Examining Student Engagement in Preschool Read Alouds

Examining Student Engagement in Preschool Read Alouds Kathleen A. Paciga, Jennifer Garrette Lisy & William H. Teale University of Illinois at Chicago To appear in K. Leander, D. W. Rowe, D.K. Dickinson, M. Hundley, R. T. Jimenez, & V. J. Risko, (Eds.), 59th Yearbook of the National Reading Conference. Oak Creek, WI: National Reading Conference. It is widely accepted that reading aloud can function as an important instructional activity for literacy in early childhood classrooms (e.g., International Reading Association & National Association for the Education of Young Children, 1998; National Early Literacy Panel, 2008; Van Kleek, Stahl, & Bauer, 2003). Descriptive research has shown classroom read alouds to be socially constructed events involving the teacher and children with resulting student learning attributable to (a) the way the teacher reads the words in the book, (b) the book itself, and (c) the discussion created among the children and the teacher (Sipe, 2008; Teale, 2003). Thus, one factor important to the impact that read alouds have on children’s early literacy learning is the teacher’s read aloud style—how she involves students and what she emphasizes in the readings of books (Collins, 2004; Dickinson & Smith, 1994; Teale, 2003; Teale, Paciga, & Hoffman, 2007). Also significant is the content and language of the books being read (Teale, Yokota, & Martinez, 2008). Finally, there is what we have come to call the orchestration of the read aloud activity. Orchestration is related to read aloud style in some ways, but it is also distinct from style because it refers to factors like the amount of time the teacher devotes to the read aloud and the ways she organizes the children in the physical space used for the activity. Interestingly, little systematic investigation has been conducted on this latter factor in read alouds across the primary grade years and virtually none has been conducted for preschool children. The present study explored how specific features of preschool teachers’ orchestration of read alouds—the time spent in the activity, the seating arrangement of the children and the books chosen for it —related to the engagement of the students in read aloud sessions. Theoretical Framework This work is theoretically rooted in a sociocultural perspective, as well as the related sociocognitive perspective. Such frameworks see the conventions of literacy as culturally specific, and learning them requires engagement of the learner coupled with a degree of assistance from more expert individuals (typically parents and teachers). Rogoff (1990), building upon Vygotsky’s (1978) sociocultural theory, proposed that cognitive development occurs through an apprenticeship process in which the expert gradually transfers responsibility to the novice as the novice becomes more competent. The importance of learner engagement in reading development is substantial, as shown by research at a wide range of age levels (Brozo, Shiel, & Topping, 2007; Guthrie & Wigfield, 2000). Applying this model to the early childhood classroom, we see that activities such as read alouds enable children to actively participate in the learning process and the content of the particular text being read because the teacher scaffolds the experience through reading the text, thinking aloud for the children, fostering discussion about the book, and otherwise orchestrating read alouds to maintain children’s engagement as thoroughly as possible. Relevant Research The research relevant to this study clusters around two main threads—research on reading aloud to young children and research on engagement and reading. Reading books aloud to children is a common, well-documented practice in early childhood classrooms. Research evidence indicates that reading aloud can have lasting effects on vocabulary development (Dickinson & Smith, 1994; Elley, 1989; Karweit, 1989), comprehension (e.g., Arnold & Whitehurst, 1994; Collins, 2004; Morrow, 1985, 1988; Teale, 2003) and, in some cases, print awareness (Justice & Ezell, 2002). Furthermore, there is a Engagement in Read Alouds 2 clear indication that teacher read aloud styles—the focus and content of teacher talk as well as the instructional strategies the teacher employs—are stable across texts and affect student vocabulary learning and comprehension (Dickinson & Smith, 1994; Martinez & Teale, 1993; Teale & Martinez, 1996). In other words, teachers have consistent, describable ways of interacting with the students and texts in read aloud activities, and their different patterns of interaction impact student learning differentially. The body of research on the motivational aspects of reading has established that there is a strong positive relationship between engaged time and higher student achievement (Cotton, 2001; Finn, Pannozzo, & Voelkl, 1995; Morrow & Weinstein, 1986; Nystrand & Gamoran, 1991), underscoring the importance of better understanding the complex relationship between student engagement and reading achievement. Additionally, research suggests that the best teachers of reading and writing are those who orchestrate literacy instruction in ways that keep students engaged—i.e., attentive (Samuels & Turnure, 1974) and involved (Farran, Kang, & Plummer, 2003) in literacy activities for extended periods of time (Pressley, Wharton-McDonald, Allington, Block, & Morrow, 1998). Engagement during reading indicates sustained personal commitment to creating understanding while one reads (Nystrand & Gamoran, 1991), or, in the case of preschoolers, while one listens to text. For research purposes, engagement is often measured according to a child’s visual focus and overt participation (verbalization or manipulation of materials) in an activity. Thus, both level of involvement and type of participation are components of child engagement. The majority of the literature on engagement in reading has typically focused on populations of elementary school and older students (Guthrie & Wigfield, 2000), but there are a few studies which examine the construct within the context of early reading instruction. Connor, Jakobson, Crowe and Meadows (2009) included engagement as a variable in their study of differentiated reading instruction, concluding that student engagement was positively related to reading achievement in first grade. Preschool children’s engagement and learning has certainly received considerable research attention (one need only look at the accumulated body of research on something like Sesame Street (e.g., Cole, Richman, & Brown, 2001) but, interestingly, it has not appeared often as a variable examined in correlational or intervention research on preschool read alouds. Watson’s (2008) investigation of the relationship between children’s involvement in whole-class narrative and expository teacher-led read alouds and vocabulary growth indicated that vocabulary outcomes increased as child involvement during book reading increased in the presence of teacher vocabulary facilitation, but little more in the way of empirical study has been conducted, despite Dickinson and colleagues’ call some years ago that, “when considering the nature of the book reading event, one should examine the teacher’s reading… and the nature of children’s engagement” (Dickinson, McCabe, & Anastasopoulos, 2002, p. 4). Certainly read alouds have the potential to foster substantive engagement in young children and so, the intersection of research on reading aloud to young children and research on engagement in learning appears to be relevant and needed. There is considerable consistency across recommendations of general strategies for how teachers can productively read aloud to preschoolers (e.g., McGee & Schickedanz, 2007; Van Kleek, et al., 2003), but this study sought to provide information about three specific features of preschool teachers’ orchestration of the read aloud activity—where/how children are seated, the duration of the read aloud, and the books used for reading—and their relation to student engagement. Research Questions The present research sought to examine relations between features of the teacher’s orchestration of the read aloud and children’s engagement. Specifically, this research addressed the following questions: What is the relationship between student verbal engagement, student nonverbal engagement and: (a) the proximity of the child to the teacher/book during the readings? (b) the duration of the read alouds? and (c) the texts employed in the read aloud? Engagement in Read Alouds 3 Method In order to provide the most in-depth analysis possible, this study employed a mixed methodology, incorporating multiple case study, discourse analysis, and post-hoc quantitative methods. The Research Context The settings and the children. This research was conducted within the context of an ongoing Early Reading First (ERF) project, from January through April, in two classrooms (Williams and Medina) that each served approximately 20 preschoolers—three- and four-year-old children. The purpose of the ERF project was to enhance classrooms as centers of early literacy excellence. The Opening the World of Learning (OWL) curriculum (Schickedanz, Dickinson, & Charlotte-Mecklenberg Schools, 2006) served as the core curriculum in both classrooms. Key to this ERF project was ongoing professional development for teachers and teacher assistants that consisted of monthly group meetings and intensive classroom-based coaching by project early literacy coaches. The schools were located in two high-risk neighborhoods in the city of Chicago. Ninety-eight percent of the children in each classroom were of African-American descent, and more than 80% of them qualified for free or reduced-price lunch. Parental consent was obtained for all participants, and all names are pseudonyms. Read aloud texts. In the OWL curriculum, read aloud books are narratives only (i.e., not concept books, songs/poems, non-narrative predictable texts, or information books). Both teachers (Williams and Medina) were observed reading the same three texts. The first text was Max’s Dragon Shirt (Max) (Wells, 1991). In this story Max and his older sister Ruby are sent to the department store to buy new pants for Max; but after a comedy of errors, they end up purchasing a shirt instead. In the next text, The Lion and the Little Red Bird (Lion) (Kleven, 1992), a lion uses things he collects in the forest (e.g., berries, grass, etc.) to make paintings of his adventures on the wall of his cave. Each morning, the lion emerges with a different color tail, and a curious bird wonders why. One stormy night the lion brings the bird into his cave. The bird sees the paintings and understands why the lion’s tail changes colors each morning. Raccoon on His Own (Raccoon) (Arnosky, 2001) was the third book observed during story time. In this text, a family of raccoons looks for food in the swamp when the baby raccoon strays away from his family and gets pulled onto the river in a boat. After a day full of adventure, baby raccoon is reunited with his family. Read aloud style. The read aloud methodology in OWL involves the whole class, and books are read aloud repeatedly to the children up to four times for different purposes each time. All observations used in this analysis were of the teachers’ first readings of each story. OWL recommends that the first reading be co-constructive (Dickinson & Smith, 1994) in nature, characterized by high amounts of teacher and student talk during the book reading, but minimal talk before or after. Analytic talk, extensive clarification, and support of student story understanding through use of comprehension supports (i.e., making connections, pointing out subplots carried out in the illustrations, summarizing) and vocabulary scaffolds (i.e., defining, rephrasing, supporting with gestures, intonation or illustrations) predominate in this style of reading. Both teachers had received extensive training in this read aloud style prior to the onset of the research, and so the read aloud styles across teachers and books were reported to be consistent by literacy coaches who observed in the classes regularly each week and by research assistants who periodically collected data in the classrooms. Data Collection and Coding Procedures Observation and video taping. Over a period of 15 weeks, live observations of six whole class, teacher-led read alouds were completed. Each read aloud was video taped using two cameras, one focused on the teacher and the other on the children. During the read aloud, researchers drew seat maps plotting each student’s location relative to the teacher and took field notes recording which student was talking during each instance of verbal engagement across the reading session (because it is often difficult to Engagement in Read Alouds 4 determine who is talking from the video artifact alone). There were several instances when more than two students said the same thing at the same moment, making it difficult to ascertain which students were the source of the verbalization in live data coding. In these instances, researchers recorded the verbalizations on a “multiple voices” list. The total time for each reading was also noted. Following the session, the entire read aloud was transcribed using the video recording and accompanying notes. Data coding—engagement. Both student nonverbal engagement and student verbal engagement were coded using the videos and transcripts. The Nonverbal Engagement Score represented the degree to which the student directly attended to the reading (i.e., eyes were on the teacher and/or the book being shown by the teacher). The seat maps, in conjunction with the videotapes, were used to code for student nonverbal engagement. The Verbal Engagement Score focused on collaborative efforts between the teacher and the students to understand the story (e.g., its plot, characters, themes). When the student said something that related to the content of the story, it was considered to be an instance of verbal engagement. The completed transcript and the observational notes about who was speaking served as the sources for coding student verbal engagement. The approach to coding nonverbal engagement was binary: each student was considered to be nonverbally engaged or nonverbally unengaged at every 2-minute interval. So, at 2 minutes (+/- 20 seconds), the researcher looked at every student and examined whether the student’s eyes were on the teacher/book (engaged) or not (i.e., the student was playing with a peer, getting tissues, simply not paying attention, etc.) and therefore was unengaged. For each read aloud, each student received an average Nonverbal Engagement (NVE) Score, calculated by dividing the total number of 2-minute intervals in which the student was engaged by the total number of intervals for the reading. To calculate inter-rater reliability for coding the NVE Score, the entirety of one of the six observations was randomly selected and double coded for nonverbal engagement, with a resulting Kappa = .83. Each student’s Verbal Engagement (VE) Score was determined by counting the number of idea units that he or she contributed to the meaning making within the reading session. Verbalizations not related to meaning making—e.g., requests to go to the bathroom, telling on a friend—were not counted. In instances where a student’s verbalization contributed to meaning making in one turn and that same student clarified his/her contribution and therefore ‘remained’ on the identical topic in subsequent turn(s), the contribution was counted once rather than multiple times. The following transcript segment and subsequent explanation illustrates how these issues were addressed in the Verbal Engagement coding: Line Speaker Verbalization 1 Williams: So here's the mother raccoon, she has baby raccoons with her, and it looks like they're using their claws to dig in the mud. They came to search for food. They're looking for their breakfast, so that's how they get food; they have to dig. What's this right here? Oh, I like the way Alexandrea raised her hand. What's this Alexandrea? 6 Alexandrea: A snake. 7 Williams: Oh. So not only do raccoons live in the swamp, but snakes live in the swamp, too. Hmm. Joe: I got something. I got something to share. There’s some raccoons that live in the open. 11 Williams: In the what? 12 Joe: In the open. and they dig in the garbage can. 2 3 4 5 8 9 10 Engagement in Read Alouds 5 13 Williams: Oh. Sometimes you can find raccoons in the alleys or next to garbage cans, where they get their food. Yes we talked about that. James? 15 James: Uh, I got something to share. Well, 16 Williams: Sorry, I can't hear my friend. Kelly, I can't hear my friend. 17 James: I have a red pen, then I tried to write. 18 Williams: This morning you were writing? 19 James: In my mama's school. 14 Alexandrea responds to Williams’ question (lines 4-5), contributing 1 idea unit to the conversation (line 6). James has three turns of discourse (lines 15, 17, and 19). None of these turns contributed to the meaning making related to the story, and therefore double strikethroughs mark the transcripts to indicate that James’ turns of conversation were tangential and unrelated to the story meaning. Joe has two turns (lines 9-10 and line 12) in which he is counted as contributing 1 idea unit to the conversation. His discourse in line 12 of this transcript is a clarification of what was stated in lines 9-10, so a double strike through was used to indicate that line 12 was not coded as contributing an idea unit even though Joe does add additional information, “they dig in a garbage can,” to his idea in lines 9-10. Earlier in the week (prior to the above conversation), Williams had led a discussion about animals and the food they need to grow. In that conversation, Joe and his classmates learned that raccoons can find their food out in the wild or that sometimes they dig in garbage cans in residential areas to find food. Because Joe’s turn in line 12 is pulling from a previous classroom discussion where the idea units in lines 9-10 and 12 had an established connection, we felt that Joe’s two turns were really acting more as a summary of the earlier discussion rather than two separate ideas contributing to the class effort to understand the story. There were only three instances in the entire data set where students’ additional, meaningful turns of conversations added additional information. Joe’s instance is an entirely new phrase. The students in the two other cases were completing a sentence or thought that had been interrupted by a peer or the teacher. One additional code existed in coding for verbal engagement. Recall that in live observation, researchers noted instances where more than a pair of students produced the same utterance at the same moment on a “multiple voices” list. Even with the help of video footage, we were unable to determine which individual students produced these utterances. We felt that these multiple voices did contribute to the teachers’ and students’ collaborative efforts to understand the story and so we coded these verbalizations as “multiple voices” in all instances where it was too difficult to ascertain all the students who were speaking. Data coding—proximity. In addition to coding for idea units, all turns of student verbal engagement were coded for proximity to the teacher. Seat maps were used to determine the physical location of each student throughout the read aloud. Talk that occurred within approximately six feet of the teacher/text was coded as “close” to the teacher/text. Talk that emerged from beyond this area was coded as “far” from the teacher/text. This second-level coding was used to examine the relationship between engagement and proximity to the text. Note about data sharing. From October through March, preliminary data and seat maps from each read aloud session were shared by the first author with each of the two teachers and the literacy coaches who were working them in their classrooms. Thus, the ongoing professional development associated with the context for data collection likely influenced the findings, but it was not our intention to withhold developing research insights from teachers. We have attempted to address this in interpreting study results, considering them as, in essence, part of a design experiment (Reinking & Bradley, 2008). Engagement in Read Alouds 6 Analysis & Results Our research questions led to analyses of three different factors. The first two analyses concerned the teacher’s orchestration of the read aloud activity, specifically examining how student engagement related to student proximity to the teacher/book and to the amount of time the reading took. The other focused on the relationship between the specific text being read and student engagement. Engagement related to proximity of the students to the teacher. Our first finding with respect to proximity is represented in Figures 1a and 1b. It can be seen that the two teachers in the study employed two different seating patterns for their read alouds at the beginning of the study’s data collection. One was what we named a Cluster pattern (Figure 1a), and this was characteristic of Medina’s classroom; the other was a Ring pattern (Figure 1b - children sat around the edge of the ‘circle time’ rug in the room), which was characteristic of Williams’ classroom. One outcome of these seating patterns is that more children in the Cluster pattern were closer to the teacher/text than there were in the Ring pattern. In addition, there was an unequal distribution of students seated Close (6 feet or less away from the teacher/text) and Far (more than 6 feet away from the teacher/text) across the two classrooms and three readings (approximately 59% of the children sat Far from the teacher/text). ---------------------Insert Figures 1a and 1b about here ---------------------Figures 2 and 3 present findings related to the above seating patterns and Nonverbal and Verbal engagement, respectively. We first examined the pattern of nonverbal engagement across both classrooms. As can be seen in Figure 2, students close to the teacher had higher Nonverbal Engagement (NVE) scores across each of the three books. Next, we examined each teacher individually (see Figure 3) for the NVE of Close and Far children, and an interesting picture emerged. In Williams’ classroom (Ring Pattern), Close children consistently had higher NVEs than Far children for each of the three books. In Medina’s classroom (Cluster pattern), results were more mixed: for Max (the first book read), Close children had an average NVE score of .92 whereas Far had an average of .79. That pattern was reversed for the other two books, with Far children’s scores being higher for both. We attributed these differences to two factors, as discussed in more detail below. One is that the proximity of the farthest of the Far children in the Ring pattern (Williams) was much greater than the farthest of the Far children in the Cluster pattern (Medina). The other is that, after the results of both teachers’ first readings were shared by the researchers, Medina made changes to the instructional aspects of her readings and in the seating patterns she employed as a result of the coaching she received related to proximity/engagement. Specifically, Medina moved the younger students up close and reported a concerted effort to call on students sitting in all areas of the read aloud space in her readings of Lion and Raccoon. Differences in NVE in Medina’s Close and Far students for Lion and Raccoon were so small that they were negated when we combined data across both teachers (Figure 2). ---------------------Insert Figures 2 and 3 about here ---------------------With respect to VE (verbal engagement) and proximity across the three readings, Figure 4 shows that for both teachers’ seating patterns, students seated Close to the text/teacher were more verbally engaged in the meaning making process than those who were seated Far. It is also important to add that this discrepancy is more pronounced in the Ring pattern than in the Cluster pattern. Engagement in Read Alouds 7 ---------------------Insert Figure 4 about here ---------------------Engagement related to the duration of the read aloud. Our next query probed the relationship between student engagement and the duration of the read aloud sessions. We were prompted to conduct this analysis because of results found during preliminary data gathering on read alouds in these preschool classrooms (conducted before the videotaping reported on in this paper that enabled the closer analyses). During that early phase of the work we noted that when readings went on for a long time, there was an accompanying drop in child engagement. To examine this phenomenon more closely and understand what might constitute ‘too long’, each reading in the data set of 6 readings (range in length: 10 min – 32 min.) was divided into segments (0-12 min; 12-20 min; more than 20 min), and an average NVE score and summed VE scores were calculated for each child for each time interval. This yielded three segments— Early, Middle, Late—for each reading. Early and Middle time segments were observed for both teachers in all three texts. Late segments were only observed in Raccoon and for both teachers. Figure 5 depicts the results of these analyses for both child nonverbal engagement and verbal engagement across the three segments for Raccoon only because this was the only instance where the duration of the read aloud and discussion exceeded 20 minutes for both teachers. For NVE, Figure 5 indicates that average engagement successively decreased across each of the three segments of the reading, to the point where, in the Late segment, it was approximately 25% less than during the Early phase of the reading. This accorded with the preliminary finding from early observations that after approximately 20 minutes, preschoolers’ nonverbal engagement tapered off quite a bit. Findings for verbal engagement over the course of the long readings of Raccoon were quite different. It can be seen in Figure 6 that the mean VE score decreased from the Early phase to the Middle phase but then increased Late in the readings. This at-first-glance puzzling finding was explained by a qualitative content analysis of what occurred verbally in the Late phase of the readings. It turned out that the teachers engaged in a protracted quizzing of students about this book, asking multiple children the same comprehension follow-up questions, as illustrated, below: 1 Williams: First question: what other kinds of animals live in the swamp besides the raccoon? Shane? 3 Shane: Snakes. 4 Williams: Snakes. Jaylen? 5 Jaylen: Alligators. 6 Student: I'm tired. 7 Williams: Alligators. Amaya? 8 Amaya: Snakes. 9 Williams: Yeah, we saw snakes. Xavier? 10 Xavier: Turtles. 11 Williams: Turtles. We saw them. Remember they scattered away when raccoon put his claw in the water. Sit down. Terrence? Do you remember any other animal? Joshua? 2 12 13 Engagement in Read Alouds 8 14 Joshua: Ducks. 15 Williams: Ducklings. A merganser and her ducklings. 16 Jackie: Williams? 17 Williams: Kerry? 18 Kerry: Alligators and snakes. 19 Williams: Yeah, we saw alligators and snakes. Tajane, Did you see anything different? 20 Thus, the verbal engagement looked high because children were required to answer questions, but other than the children who were being questioned, engagement in general had dropped off by Late in the reading, as suggested by the NVE scores. ---------------------Insert Figures 5 and 6 about here ---------------------Engagement related to the text that is read. The last set of queries examined the relationship between the book involved in the read aloud and student engagement, both nonverbal and verbal engagement. Although these were not a central facet of our initial data analysis plan, they were conducted because the “feel” of read alouds was different across the three texts; it appeared as though certain texts were more—or less—riveting than others. The analyses that follow represent our attempt to more systematically examine these anecdotal observations. For nonverbal engagement, our coding allowed for scores ranging from 0 to 1 (see discussion in Methods section describing how this ratio was calculated). Table 1 presents the observed range, mean, and standard deviations of Nonverbal Engagement Scores for each book across the two teachers’ readings. -----------Insert table 1 about here -----------Split-plot ANOVA was used to determine the relationships among students’ nonverbal engagement, the teachers, and the texts during each read aloud session, with teacher serving as the between-subjects factor and the texts serving as the within-subjects factor. The NVE Scores served as the dependent measure. Assumptions of independence, normality, and sphericity were met for the analysis, but the assumptions of homogeneity and additivity were not met. Tables 2 and 3 present results from these analyses. Table 2 shows no significant relationships for either the within-subjects factor of text or the interaction of text and teacher. Table 3 shows no significant effect for the between-subjects factor of teacher. In other words, the children’s nonverbal engagement was not related to the particular book being read, either with respect to each teacher’s reading or for both teachers’ readings considered together. -----------Insert tables 2 and 3 about here -----------Verbal Engagement scores were based on frequency counts of the number of child turns of conversation that contributed to story understanding. The possible range of scores for VE was from zero Engagement in Read Alouds 9 to infinity. Table 4 shows the observed range of scores, means, and standard deviations for students’ verbal engagement for each text. Three (1 x 2) Hotelling’s T tests (using a Bonferroni adjustment to account for inflated error effects) were used to explore the interaction between overall mean student verbal engagement and text. The N in Table 4 differs from the N in Table 1 because of the “multiple voices” coding, as previously explained. -----------Insert tables 4 and 5 about here -----------Table 5 indicates that significant effects were found for all comparisons, suggesting that the student verbal engagement varied significantly across texts. The most student verbal interaction occurred in the reading of Raccoon, and students offered the least amount of verbal interaction in the reading of Lion. Why did Raccoon engender the most verbal engagement from students? We see two viable possibilities. One is that the book itself was more (1) ‘interesting’ or (2) ‘difficult’ and therefore either (a) invited or (b) required more talk. A second is that because of the ongoing data sharing and coaching about how to deepen students’ comprehension, teachers were trying to facilitate more student talk in this third reading of the study than they had in the first two that were observed/videotaped. This possibility is supported by the fact that Medina, after discussion with her coach following the first two readings, actually moved children to different locations in the cluster of children surrounding her during the Raccoon reading and reported that she was making a conscious effort to call on children who, up to that time, had not been very verbally engaged. Qualitative analyses indicated that the greater verbal engagement found in Raccoon also related to the kinds of things children talked about during the read aloud. The following excerpt from one classroom’s discussion illustrates the differences identified in the qualitative analysis: 1 Text: The raccoon reached up to grab a sturdy branch 2 Medina: That means a nice strong branch. But he's a baby, he doesn't reach. 3 Text: And climb out of the moving boat. But the branch was too high. 4 Medina: He's a baby. 5 Joshua: He can go up to the tree and hold onto that tree. 6 Medina: But Joshua, he's kind of far away from that branch, from the trunk of the tree, and it's a big tree. And he's a baby and he's trying to grab the branch, but he can't, it's too high. Yes, Deborah. 9 Deborah: Uh, he might fall into the water. 10 Medina: That's smart. He might fall in the water, you guys. He's scared. Let's continue reading. 12 Text: The boat glided under another branch. 13 D’Asia: A snake! 14 Joshua: A snake! 15 Medina: That means it's going under the branch nice and slow. 16 Text: This branch was low enough to climb. 17 Medina: That means he can climb on this branch. 7 8 11 Engagement in Read Alouds 10 18 James: I see the snake! 19 Medina: Mmm hmmm… (‘let’s see…’) 20 Text: But the raccoon saw a shape of a long snake draped across the branch, and he ducked down. Medina: He saw the snake, so he ducked because he didn't want the snake to see him, and the snake is hanging like drapes. That means he's just hanging on the branch, you know when snakes hang, they kind of just hang. And he couldn't jump on the branch cause otherwise the snake was going to attack him. So he's nice and quiet. Cause he knows that if he moves, the snake might what? 28 Imani: It might get him. 29 Medina: Smell him or get him, yeah. Do you think he's scared right now? 30 Students: Yes. 31 Medina: Oh, I think so. Yes, Ariel. 32 Ariel: If the raccoon climbs on the branch, the tree, that means he can't, that means he's not old enough. 34 Medina: You don't think the raccoon is old enough? 35 Ariel: Yes. 36 Joshua: He could barely reach. 37 Medina: Ok, that's a good observation. He could barely reach, ok. 38 Ariel: But if he go on the branch the snake bites him. 39 Medina: That's why he was smart enough to stay still. He's not moving. 40 Joshua: He be like this. (making a freezing pose) 41 Morgan: What's that? 42 Medina: Yes. He's still. He froze. Good job. Yes, let's read. 21 22 23 24 25 26 27 33 In the preceding example, Deborah refers to pictures to support her reasoning (line 9). One student explains why she thinks the raccoon cannot reach the branch, despite ineloquent vocabulary and syntax (Ariel, lines 32-33). In lines 32-39, one can see glimpses of distributed meaning construction when Joshua helps clarify what Ariel means. In this reading, as contrasted with the other books, children helped one another understand the plot, as well as made more predictions and connections. There were comparatively few such verbal interactions with Max or Lion. Discussion Both the statistical and qualitative analyses conducted in this study yielded insights into the relationships between preschool students’ engagement in read alouds and the books used in read alouds as well as how the read alouds were orchestrated. The first factor that emerged as related to engagement was proximity of the children to the teacher reader/book. Results showed that children Far from the teacher Engagement in Read Alouds 11 (i.e., more than 6 feet) tended to be less engaged overall than children closer (see Figure 1b). The seating patterns that the two teachers employed during their readings helped surface this result more clearly. Williams’ Ring pattern (a seating pattern used by many preschool teachers) resulted in quite a few children being at a great distance from the teacher. The illustrations are a central part of the read aloud experience for children. Preschool teachers most often share trade-book sized picture books (rather than big books) with children; thus, the size of the illustrations becomes an issue to be considered. Children at the corners of the typical preschool or kindergarten rug have greater difficulty seeing this important part of the activity, a factor that likely influences a greater degree of non-engagement. Children seated closer to the teacher/text have a clearer view of the images used to convey the story’s meaning and may, therefore, become more involved in meaningful ways with the reading and discussion. Also, it was clear that Medina found great face validity in the finding of proximity being related to engagement. Her efforts at moving children to different locations in the cluster (see Figure 1a) represented a way of increasing engagement for specific children that became part of her instructional strategies during read alouds. We believe that this proximity-engagement facet of read alouds warrants more controlled research; especially interesting would be systematically studying this factor as it relates to the engagement of children who exhibit different degrees attention in the range of classroom activities and with children of different ages. The next trend that emerged from the data related to children’s engagement in the read aloud activity vis-à-vis the elapsed time of the book reading/discussion. Among these preschoolers, nonverbal engagement tapered off considerably after 20 minutes of the read aloud experiences. Verbal engagement was maintained beyond 20 minutes, but it seems only because teachers directly questioned individual children in ways that required responses during this period of the read alouds studied. Our suspicion, based on the NVE finding for the readings of Raccoon, is that the direct questioning is masking what typically would happen with a more ‘invitational’ approach by the teacher to children’s verbal responses—that three- and four-year-olds would not have as much to offer by this point in a read aloud as they had previously because the attention of many in a large group setting like this would wane substantially. This is an important factor for teachers and curriculum coordinators to consider. Longer texts and corresponding discussions may yield less in terms of student learning, unless the book or the topic is particular gripping to the children involved. It is also possible, however, that the point in the school year may play a role in the degree of engagement—longer read alouds may work better in the spring or summer than they did the previous fall. This would be an interesting point for future systematic study. At present, given our observations, we suspect that read alouds lasting 15-20 minutes represent an ideal for this age group–enough time to engage students in some conversation about the text, but not so long that one sacrifices students’ “unforced” engagement. The various book readings we have observed within the context of our Early Reading First project that lasted longer than 20 minutes exhibited a preponderance of management talk during the Late phase (e.g., moving students from one place to another, teachers waiting until everyone was quiet, teachers responding to one child “telling” on the other) and otherwise ‘off-task’ behaviors by the children. The third insight related to the textual (book) factor. The data on nonverbal engagement indicated that, for both teachers and across all texts, on average the preschoolers were engaged over 80% of the time, showing that these children were quite engaged in the read alouds the teachers conducted. These data are comparable to the findings of other researchers investigating young children’s engagement in reading (Connor, et al., 2009; Watson, 2008) and may be a consequence of the read aloud style (which was constant across classrooms because teachers were conducting their readings as outlined in the OWL curriculum), high interest texts, or, as is likely, a combination of the two. It is worth noting that Max’s Dragon Shirt produced the highest nonverbal engagement in both classrooms. This could be because Max and Ruby are popular characters on television in addition to storybook characters. Or, conversely, it could be related to the language complexity of the story. Isolation and manipulation of text and style variables in future research would be helpful in sorting out the causes of differing levels of student engagement. Engagement in Read Alouds 12 Perhaps the methods recently employed by Price, van Kleek, and Huberty (2009) – systematically examining the relationship between student engagement and the number and function of the illustrations, the total number of sentences, mean length of sentences, vocabulary diversity, and number of new different words presented in the text – would be a useful next step in this line of research. Our final consideration relates to the limitations of our data sources. Both teachers had received significant amounts of targeted professional development, read in lively, engaging ways, and they made good use of the vocabulary and story understanding supports recommended in the OWL curriculum, so they may not be representative of ‘typical’ preschool teachers in that respect. Also, the context of the present research meant that we shared our formative findings with the classroom teachers and their literacy coaches, thus intervening in how they conducted the read alouds and, consequently, affecting study findings. Had we not discussed our thoughts about seating arrangements and distribution of verbal engagement, teachers may not have made concerted efforts to call on students sitting at the back of the cluster in our observations of The Lion and the Little Red Bird (Kleven, 1992) or Raccoon on his Own (Arnosky, 2001), and we may not have seen Medina work to rearrange her seating pattern in the attempt to incorporate more discussion for Raccoon on his Own. Thus, our ongoing analyses appear to have unintentionally replicated certain features of design based research (Design-Based Research Collective, 2003). Specifically, our “methods respond[ed] to emergent features of the setting” (p. 6), driving the development of an intervention. Our final analyses, however, treated the study variables as though they were not manipulated. Such a situation may not be ideal for describing the issues highlighted for study, but we put precedence on our obligation to improve teachers’ practice and extend children’s learning. Certainly, it remains for more systematic studies to be conducted on these topics, but we feel that the circumstances under which the data for this study were collected did not negate the preliminary insights derived from the work. In fact, our confidence in the findings increased because the results of proximity and reading length held up even when the teachers were sensitized to our ongoing thinking about patterns and trends. Conclusion Every day in early childhood classrooms across the United States young children are invited to participate in story time. Teachers choose books and read those books in ways intended to support vocabulary development, story understanding, and other early literacy skills and dispositions. Our findings from this exploratory study indicate that, depending on factors such as where children sit, how much time the teacher devotes to the reading, and the books chosen for the read aloud, student engagement (and therefore story understanding and student learning) can be enhanced or undermined. These variables warrant further manipulation and more detailed analysis. Engagement in Read Alouds 13 References Arnold, D. S., & Whitehurst, G. J. (1994). Accelerating language development through picture book reading: A summary of dialogic reading and its effects. In D. K. Dickinson (Ed.), Bridges to literacy: Approaches to supporting child and family literacy (pp. 103-128). Cambridge, MA: Blackwell. Arnosky, J. (2001). Raccoon on his own. New York: Puffin. Brozo, W., Shiel, G., & Topping, K. (2007). Engagement in reading: Lessons learned from three PISA countries. Journal of Adolescent and Adult Literacy, 51(4), 304-315. Cole, C., Richman, B., & Brown, S. (2001). The world of Sesame Street research. In S. Fisch & R. Truglio (Eds.), G is for growing: Thirty years of research on children and Sesame Street (pp. 147-180). Mahaw, NJ: Erlbaum. Collins, M. F. (2004). ESL preschoolers' English vocabulary acquisition and story comprehension from storybook reading. (Doctoral dissertation, Boston University, 2004). Dissertation Abstracts International, 65, 824. Connor, C. M., Jakobson, L. K., Crowe, E. C., & Meadows, J. G. (2009). Instruction, student engagement, and reading skill growth in Reading First classrooms. Elementary School Journal, 109(3), 221. Cotton, K. (2001). Educational time factors. School improvement research series, Close-up no. 8. Retrieved February 16, 2007, from http://www.nwrel.org/scpd/sirs/4/cu8.html. Design-Based Research Collective (2003). Design-base research: An emerging paradigm for educational inquiry. Educational Researcher, 32(1), 5-8. Dickinson, D. K., McCabe, A., & Anastasopoulos, L. (2002). A framework for examining book reading in early childhood classrooms. CIERA Report #1-014. Retrieved June 8, 2005, from http://www.ciera.org/library/reports/inquiry-1/1-014/1-014p.pdf. Dickinson, D. K., & Smith, M. W. (1994). Long-term effects of preschool teachers' book readings on low-income children's vocabulary and story comprehension. Reading Research Quarterly, 29(2), 174-187. Elley, W. B. (1989). Vocabulary acquisition from listening to stories. Reading Research Quarterly, 24(2), 174-187. Farran, D. C., Kang, S., & Plummer, C. (2003). Child observation in preschool manual. Unpublished Manuscript. Nashville, TN: Vanderbilt University. Finn, J. D., Pannozzo, G. M., & Voelkl, K. E. (1995). Disruptive and inattentive-withdrawn behavior and achievement among fourth graders. The Elementary School Journal, 95, 421-434. Guthrie, J. T., & Wigfield, A. (2000). Engagement and motivation in reading. In P. D. Pearson, R. Barr, M. Kamil & P. Mosenthal (Eds.), Handbook of reading research (Vol. III, pp. 403-422). Mahaw, NJ: Erlbaum. International Reading Association, & National Association for the Education of Young Children (1998). Learning to read and write: Develomentally appropriate practices for young children. Washington, D.C.: National Association for the Education of Young Children. Justice, L. M., & Ezell, H. K. (2002). Use of storybook reading to increase print awareness in at-risk children. American Journal of Speech-Language Pathology, 11(1), 17-29. Engagement in Read Alouds 14 Karweit, N. (1989). The effects of a story-reading program on the vocabulary and story comprehension skills of disadvantaged prekindergarten and kindergarten students. Early Education and Development, 1, 105-114. Kleven, E. (1992). The lion and the little red bird. New York: Puffin Books. Martinez, M. G., & Teale, W. H. (1993). Teacher storybook reading style: A comparison of six teachers. Research in the Teaching of English, 27, 175-199. McGee, L. M., & Schickedanz, J. A. (2007). Repeated interactive read-alouds in preschool and kindergarten. Reading Teacher, 60(8), 742-751. Morrow, L. M. (1985). Retelling stories: A strategy for improving young children's comprehension, concept of story structure, and oral language complexity. The Elementary School Journal, 85(5), 647-661. Morrow, L. M. (1988). Young children's responses to one-to-one story readings in school settings. Reading Research Quarterly, 23(1), 89-107. Morrow, L. M., & Weinstein, C. S. (1986). Encouraging voluntary reading: The impact of a literature program on children's use of library centers. Reading Research Quarterly, 21(3), 330-346. National Early Literacy Panel (2008). Developing early literacy: Report of the National Early Literacy Panel. Washington, DC: National Institute for Literacy. Nystrand, M., & Gamoran, A. (1991). Instructional discourse, student engagement, and literature achievement. Research in the Teaching of English, 25, 261-290. Pressley, M., Wharton-McDonald, R., Allington, R., Block, C. C., & Morrow, L. (1998). Nature of effective first-grade literacy instruction. CELA Research Report 11007. Retrieved February 17, 2007, from http://cela.Albany.edu/1stgradelit/index.html. Price, L. H., Kleeck, A. v., & Huberty, C. J. (2009). Talk during book sharing between parents and preschool children: A comparison between storybook and expository book conditions. Reading Research Quarterly, 44(2), 171-194. Reinking, D., & Bradley, B. A. (2008). What are the methods of formative and design experiments? In D. Reinking & B.A. Bradley (Eds.), On formative and design experiments: Approaches to language and literacy research (pp. 43-88). New York, NY: Teachers College Press. Rogoff, B. (1990). Apprenticeship in thinking: Cognitive development in social context. Oxford: Oxford University Press. Samuels, S. J., & Turnure, J. E. (1974). Attention and reading achievement in first-grade boys and girls. Journal of Educational Psychology, 66(1), 29-32. Schickedanz, J., A., Dickinson, D. K., & Charlotte-Mecklenberg Schools (2006). Opening the world of learning. New York: Pearson. Sipe, L. (2008). Storytime: Young children's literary understanding in the classroom. New York: Teachers College Press. Teale, W. H. (2003). Reading aloud to young children as a classroom instructional activity: Insights from research and practice. In A. Van Kleeck, S. A. Stahl, & E. Bauer (Eds.), On reading books to children: Parents and teachers (pp. 114-139). Mahaw, NJ: Erlbaum. Teale, W. H., & Martinez, M. G. (1996). Reading aloud to young children: Teachers' reading styles and kindergartners' text comprehension. In C. Pontecorvo, M. Orsolini, B. Burge, & L. B. Resnick (Eds.), Children's early text construction (pp. 321-344). Hillsdale, NJ: Erlbaum Engagement in Read Alouds 15 Teale, W. H., Paciga, K. A., & Hoffman, J. L. (2007). Beginning reading instruction in urban schools: The curriculum gap ensures a continuing achievement gap. The Reading Teacher, 61(4), 344-348. Teale, W. H., Yokota, J., & Martinez, M. G. (2008). The book matters: Evaluating and selecting what to read aloud to young children. In A. DeBruin-Parecki (Ed.), Effective early literacy practice: Here’s how, here’s why (pp. 101-121). Baltimore, MD: Paul Brookes. Van Kleek, A., Stahl, S., & Bauer, D. M. (2003). On reading books to children: Parents and teachers. Mahaw, NJ: Erlbaum. Vygotsky, L. S. (1978). Mind in society: Cambridge, MA: Harvard University Press. Watson, B. G. (2008). Preschool book reading: Teacher, child and text contributions to vocabulary growth. Doctoral dissertation, Vanderbilt University. Retrieved December 12, 2008 from http://etd.library.vanderbilt.edu/available/etd-03282008104911/unrestricted/BWatsondissertation.pdf. Wells, R. (1991). Max's dragon shirt. New York: Puffin. Engagement in Read Alouds 16 Tables Table 1. Summary of Nonverbal Engagement Scores (NVE) N Range Mean SD Max 33 .43-1.00 .87 .17 Lion 33 .22-1.00 .77 .23 33 .20-1.00 .81 .18 Raccoon Table 2. Within-subjects Effects: Nonverbal Engagement Source Text Type III Sum of Squares .166 Df Mean Square F Sig. 2 .083 3.061 .054 .798 .455 Text x Teacher .043 2 .022 Error (Text) 1.677 62 .027 Table 3. Between-subjects Effects: Nonverbal Engagement Source Teacher Error Type III Sum of Squares .000 1.884 Df Mean Square F Sig. 1 .000 .008 .931 31 .061 Table 4. Summary of Verbal Engagement N Minimum Maximum Mean Std. Deviation Max 33 .00 11.00 1.50 2.43 Lion 33 .00 7.00 1.21 1.67 33 .00 17.00 5.03 4.48 Raccoon Engagement in Read Alouds 17 Table 5. Examination of the Relationship between Verbal Engagement and Text Verbal Engagement Variable 1 Verbal Engagement Variable 2 n M SD n M SD F MAX x LION 34 1.88 2.99 34 1.71 2.71 8.08* MAX x RACCOON 34 1.88 2.99 34 5.65 4.92 22.62* LION x RACCOON 34 1.71 2.71 34 5.65 4.92 22.09* * p < .01 Figures Figure 1a. Seating Patterns: Cluster Engagement in Read Alouds 18 Figure 1b. Seating Patterns: Ring Figure 2. Nonverbal Engagement Score related to proximity: Combined across classrooms Engagement in Read Alouds 19 Figure 3. Nonverbal Engagement Score related to proximity: Individual classrooms Figure 4. Verbal Engagement Score related to proximity: Combined across classrooms Engagement in Read Alouds 20 Figure 5. Nonverbal Engagement Score related to duration of the read aloud: Combined across classrooms Figure 6. Verbal Engagement Score related to duration of the read aloud: Combined across classrooms