Abstract
Shared reading is a powerful literacy intervention for the development of early communication and literacy skills. For students with extensive support needs who are beginning communicators, shared reading has the potential to support the development of initiation and expressive symbolic communication skills. This descriptive, multicase, mixed methods study investigated changes in teacher shared reading practices and perspectives, as well as the communication skills of students with extensive support needs associated with completing professional development focused on a responsive and interactive approach to shared reading. The sample included four classroom teachers and 13 students with extensive support needs in one rural U.S. public elementary school. Teachers increased their use of targeted interactive shared reading practices from 50% to 90%. Students increased the number of expressive communication skills they exhibited (p = .031, r = .60), as well as the highest level of their symbolic communication (z = 2.16, p = .031, r = .60). Thematic analysis of teacher interviews identified shifts in their perceptions of their role, expectations, and ability to support student communication development during shared reading, which converged with the changes noted from fall to spring during observations of teacher practice and students’ communication skills during shared reading.
Keywords
Shared reading is defined as the interaction between an adult and one or more children as they read a book together (Ezell & Justice, 2005). Language-based interactions during shared reading are known to positively impact children’s oral language development (National Early Literacy Panel, 2008). Unfortunately, many children with extensive support needs (ESN) cannot participate in these language-based interactions during shared reading because they are beginning communicators (i.e., communicate exclusively in nonsymbolic ways; Erickson & Geist, 2016). Supporting the language and communication development of these beginning communicators requires responsive approaches to interaction during which adults accept and contingently respond to the children’s unique and idiosyncratic communication behaviors, while providing language input that aligns with the child’s presumed interests and intent (Ainsworth et al., 1978; Bornstein & Tamis-LeMonda, 1989; Landry et al., 2001). In a shared reading context with children with ESN, adults implementing a responsive approach work to attribute meaning to all child behaviors, rely on comments that reflect perceived child interests in the book, and encourage child participation without requiring it (Erickson & Koppenhaver, 2020).
Children With ESNs in Rural Schools
As a subgroup of children receiving special education services under a variety of IDEA (Individuals with Disabilities Education Improvement Act, 2004) eligibility categories (i.e., autism, multiple disabilities, intellectual disability), children with ESN require (a) extensive, repeated and individualized instruction, (b) substantially adapted and modified instructional materials, and (c) individualized methods of accessing information to learn, maintain, generalize, demonstrate, and use skills across settings (Erickson, 2013). More than 90% of children with ESN receive this instruction within separate special education classrooms or separate schools where they have little to no interaction with their peers without disabilities (Burnes & Clark, 2021; Morningstar et al., 2017). This is despite evidence that there are systemic problems with segregated special education settings (Erickson et al., 2021; Kleinert et al., 2015) and children with ESN in general education settings have greater access to literacy instruction (Ruppar et al., 2018) with improved overall academic outcomes (Shogren et al., 2015). As we work to create a system that provides greater access to general education settings, we must continue to find ways to improve instruction in separate special education settings (Brock, 2018; Morningstar et al., 2017). Shared reading has the potential to support instruction and promote positive language, literacy, and communication outcomes for children with ESN regardless of setting. However, the adults who read with them must have the knowledge and skills required to teach, build, and scaffold communication and initiation during interactions around books.
The challenges of separate special education settings are likely more pronounced for the approximately 15% of children with ESN who are served in rural communities (Erickson & Geist, 2016). Many rural communities struggle to provide quality special education services due to personnel shortages, difficulties in recruiting and retaining qualified special education teachers, and disparities in community resources relative to urban communities (Rude & Miller, 2018). Rural special education teachers of children with low incidence disabilities report that they experience isolation as a result of being the only teacher with expertise in the needs of their students within their entire district (Abell et al., 2014). As is reported more broadly among special education teachers, it is likely that special education teachers in rural contexts have knowledge gaps and limited self-efficacy in early literacy instruction, as well as little training in communication despite the complex needs of their students (McCombes-Tolis & Feinn, 2008; Pennington et al., 2009). Thus, they require professional development focused on early literacy and communication instruction to effectively address the learning needs of their students with ESN.
Approaches to Shared Reading for Students With ESNs
Most of the evidence base in shared reading with students who have ESN supports structured, teacher-directed instruction that requires students to demonstrate specific responses to stimuli or questions. Initially, this body of evidence focused on teaching students to respond correctly to specific stimuli (e.g., name embedded in story, prompt to turn page) and/or known-answer questions (Browder et al., 2008; Mims et al., 2009; Mucchetti, 2013). Subsequently, studies in shared reading using frameworks of systematic instruction focused on teaching students with ESN strategies for answering questions requiring more than literal recall. These strategies involved highlighting sections of text where answers could be found (Mims et al., 2018), providing increasingly focused sections of highlighted text with the correct answer (Spooner et al., 2015), and using graphic organizers (Mims et al., 2012). These approaches have increased the ability of students with low incidence disabilities who have symbolic communication to respond to teacher-directed instruction during shared reading, but they have not focused on supporting students with ESN, who are beginning communicators, in initiating and otherwise leading interactions during shared reading.
Adult-directed interactions during shared reading, especially with students with ESN who are beginning communicators, are often unbalanced and fail to support the development of the students’ ability to initiate comments or pose their own questions (Fleury & Schwartz, 2017). In addition, when adults control shared reading interactions with a focus on eliciting predetermined answers, less time is spent supporting students in making connections with the text or learning to initiate and sustain interactions. Emergent and beginning readers must learn to connect and eventually integrate information in the text with their prior knowledge of the world (Duke et al., 2011; Kintsch, 2004). Learning to do this requires adult reading partners to demonstrate these types of connections and create opportunities for students to initiate and engage in this type of thinking during shared reading.
A responsive and interactive approach to shared reading, in which adult contributions are student-oriented and provide language models that are just beyond the students’ current language ability, can facilitate interaction, promote student initiations, and support communication development for children with a range of abilities, including ESN (Cole et al., 2002; Erickson, 2017; Landry et al., 2017). Specific responsive strategies might include following the student’s lead, providing sufficient wait time to encourage student contributions, using open-ended questions or topic-continuing comments linked to student interest, monitoring student engagement, or repeating and expanding on student communication attempts (Erickson & Koppenhaver, 2020; Landry et al., 2012; Skotko et al., 2004). These responsive behaviors during shared reading increase initiations and responses among children with typical development who are at-risk for literacy learning challenges (Landry et al., 2012, 2017). When employed during shared reading with children who have ESN, these responsive behaviors are known to improve communication and initiation outcomes. For example, these responsive behaviors during shared reading increased labeling, commenting, and other initiations among children with Rett syndrome (Skotko et al., 2004), and increased attention to text, interaction, and initiations among other students with ESN (Cheek et al., 2022).
A responsive and interactive approach to shared reading may be particularly effective with students with ESN who are beginning communicators, do not initiate communication around books, or are not engaged during shared book reading. A responsive and interactive approach provides a context in which adults can attribute meaning to the behaviors of students who are beginning communicators, which subsequently teaches students how their actions affect others, regardless of whether or not the behavior was intentional or symbolic (Cress et al., 2007; Yoder et al., 2001). Furthermore, when this responsive and interactive approach is combined with adult efforts to make comments paired with key graphic symbols displayed on an organized augmentative and alternative communication (AAC) system, they can support student comprehension and help students communicate symbolically about books (Barker et al., 2013; Sennott et al., 2016). When adult contributions follow the student’s lead and are contingent on the student’s communicative attempts, it reduces the information processing load, supports understanding, and increases the student’s motivation to attend as well as initiate and respond, while promoting more frequent, varied, and complex communication efforts (Girolametto, 1988; Girolametto et al., 2007; Girolametto & Weitzman, 2002).
An Interactive and Responsive Approach to Shared Reading
It is important that special education teachers in rural settings are empowered to support students in participating in shared reading interactions in ways that result in children who initiate, make connections with their prior knowledge and experience, and respond to adults. To begin to address this need, the authors, as part of a larger team and project, developed a set of professional development modules focused specifically on helping teachers improve the quality of their shared reading interactions with a diverse range of students who have ESN and other complex needs. These modules centered on teaching an interactive and responsive approach to shared reading based on an early language and literacy framework for young children (Cole et al., 2002). The approach, Follow the CAR, involves (a) Commenting and waiting, (b) Asking for participation and waiting, and (c) Responding by adding a little more. The approach supports students in learning to take the lead, and it improves their language, literacy, and communication outcomes regardless of the students’ ages. To support beginning communicators in interacting, building language skills, and learning how to take the lead during shared reading interactions, this framework has been adapted and applied to shared reading with students who have ESN while reading a book with an adult partner (Erickson & Koppenhaver, 2020), with recently reported evidence of its positive impact on this population (Cheek et al., 2022). When adults implement Follow the CAR, they are encouraged to use comments rather than questions (Justice et al., 2009) for the purpose of demonstrating and encouraging students to construct understandings and use language during shared reading. In addition, Follow the CAR focuses on adult responsivity to perceived student interests, intents, and initiations (Barnes et al., 2017; Bellon & Ogletree, 2000) for the purposes of building joint attention, communication, and language. During this study, these modules were presented in person; however, they are now available as open-source, online, on-demand resources (https://sharedreader.org). This is particularly relevant to special education teachers in rural areas who are likely to experience isolation and limited access to training and specialist knowledge regarding their students with low-incidence disabilities (Abell et al., 2014; Rude & Miller, 2018).
The Tar Heel Shared Reader model has a dual focus of positively impacting change in teacher practice and improving student communication and literacy outcomes through shared reading. The goal of the project was to iteratively develop and evaluate the supports and resources teachers need to provide a responsive and interactive approach to shared reading. During this iterative process, descriptive data were collected about students and teachers as shared reading was included as part of the daily classroom instruction. The process served to identify parts of the Tar Heel Shared Reader implementation model that needed to be refined; however, the results also provide preliminary evidence of the impact of the model on teacher behaviors and perceptions, as well as student communication skills. The purpose of this study was to explore the impact of the model’s professional development regarding responsive and interactive shared reading on targeted teacher practices and perceptions, as well as the communication outcomes of children with ESN. The data are quantitative and qualitative. The quantitative data were generated from structured observations of teacher behaviors during shared reading and assessments of student communication skills. The qualitative data were generated during teacher interviews and classroom observations. The data were used to answer the following research questions:
Method
A descriptive, multicase investigation (Greene et al., 1989) was conducted using mixed quantitative and qualitative methods to enhance the researchers’ understanding and interpretation of the impact of the professional development modules (Collins et al., 2006). A descriptive multicase investigation with mixed methods was selected because it supports a data-rich description of the complex interactions and behaviors of teachers and students during implementation of the shared reading intervention within the natural context of the classroom from a variety of lenses (Baxter & Jack, 2015). By converging and corroborating quantitative and qualitative data, it increases the validity of results for complex phenomenon associated with classroom interactions while simultaneously minimizing bias associated with method, investigator, theory, and context (Greene et al., 1989), and enhancing the results (Collins et al., 2006). The study was approved by the Institutional Review Board at the university where the authors are employed and the school system. All teachers and classroom staff as well as the parents of the child participants provided written consent for their participation in the study. The child participants in this study lacked the receptive and expressive language skills to meaningfully provide assent for their participation. However, their behaviors were carefully monitored with teachers encouraging their participation in the ways they would encourage all students to participate in all aspects of the school day. Children were never forced to participate in any shared reading interactions that were part of this study.
Settings
The current data were collected during the second year of a 5-year project from four special education classrooms serving students with ESN in one general education public school in a rural school system in the southeastern United States. The four classrooms were self-contained special education classrooms (i.e., classrooms in which one teacher provides instruction to the same group of students for the majority of the instructional day). In the year this study was completed, the school had a total enrollment of approximately 665 students, with about 65% of the students receiving free or reduced lunch. The racial demographics of students in the school were as follows: 50% White, 10% Hispanic, 34% Black, 1% Asian, and 5% Multiracial.
Participants
In each of the classrooms, the teachers and instructional assistants consented to participate and at least three caregivers of students with ESN consented for their child to participate. One teacher (Teacher A) was in the 35- to 45-year-old range with 15 years of experience, two teachers (Teacher B and Teacher C) were in the 25- to 35-year-old age range with 5 to 7 years of teaching experience, and one teacher (Teacher D) was in the 45- to 55-year-old range with 25 years of experience. The student participants ranged from pre-kindergarten to fifth grade. Students at Grade 3 and higher all participated in alternate assessments based on alternate achievement standards as mandated by the state. Some participants in pre-kindergarten to Grade 2 received special education services under the category of developmental delay, which is not used in all states but could be used through age 7 in the state where this study was conducted. The students receiving special education services under the eligibility category of developmental delay in this study all had multiple disabilities impacting their cognition, motor, and communication skills. More information about student participants is provided in Table 1.
Student Participants in Shared Reading Study.
Note. Highest Beginning Communication Level: 1 = preintentional behaviors; 2 = intentional behaviors; 3 = unconventional communication; 4 = conventional communication; 5 = concrete symbols; 6 = abstract symbols; 7 = language.
Materials
Professional development for the shared reading approach included five sessions that focused on helping teachers learn how to support student engagement and interaction, engage in self-reflection, and use Tar Heel Shared Reader technology. These sessions included training content, active learning activities, and application assignments to be completed before the next training. The series of five modules, each requiring a total of 60 to 90 minutes, were developed to provide (a) an overview of the project; (b) a description of the specific approach to shared reading, Follow the CAR (c) an introduction to the Tar Heel Shared Reader technology; (d) an overview of the importance of self-reflection and an introduction to a self-reflection aligned with the Follow the CAR intervention; and (e) a description of factors to consider when selecting books for shared reading (see Appendix A for more information about module topics). These modules are just one part of the larger Tar Heel Shared Reader project, which also provides a variety of supports to improve professional practice in shared reading, including planning forms, self-reflection guides, and coaching resources. Tar Heel Shared Reader is intended to address the need for high-quality training for rural special educators to address gaps in literacy instruction (McCombes-Tolis & Feinn, 2008) and communication training (Pennington et al., 2009).
Measures
Teacher behaviors during shared reading instruction were assessed using a measure called the Shared Reading Self-Reflection and Observation (see Appendix C). This author-developed approach to observation reflects the content of the professional development modules, data collected during shared reading observations, and teacher feedback. The measure includes a list of 13 teacher behaviors that address areas such as book selection, communication access, commenting, pausing, and attributing meaning to student behaviors. Behaviors are marked as being present, absent, or not applicable (e.g., attributing meaning to communicative behaviors would not be expected if the students sat quietly).
The Communication Matrix (Rowland, 2004) was used to assess student expressive communication skills. This assessment, specifically developed for use with children with ESN, uses structured observations of early communicators’ behaviors for the purpose of building a communication profile. Across four communication purposes (i.e., refuse, obtain, social interaction, exchange information), the examiner records the presence of 24 behaviors (see Table 3 for a list of the behaviors). Each observed behavior is then described with reference to a level of communication, from preintentional behaviors (Level 1) to language expressed with two- to three-word combinations (Level 7). The resulting profile reflects the range of communication behaviors a child uses as well as the complexity of their communication across seven levels.
At the beginning and end of the study, teachers participated in a 10- to 15-min interview designed to elicit information about their understandings of and experiences in engaging students with ESN in shared reading (see Appendix B). The interview was semi-structured, with open-ended questions. Five primary topic areas were probed during the interview (a) understanding of shared reading, (b) practices during shared reading, (c) decisions about books for shared reading, (d) approaches to supporting students who are unable to use speech, and (e) challenges and successes with shared reading.
Procedures and Data Collection
Following approval from the university’s Institutional Review Board and the school district, four researchers with Ph.D.s who have experience teaching or providing speech-language services to students with ESN collected data from teachers and students. Student communication profiles were assessed with the Communication Matrix during 30- to 60-min structured classroom observations of each student over multiple instructional routines at the beginning and end of the school year. Teacher interviews took place in the participants’ classrooms while other classroom staff engaged students in instruction, free-choice activities, or outside play. The participants’ responses were written down by the interviewer, transcribing direct responses to the greatest extent possible. Teacher behaviors were assessed using the Shared Reading Self-Reflection and Observation Form using either videos or live observations. In the fall, videos were used for initial observations of teacher practice within 4 to 6 weeks of the first professional development session. In the spring, final observations of shared reading sessions were observed by the first author and another researcher approximately 1 month after the final professional development session.
Professional development for the shared reading implementation model was delivered in person by researchers over five sessions that occurred between September and March (see Appendix A), with three in the fall semester and two in the spring semester. These sessions lasted approximately 60 to 90 minutes. Classroom observations of shared reading were conducted by researchers approximately 1 time per month. During classroom observations, researchers took field notes (Gambold, 2010) recording information about teacher behaviors, student engagement, and student communication during the shared reading sessions and documenting interactions with teachers and staff.
Data Analysis
For the Shared Reading Self-Reflection and Observation Form, the percentage of observed behaviors was computed at the initial and final classroom observations over a 5-month period. Descriptive analysis was used to examine changes in teacher shared reading behaviors during classroom observations. In the spring, the Shared Reading Self-reflection and Observation Form was independently scored by two researchers present during live shared reading sessions. Afterward, videos from the fall were independently scored by the same two researchers during a single real-time viewing, to replicate live conditions used in the spring. Across all observed sessions, Cohen’s Κappa indicated strong agreement (Κ = .881, p = .000). All disagreements between coders were resolved with discussion.
For the Communication Matrix, frequency counts were calculated for the total number of behaviors observed out of a possible 24 across all communication purposes during 30- to 60-minute observations. Paired samples t-tests were used to analyze the change in the number of behaviors observed from the beginning to the end of the year for individual students, as well as the change in the number of students observed to demonstrate individual expressive communication skills. Next, the highest level of communication (i.e., 1, preintentional to 7, language use) was identified across all purposes for communication for each individual student.
Teacher interviews from the beginning and end of the year were analyzed using thematic analysis with an inductive, latent analysis approach (Braun & Clarke, 2006). The first author began with data familiarization and then parsed the data into information-rich quotations and reflections, which composed the data set for analysis. Using a “bottom-up” procedure, the content of reflections and quotes were coded using codes that were identified inductively from the data. Information with the same code or codes that appeared to relate was collapsed under the same theme. After multiple passes with the data, preliminary categories of codes and themes were modified, expanded, or collapsed. Ultimately three thematic categories were retained. These categories related to the overall research question about the impact of the shared reading model on teacher knowledge and skills.
Quantitative and qualitative data were first analyzed independently. Quantitative data regarding teacher shared reading behaviors and child communication skills were analyzed using descriptive statistics, paired sample t-tests, and Wilcoxon signed rank tests. Teacher interviews were analyzed using thematic analysis to examine changes in teacher self-perceptions of their practices, students, and understandings of shared reading. The data were not mixed until the data interpretation stage, with both forms of data being given equal status with regard to teacher perceptions of shared reading, teacher behaviors during shared reading, and student communication skills (Collins et al., 2006). The mixed data analytic strategy involved linking data and integrating results from analyses of the separate quantitative and qualitative data components (Bazeley, 2012; Onwuegbuzie & Corrigan, 2014). Specifically, quantitative analyses of teacher behaviors during shared reading and student expressive communication profiles were linked with findings from the qualitative analyses of teacher interviews. The focus was on teacher behaviors and student outcomes and their relationship to teacher perceptions at both the beginning and the end of the year. Themes and quotes from the interview data were extracted to illustrate the relationships between this data and the descriptive statistics from the teacher shared reading behaviors and child communication behaviors (Bazeley, 2012).
Results
Changes in Teacher Behaviors During Shared Reading
From the initial observation following the first professional development session in the fall to the final observation following the last professional development in the spring, teachers demonstrated an average increase from 50% to 90% of the targeted shared reading behaviors as listed on the Shared Reading Self-Reflection and Observation Form (see Table 2). Five behaviors increased from 0% to 100%. They included: (a) providing students in need of AAC access to a personal communication system, (b) preplanning comments, (c) using comments, (d) attributing meaning, and (e) repeating and adding more.
Teacher Behaviors During Shared Reading in Fall and Spring.
Note. AAC = Augmentative and alternative communication.
Change in Student Skills
A paired samples t-test indicated a statistically significant fall (51.9% ± 20.9%) to spring (66.0% ± 22.0%) increase in the percentage of all communication behaviors, across range and complexity, on the Communication Matrix for individual students, t(12) = 2.757, p = .017, d = .66. Similarly, a paired samples t-test indicated a statistically significant increase in the average number of communication skills demonstrated by students from fall (6.8 ± 4.4) to spring (8.6 ± 3.7), t(3.573), p = .002, d = 2.556. There was also an increase from fall to spring in the total number of students demonstrating 18 of the 24 behaviors assessed on the Communication Matrix, with an average increase of 2.78 students (range = 1 – 8) per behavior across the 18 behaviors. Students’ average highest level of communication for individual expressive skills increased from the unconventional level of communication in the fall (3.4 ± 2.0) to the conventional level of communication (4.0 ± 1.6) in the spring. A Wilcoxon signed rank test indicated that this gain in the highest level of communication observed across all purposes for communication was statistically significant, z = 2.16, p = .031, r = .60. See Table 3 for the total number of students demonstrating each behavior in the fall and spring and the average highest level of communication observed for each item on the Communication Matrix in fall and spring.
Number of Students and Average Highest Level of Communication for Each Item on the Communication Matrix.
Changes in Teacher Perceptions
Over the course of the year, there was a shift in the teachers’ perceptions of their role during shared reading. For example, in the fall, most teachers were focused on their role as reader. As Teacher C said, “I sit in front of them and read.” If teachers mentioned engaging students, it was through teacher-directed strategies such as asking questions or prompting. As Teacher A described, “We prompt the whole day because we think it’s what we have to do . . . We just want to have them respond quickly and so we prompt.” Although they recognized the communication challenges their students faced, teachers did not mention any role they had related to supporting student communication about the text or initiation of topics of interest. In fact, Teacher A explicitly stated that she didn’t work on communication and that it “is on the back burner to be honest.” By the spring, the teachers described their primary role as empowering, engaging, and scaffolding student engagement and communication while reading books. As Teacher A shared, “The focus is on getting students engaged, getting them to communicate, and honoring their communication and making sure each student has a way to communicate.” As teachers embraced the role of supporting student interaction, teacher-centered control was relinquished. For example, teachers reported shifting the balance of power by attending to and “reacting to their reactions” (Teacher D). As shared reading became less of something teachers did to students to more of something teachers did with students, teachers described their role as scaffolding student-led interactions. For example, teachers reported attributing meaning to the communication behaviors of presymbolic communicators. Teacher B explained,
One student started touching the [graphic symbols on the communication] board and looking for acknowledgment. He looks for me to make meaning . . . He realizes that when he selects [the graphic symbol] people are putting meaning to it.
Teacher C added, “I’m not reading at them. I’m reading to them. Now they’re talking about the book.”
Teacher perceptions about student communication potential and skill also changed. In the fall, only students with verbal, abstract language were considered to have the potential for successful interactions around books. Teacher C said, “The students who are verbal, you can interact well with books.” For students who do not yet use abstract language, Teacher A said that she “just look[ed] for engagement and attention.” As teachers embraced the role of supporting interaction, they began to see all students as communicators, including those who were beginning communicators and just learning to use AAC (i.e., communication boards). “I have been pleasantly surprised by the fact that they are so willing to use their communication boards” (Teacher B). Furthermore, all students were seen as capable of learning. Teacher A reported, “Kids are learning how to do it. Learning that they are communicating.” Teachers not only began to see their students as capable of responding, but also of initiating. In the fall, Teacher C explained, “I’ll read the book to myself several times . . . I’ll do that to figure out what this child is going to say about this.” By the spring, students were considered capable of communicating unique, unpredictable, and unscripted thought. “It’s always surprising about their interests and what they end up saying. It’s important to keep an open mind and realize they might want to talk about things you don’t expect” (Teacher B).
Teachers’ perceptions of student behaviors also changed over the course of the year. In the fall, teachers focused on behaviors as challenges, viewing them as disruptive or unproductive for shared reading. Teacher B stated that “behaviors are probably the biggest barrier.” Teachers reported shared reading was difficult when students were “not interested” (Teacher A) or were having difficulties with “alertness” (Teacher A) and “attention” (Teacher C). As teachers implemented a more responsive and interactive approach to shared reading, student engagement increased, and disruptive or inattentive behaviors decreased. In the spring, Teacher C reflected, “Now they’re actually looking at [books] instead of ripping them up.” She continued, another student used to have a “pure melt down every time we pulled her for reading” and now she asks to do reading, as “it has become one of her favorite things to do.” Furthermore, teachers began recognizing more behaviors as communicative and viewing them as on-ramps to engagement and learning. Teacher B said that she “put meaning to what they touch, even if it is accidental.” Similarly, Teacher A described, “You can turn any word [graphic symbol] a student says into a comment that relates to what is in the book.” Teacher A described this process as “honoring their communication” and letting them know “you are telling us something.” Other student behaviors, formerly seen as disruptive in a teacher-directed, shared reading context, were increasingly acknowledged and honored as bids for attention in an interactive context. “One student was hitting everything he could hit [on his communication board]. Once I acknowledged he was trying to communicate by responding to him, he slowed down and thought about what he wanted to touch and communicate” (Teacher B).
Teacher perspectives of barriers to successful shared reading also changed over time. In the fall, teachers primarily identified barriers that were intrinsic to students. “I thought about what they do during general education and didn’t think it would really work with my kids” (Teacher B). Disruptive behaviors, inattentiveness, or the inability to use speech to communicate were all viewed as student faults that inhibited quality shared reading interactions. After experiencing meaningful interactions with students, teachers began to identify barriers intrinsic to themselves. As Teacher A reflected, she believed that she had created a prompt dependency that decreased the likelihood of her students expressing their own ideas or understanding. She noted, “The students look at you to prompt them, like ‘This is what you’ve taught me’.” She was not alone in reporting her understanding that the more she exerted control, the fewer opportunities students had to learn, experiment, and grow.
Chaos is not a bad thing . . . I quickly learned that chaos is the first start to communication to a certain extent. And exploring is an important part of that process. I’ve prided myself in the fact that I encourage exploring in other areas, but I had never thought about it with communication and letting them touch everything [on the communication board]. (Teacher B)
Furthermore, Teacher B realized that one student’s “behaviors” were really bids for attention that increased because she had ignored them. Teacher B explained that the student “is realizing that she doesn’t have to touch everything [on the communication board] to get your attention. Once I give her attention, she slows down.” In different ways, all four teachers reflected on the ways that their own behaviors resulted in participation barriers for their students. Teacher B also realized that her low expectations led to low outcomes.
I’ve realized that even my weakest gross motor student is capable [of using a communication device] . . . In years to come, I’ll never second guess any of my students’ abilities to use a [communication] board. I thought I had high expectations.
Mixed Data Analysis
When linking data from teacher interviews and observations in the fall and spring, the shifts in teacher behaviors observed during shared reading were also reflected in their own descriptions of their practice. For example, at the beginning of the year, teachers reported that questions or prompting were required to elicit student participation. This was documented in high rates of physical prompting during observations in the fall. During interviews in the fall, three of the four educators failed to mention communication and the fourth explicitly stated that it was on the “back burner” during shared reading. This omission was also documented through observations of students needing AAC to communicate having no access to personal communication systems during shared reading in three of the four classrooms. Furthermore, teachers were not observed demonstrating comments at the student’s language level, attributing meaning to student communication efforts, or building communication skills by using repetitions or expansions. By the end of the year, all the teachers described their primary role as empowering and scaffolding student engagement and communication while reading books together. For example, teachers reported providing all students with access to their own AAC system and attributing meaning to nonsymbolic communication behaviors. Teachers also reported making comments to demonstrate ideas their students might want to say and using wait time to share the power of topic control with all communicators. These behaviors were observed during shared reading sessions conducted by all the teachers in the spring, when three out of four teachers also preplanned comments to support communication development.
Teacher perspectives regarding challenges in shared reading shifted from the fall to the spring and differences in patterns of their shared reading behaviors provide further evidence of this shift. In the fall, teachers reported that variables intrinsic to their students were the greatest challenges to shared reading. For example, their stated belief that verbal students were most likely to benefit from shared reading paired with no observed efforts to attribute meaning to student’s nonsymbolic communication efforts. Furthermore, teachers perceived many nonsymbolic communication efforts as negative or disruptive, which explains why they were never observed labeling, repeating, or expanding on nonsymbolic student communication. Nevertheless, by the end of the year, teachers reported high expectations for student participation and perceived student nonsymbolic behaviors as meaningful and communicative. This shift was accompanied by increased use of wait time to allow for student initiations, use of attributing meaning to honor communication efforts, and use of effective language intervention strategies like repeating and expanding upon student communication.
Changes in teacher perceptions about student potential and skill as communicators was accompanied by fall to spring changes in student scores of the Communication Matrix. The design of the study does not make it possible to know if changes in teacher shared reading behaviors were related to changes in student expressive communication skills, but the changes in teacher perceptions were clear, regardless of any relationship. In the fall, the only nonsymbolic communication behaviors described by teachers were those considered negative and disruptive. Average student levels on the Communication Matrix reflected unconventional, nonsymbolic communication, with the majority of students only showing a few of the beginning skills for each of the four purposes for communication measured. In the spring, teachers described seeing all their students as capable communicators, with a wide range of behaviors considered to be positive additions to the interaction, including protesting, directing adult attention, and commenting. This shift is similarly reflected in the 27% increase in the total number of expressive communication behaviors captured on the spring administration of the Communication Matrix. Teachers also described recognizing more nonsymbolic behaviors as communicative and reinforcing these behaviors because they considered them to be on-ramps to engagement, which then led to further developments in student communication. These developments were also noted in the students’ Communication Matrix profiles, as more students were observed using 18 out of 24 expressive communication skills, and there was an overall increase in the average of the highest level of communication observed across the 24 skills.
Discussion
This descriptive, multicase, mixed methods study explored changes in teacher perceptions, teacher shared reading behaviors, and student communication outcomes in a rural educational setting as teachers worked to implement a responsive and interactive approach to shared reading highlighted through a set of professional development modules. Although the design of the study does not support causal claims, it does offer a description of changes in teacher practices, teacher perspectives, and student outcomes that occurred over 6 months while teachers completed professional development focused on an interactive approach to shared reading. Many factors may have supported improved student communication outcomes, including: (a) teacher training in interactive shared reading practices, AAC demonstration, and book selection; (b) increased teacher use of targeted interactive shared reading practices; (c) increased use of texts that were at or slightly above the students’ language level, or (d) increased teacher demonstration of AAC combined with increased student access and use of AAC. It is worth noting that the teachers attributed the changes in their practice and their students’ communication skills to the professional development they received. In the interviews, teachers reported changes in their shared reading interactions that reflect changes documented in the shared reading observations and quantitative gains in student communication.
One exciting finding in this study is that this group of students with ESN who, by definition, are beginning communicators, demonstrated significant gains in their communication skills, with a medium to large effect size. While meaningful growth among students with any disabilities is important, it is particularly exciting to find it among a group of students who are often reported to show little growth in communication skills across their educational career (Erickson & Geist, 2016). Therefore, even small effect sizes can be considered relevant, meaningful, and useful with this population (Dunst & Hamby, 2012), especially since the gains occurred on a generalized measure of communication (i.e., the Communication Matrix) rather than measures directly tied to behaviors taught in the intervention.
Over the year, teachers’ instructional practice shifted from more teacher-directed and controlled to more student-centered and interactive. In the fall, teachers sought to engage students by prompting responses to known-answer questions, which has been reported to be one of the most, if not the most, used instructional strategy in classrooms serving students with ESN (Ruppar, 2017; Ruppar et al., 2011). When teachers directed the interaction with questions, they reported having to focus more on student compliance due to high levels of inattentive and disruptive behaviors. As teachers focused on encouraging student-initiated contributions, responding to them, and helping them connect meaningfully to the text, they reported that students became more attentive, interested, and engaged. This inverse relationship between teacher control and student engagement was previously reported among teachers of students with ESN (Bock & Erickson, 2015). In the current study, teachers became more student-centered, inviting participation and supporting communication. This responsive and interactive approach has reportedly been associated with more frequent and expansive opportunities for students to explore, use, and develop language and communication skills (Bock & Erickson, 2015), similar to what has been observed among students with autism (Girolametto et al., 2007) and less severe disabilities during play (Girolametto, 1988; Girolametto & Weitzman, 2002).
Teacher expectations of their students increased over the year, as students made gains in their communication skills. It is not possible to determine whether teacher expectations or student performance influenced change in each other, but others have suggested a link between teacher attributions and student achievement (Jorgensen et al., 2007). It is the case that misjudging the cognitive and communication potential of students with ESN can lead teachers to prohibit access to or restrict the learning opportunities that are afforded to students with severe disabilities (Ruppar, 2017; Ruppar et al., 2015). Similar to what has been reported elsewhere (Ruppar, 2017), teachers in the current study initially expected participation only from those who used spoken, abstract language. As a result, students without those skills were infrequently called upon and were not provided with access to the AAC systems they needed to initiate or respond. Over time, all students had access to personal AAC systems, and teachers began demonstrating the use of the AAC systems and seeing all their students as capable communicators. Furthermore, teachers expressed increased self-efficacy in supporting the participation and communication of all their students, including those who did not use speech to communicate. Consistent with Soto (1997), teachers’ beliefs that they could teach their students to communicate more effectively co-occurred with higher expectations of student achievement and beliefs that their students were capable of learning.
As teachers’ perceptions of their students’ communicative competency changed, so did their interpretation and perception of student behaviors. For example, some student behaviors initially described as disruptive were eventually ascribed communicative value and honored. Rather than interpreting behaviors as nonconformist and dysfunctional, teachers came to view them as meaningful and responded to them in supportive ways that offered symbolic alternatives. These affirming responses coincided with higher levels of varied, intentional student communication. As a result, student participants began experiencing membership in a literate community that acknowledged, valued, and expected their meaningful participation (Kliewer, 1998).
Implications
This study documents changes in teacher practice and perspective as well as changes in student communication that occurred in classrooms serving children with ESN in rural settings as teachers completed professional development focused on an interactive and responsive approach to shared reading featuring Following the CAR. Now that these resources are available in an open-source, online, on-demand format, they are particularly well-suited to address the needs of rural special education teachers who often face issues of isolation, reduced self-efficacy, and limited self-confidence (Abell et al., 2014; Rude & Miller, 2018). The teachers in this study reported increased confidence in their ability to use shared reading to support their students’ communication development, which is a specifically recognized need among rural special education teachers (Pennington et al., 2009). By increasing self-efficacy in shared reading practice, teachers may have higher expectations of their students and provide more literacy learning opportunities (Soto, 1997). This study suggests that students with ESN, including those who do not actively participate in shared reading in the beginning, may benefit from a responsive and interactive approach to shared reading. This approach may be one way to increase engagement, initiations, and complexity of symbolic communication while decreasing disruptive behaviors of students with ESN, allowing teachers to focus on engagement and interaction while supporting student ability to make meaning, connect with, and communicate about texts.
Limitations and Future Research
Given the design of the study and the small sample size, caution must be used in interpreting the results or attributing the teacher and student outcomes solely to the professional development teachers completed. It does, however, suggest that teachers may be able to acquire the skills needed to implement the responsive and interactive approach to shared reading by completing the professional development modules, but more evidence causally linking these modules to teacher outcomes is warranted. Further, more evidence is needed to make causal claims regarding the responsive and interactive approaches to shared reading in this study and improved student communication outcomes. Future research should also consider examining the effects of teacher behaviors on the communication behaviors of students during shared reading.
Another limitation of the current study is the fact that teachers and students were studied in separate special education settings. Although more than 90% of students with ESN are educated in similar settings across the United States (Brock, 2018; Erickson & Geist, 2016; Morningstar et al., 2017), these settings are associated with decreased access to literacy instruction (Ruppar et al., 2018) and reduced academic outcomes compared with general education settings (Shogren et al., 2015). Future studies should examine the benefits of shared reading for students with ESN in inclusive settings in small groups that include their peers without disabilities.
Conclusion
This study contributes important information to the small body of literature about shared reading with students with ESN in rural, separate special education settings. The results suggest that professional development targeting a responsive and interactive approach may promote teacher behaviors that support student communication skills and result in increases in the amount, variety, and complexity of communication exhibited by students with ESN.
Footnotes
Appendix A
Appendix B
Appendix C
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
This document was produced in part under U.S. Department of Education, Office of Special Education Programs Grant No. H327S160005. The views expressed herein do not necessarily represent the positions or policies of the Department of Education. No official endorsement by the U.S. Department of Education of any product, commodity, service, or enterprise mentioned in this publication is intended or should be inferred. Project Officer, Susan Weigert.
