to maximize comprehension gains so that the benefits of the Text Talk Method can be easily identified. It would also be valuable to compare the method at various developmental levels, such as kindergarten, first, and third grades. The potential for yielding significant results is high. The Text Talk Method provides educators with an effective strategy for increasing primary age students? language development and comprehension. Pictures undeniably capture children?s curiosity and entice them to open the pages of a book. The potential power of illustrations cannot be argued. However, the potential for illustrations to facilitate the construction of meaning from narrative prose is still undetermined. The use of illustrations is pervasive, both in narrative prose, informational texts, and multimedia (Carney & Levin, 2002). Research demonstrates that pictures can significantly aid students? learning from informational texts (Carney & Levin, 2002; Levie & Lentz, 1982). It remains uncertain how illustrations can be used to rouse the cognitive processes that facilitate young children?s construction of meaning from narrative prose (Levie & Lentz, 1982). Further research should explore how illustrations utilized at various developmental levels can enhance young children?s construction of meaning from decontextualized language so that students are able to produce thoughtful, complex responses to the text. 52 Enhancing young children?s literacy development requires educators to diagnose each student's individual learning style and instructional needs. In order for teachers to prescribe appropriate methods of instruction, teachers must first be knowledgeable of which strategies facilitate the unique needs of his/her various students. Further research is warranted to investigate how various combinations of read-aloud style, presence/absence of illustrations, and learner characteristics best facilitate young children?s comprehension performance. The ultimate objective for future research should be to provide educators with proven research-based methods that result in the significantly deeper processing of text, enabling young children to build meaning from the decontextualized language of narrative storybooks. 53 REFERENCES Adams. M. (1990). Beginning to read: Thinking and learning about print. Cambridge, Massachusetts: The MIT Press. Allison, D., & Watson, J. (1994). The significance of adult storybook reading styles on the development of young children?s emergent reading. Reading Research and Instruction, 34, 57-72. Anderson, R., Hiebert, E., Scott, J., & Wilkinson, I.A.G. (1985). Becoming a nation of readers: The report of the commission on reading. Washington, DC: The National Institute of Education. Barrentine, S. (1996). Engaging with reading through interactive read alouds. The Reading Teacher, 50, 36-43. Beck, I.L., & McKeown, M.G. (2001). Text talk: Capturing the benefits of read aloud experiences for young children. The Reading Teacher, 55, 10-20. Beck, I.L., & McKeown, M.G. (2003). Taking advantage of read-alouds to help children make sense of decontextualized language. In A. Van Kleeck, S. A. Stahl, & E. B. Bauer (Eds.), On reading books to children: Parents and teachers (pp. 159-176). Mahwah, NJ: Lawrence Erlbaum Associates. Brabham, E.G., & Lynch-Brown, C. (2002). Effects of teachers? read aloud styles on vocabulary acquisition and comprehension of students in the early elementary grades. Journal of Educational Psychology, 94, 465-473. 54 Brett, J. (1989). The mitten. New York: G.P. Putnam?s Sons. Brookshire, J., Scharff, L.F.V., & Moses, L.E. (2002). The influence of illustrations on children?s book preferences and comprehension. Reading Psychology, 23, 323-339. Carney, R.N., & Levin, J.R. (2002). Pictorial Illustrations still improve students? learning from text. Educational Psychology Review, 14(1), 5-25. Dickinson, D., & Smith, M. (1994). Long-term effects of preschool teachers? book reading on low-income children?s vocabulary and story comprehension. Reading Research Quarterly, 29, 104-122. Elley, W.B. (1989). Vocabulary acquisition from listening to stories. Reading Research Quarterly, 24, 174-187. Fang, Z. (1996). Illustrations, text, and the child reader. What are pictures in children?s storybooks for? Reading Horizons, 37, 130-142. Ferreiro, E., &Teberosky, A. (1979). Literacy before schooling. New Hampshire: Heinemann Education Books. Filippatou, P., & Pumfrey, P.D. (1996). Pictures, titles, reading accuracy and reading comprehension: a research review (1973-95). Educational Research, 38(3), 259-290. Galda, L., & Cullinan, B.E. (2000). Reading aloud from culturally diverse literature. In D.S. Strickland & L.M. Morrow (Eds.) Beginning reading and writing (134-142). New York: Teachers? College Press. 55 Guttmann, J., Levin, J.R., & Pressley, M. (1977). Pictures, partial pictures, and young children?s oral prose learning. Journal of Educational Psychology, 69, 473-480. Hoffman, J.V., Roser, N.L., & Battle, J. (1993). Reading aloud in classrooms: From the modal to a ?model?. The Reading Teacher, 46, 496-503. Holdaway, D. (1979). The foundations of literacy. Sydney: Ashton Scholastic. Lesgold, A.M., DeGood, H., & Levin, J.R. (1977). Pictures and young children?s prose learning: A supplementary report. Journal of Reading Behavior, 9(4), 353-360. Leslie, L., & Caldwell, J. (2000). Qualitative reading inventory-3(3 rd ed.). USA: Allyn & Bacon. Levie, W.H., & Lentz, R. (1982). Effects of text illustrations: A review of research. Educational Communication and Technology: A Journal of Theory, Research, And Development, 30(4), 195-232. Lickteig, M., & Russell, J. (1993). Elementary teachers? read aloud practices. Reading Improvement, 30, 202-208. Martinez, M., & Teale, W.H. (1993). Teacher storybook reading style: A comparison of six teachers. Research In the Teaching of English, 27, 175-199. Martinez, M., & Teale, W.H. (2001). Classroom storybook reading: The creation of texts and learning opportunities. Theory Into Practice, 28, 126-135. Meyer, L., & Wardrop, J., (1994). Effects of reading storybooks aloud to children. Journal of Educational Research, 88, 69-85. Morrow, L. M. (2001). Literacy development in the early years: Helping children read and write. Boston, MA: Allyn & Bacon. 56 Morrow, L.M., & Brittain, R. (2003). The nature of storybook reading in the elementary school; current practices. In A.Van Kleeck, S. Stahl, & E. Bauer (Eds.), On reading books to children: Parents and teachers(140-158). Mahwah, NJ: Lawrence Erlbaum Associates. Olson, D.R. (1977). From utterance to text: The bias of language in speech and writing. Harvard Educational Review, 47, 257-282. Pellegrini, A. D., & Galda, L. (2003). Joint reading as a context: Explicating the ways context is created by participants. In A.Van Kleeck, S. Stahl, & E. Bauer (Eds.), On reading books to children: Parents and teachers (321-335). Mahwah, NJ: Lawrence Erlbaum Associates. Pressley, M. (1977). Imagery and children?s learning: putting the picture in the developmental perspective. Review of Educational Research, 47(4), 585-622. Protheroe, P. (1993). Are picture books harmful? New Scientist, 138, 44-46. Reese, E., & Cox, A. (1999). Quality of adult book reading affects children?s emergent literacy. Developmental Psychology, 35, 20-28. Rosenhouse, J., Feitelson, D., Kita, B., & Goldstein, Z. (1997). Interactive reading aloud to Israeli first graders: It?s contribution to literacy development. Reading Research Quarterly, 32(2), 168-183. Ruch, M.D., & Levin, J.R. (1979). Partial pictures as imagery-retrieval cues in young children?s prose recall. Journal of Experimental Child Psychology, 28, 268-279. Samuels, S.J. (1970). Effects of pictures on learning to read, comprehension, and attitudes. Review of Educational Research, 40, 397-407. 57 Simons, H.D., & Elster, C. (1985). The functions of illustrations in beginning reading texts. In J. Niles & A. Lalik (Eds.), Issues in literacy: A research perspective. Thirty-fourth Yearbook of the National Reading Conference, 320-323. Simons, H.D., & Elster, C. (1990). Picture dependence in first-grade basal texts. Journal of Educational Research, 84(2), 86-92. Sipe, L. (2000). The construction of literacy understanding by first and second graders in oral response to picture storybook read-alouds. Reading Research Quarterly, 35(2), 252-275. Snow, C.E. (1983). Literacy and language relationships during the preschool years. Harvard Educational Review, 53, 165-189. Strickland, D.S., & Morrow, L.M. (2001) Beginning reading and writing. New York: Teachers? College Press. Teale, W.H., & Yokota, J. (2000). Beginning reading and writing: perspectives on Instruction (3-21). In Strickland, D.S. & Morrow, L.M. (2001) Beginning reading and writing. New York: Teachers? College Press. Tovani, C. (2000). I read it but I don?t get it: Comprehension strategies for adolescent readers. USA: Stenhouse Publishers. Trelease, J. (1989). The read aloud handbook. New York: Penguin Books. Vygotsky, L.S. (1978). Mind in society: The development of psychological processes. Cambridge, MA: Harvard University Press. Williamson, P.A., & Silvern, S. (1983). The effects of text elaboration on children?s aural language comprehension. Educational Psychology, 3(1), 5-14. 58 Zion, G. (1984). Harry the dirty dog. New York; Harper Collins. 59 APPENDIX A Posttest Rubric for The Mitten Student Code# _______ Vocabulary Questions Student Response Complete Response 2 Partial Response 1 Incorrect Response 0 Question #1: In the story, the snowshoe rabbit stops to admire his winter coat. What does it mean to ?admire?? Likes it, is proud of it Question #2: The owl was attracted to the mitten by the commotion. What is commotion? Loud noise and lots of movement Question #3: The sight of the mitten made the fox drowsy. What does it mean to be drowsy? To feel sleepy Question #4: The great bear lumbered by the mitten. What does the author mean by ?great?? Large or big Question #5: With so many animals inside the mitten, it was pulled and it bulged. What does it mean to ?bulge?? To stretch Question #6: When the mole discovered the mitten, he burrowed inside. What does the author mean by ?burrowed?? Dug a hole Question #7: The mole and the rabbit were bumped and jostled when the hedgehog moved into the mitten. What does it mean to be ?jostled?? Pushed and shoved 60 APPENDIX B Posttest Rubric for The Mitten Student Code# _______ Comprehension Questions Student Response Complete Response 2 Partial Response 1 Incorrect Response 0 Question #1: (implicit) What did Baba mean when she told Nicki, ?If you drop one in the snow you?ll never find it.? The mitten will blend in with the snow because it is as white as snow. Question #2: (explicit) Why did the mole choose to burrow inside of the glove? He was cold and tired. Question #3: (implicit) When the fox comes upon the mitten, just the sight of it makes the fox feel drowsy. Why would the mitten make him feel sleepy? It looked warm and cozy, like a good place to rest, which made him tired. Question #4: (implicit) It says that ?the mitten swelled and bulged, but Baba?s good knitting held fast.? What does ?held fast? mean? It means remained tight and strong. Question #5: (explicit) What finally made all of the animals leave the mitten? When the mouse climbed onto his nose, Bear sneezed shooting the mitten into the air. Question #6: (explicit) How was Nicki able to find his mitten after it was lost in the snow? He saw it floating in the air, silhouetted against the blue sky. 61 APPENDIX C Application For IRB Approval