Abstract
This systematic literature review updates and extends the findings of Hudson and Test’s 2011 review of literature on shared reading, an interactive read-aloud practice, to promote literacy skills for students with extensive support needs (ESN). Thirty-two shared reading studies are assessed and described in terms of literacy skills investigated, intervention methods, materials used, interventionists, and settings in which shared reading has been evaluated for students with ESN. The quality of each included study was analyzed and the assessment of level of research evidence was conducted using systematic processes. Results indicate that there is a moderate level of evidence for shared reading as an intervention to promote comprehension, vocabulary, and emergent literacy skills for students with ESN. Implications for practice and areas of need for future research are presented.
Access to literacy instruction is a human right and essential skill within all aspects of modern society (Keefe & Copeland, 2011). Literacy, which includes listening, speaking, and interacting as well as reading, writing, and spelling enhances and creates opportunities for students to interact in their home, community, and school environments, as well as to engage within established adult roles, such as employment (Ruppar et al., 2017). Students with extensive support needs (ESN) are 1% of students with low-incidence disabilities (i.e., intellectual disability, multiple disabilities, autism spectrum disorders, and deaf-blindness) who qualify to take their state’s alternate assessment and/or have extensive needs for support across domains such as communication, activities of daily living, and behavior (Kennedy, 2004; Thompson et al., 2018). These students have historically faced barriers to literacy instruction and access to equitable educational services in the general education setting (Morningstar et al., 2017). Fortunately, recent advances in learning theory and legislation have promoted an increase in research on literacy instruction strategies for this population (Toews & Kurth, 2019). This systematic literature review on shared reading for students with ESN replicates and advances previous work done by Hudson and Test (2011). This investigation identifies the strength of evidence supporting the use of shared reading as an intervention to support specific literacy skill development for students with ESN and outlines common characteristics of the intervention.
Shared Reading
Shared reading is a text-based interaction between a reader and listener in which the reader pauses intermittently to engage the listener with the book through discussion, question and response, or physical interaction (Hudson & Test, 2011). Within the published research, shared reading has been referred to by many names, including read-aloud (Shurr & Taber-Doughty, 2017), dialogic reading (Whitehurst et al., 1999), interactive reading (Fisher et al., 2004), storybook lessons (Nielsen & Friesen, 2012), story-based lesson (Spooner et al., 2009), and shared storybook reading (Golloher, 2018). Despite diversity in names, all shared reading interventions involve reading a text aloud and providing support for the student to interact with both the text and the reader (Hudson & Test, 2011). Shared reading is a research-based literacy strategy for typically developing children at home and in school (Fisher et al., 2004), students who have a high-incidence disability such as a learning disability or specific language impairment (Justice et al., 2005), and students with ESN (Hudson & Test, 2011).
The interactive nature of the read-aloud between the reader and listener is a central characteristic of shared reading. Hudson and Test (2011) identified several strategies for implementing shared reading for students with ESN, including adapted age-appropriate books with simplified text, increased picture supports, repetitive text lines (Spooner et al., 2009), use of objects to support comprehension (Browder et al., 2011), use of varied response modes including augmentative and alternative communication (AAC) devices (Koppenhaver, Erickson, & Skotko, 2001), integration of systematic prompting using time delay and system of least prompts (Mims et al., 2009), and use of a task analysis to support systematic, planned interaction with the student during the shared reading activity (Browder et al., 2007). Hudson and Test (2011) found sufficient evidence to identify shared reading as a practice with moderate research support and noted additional research was needed to identify the practice as having strong research support.
A Need for a Current Review of Shared Reading Research
Since the 2011 literature review, new independent research teams have investigated shared reading for students with ESN, providing a rationale for the current literature review (e.g., Finke et al., 2017; Golloher, 2018; Rivera et al., 2017; Ruppar et al., 2017). The goal of educational research on instructional interventions is to evaluate and disseminate information on effective practices that can then translate to teaching practice in schools. The location of intervention, materials used, and characteristics of the interventionist are some of the key factors necessary to understand, replicate, and incorporate research-based instructional practices.
In their 2011 review, Hudson and Test identified limitations of then-extant shared reading studies, including the limited number of studies and research teams, and that most studies of shared reading occurred in homes or separate special education classrooms, leaving it unclear if these practices can be considered evidence-based in inclusive settings. However, new research teams have investigated shared reading with students with ESN in general education classrooms in one-to-one student to teacher groupings (Roberts & Leko, 2013; Ruppar et al., 2017) and embedded within whole group instruction (Courtade et al., 2013). Considering the ongoing analyses of shared reading, a synthesis of this research is needed to understand what literacy skills shared reading interventions support, intervention methods, materials, interventionists, and settings in which shared reading can be considered a research-based practice for students with ESN. This need is addressed in the current review through the following research questions:
Method
Search Procedure
The following section provides a description of the literature search and inclusion review process, which follows the Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines (Figure 1; Moher et al., 2009). A literature search was conducted within three search platforms, including ProQuest, EBSCOhost, and PubMed, from the University of Kansas library system. Search terms were selected by replicating the search conducted by Hudson and Test (2011) and adding relevant terms that have emerged for shared reading since 2011. Each search was limited to peer-reviewed studies or dissertations written in English published between 1975 and February 2020.
Search procedure PRISMA flowchart.
Search terms for participants with ESN and terms related to shared reading were used in combination, including: “severe disabilit*,” “intellectual disabilit*,” “significant disabilit*,” “multiple disabilit*,” “moderate disabilit*,” “cognitive disabilit*,” “autism,” “special education,” and “support needs.” The following terms for shared reading were used: “shared reading,” “shared story,” “story reading,” “literacy based lesson,” “literacy-based lesson,” “interactive read*,” “read aloud,” “storybook reading,” “story based lesson,” and “story-based lesson”.
A total of 890 results were collected across all searches. Results from each of the nine searches were downloaded and combined in a single Microsoft Excel document. There were 502 unique studies after duplicates were removed. A hand search of titles and abstracts was conducted in eight top journals in the field of special education (see Figure 1) for dates represented in the original search to ensure no relevant articles were missed in the database search. Five additional studies were identified for review from the hand search.
Inclusion Criteria
All studies met the following inclusion criteria: (a) peer-reviewed article or dissertation; (b) published in English between 1975 and February 2020; (c) experimental or quasi-experimental study (randomized controlled trials, single-case methods, mixed methods) conducted in the United States; (d) at least one participant with ESN, (e) participants in grades K-12; (f) shared reading is the independent variable; and (g) the dependent variable is at least one component of literacy as defined by the National Reading Panel (i.e., phonemic awareness, phonics, fluency, vocabulary, or comprehension; National Institute of Child Health and Human Development, 2000) or emergent literacy skills (i.e., interaction or engagement with text; Justice & Pullen, 2003).
Screening Procedure
A two-step process was used to ensure all studies met the inclusion criteria: (a) a title and abstract screening followed by (b) full-text screening. The abstract and titles of all 507 studies were reviewed by the first author, with the second author reviewing 138 (27%). Interrater agreement for the title and abstract was calculated at 96.4% based on point by point agreement. The first and second authors met to come to consensus on all disagreements prior to proceeding to the next step of the screening process. Forty-three studies were identified for full-text screening after consensus was met. During full-text screening, the first two authors reviewed study design, participants, intervention, materials, and results. The second author screened 13 (30%) studies, with interrater agreement calculated using point by point agreement at 99.4%. After full-text screening, 11 articles were excluded, resulting in 32 studies included in the final review. Reasons for excluding these 11 articles included, participants did not have ESN, the dependent variable was not a literacy skill (e.g., social skill), the study was not an experimental design (e.g., an A-B design), shared reading was part of a multicomponent intervention package, or the article was a duplicate or review of an article that is included in the review.
Quality Indicator Coding and Level of Evidence
The quality of studies included in this review was assessed using a 20-item quality indicator checklist based on the quality indicators recommended by Horner et al. (2005). A dichotomous, Yes (indicator met) or No (indicator not met), rating was used to assess each of the 20 items on the quality indicator checklist. Interrater agreement for quality indicator coding was completed for 30% of included studies and calculated at 94.2%. Each article was evaluated to identify if a functional relation between the intervention and the predicted effect was achieved using the visual analysis process described by Cooper et al. (2007). There exist emerging parametric (e.g., regression or multilevel modeling) and nonparametric (e.g., percentage of nonoverlapping data, percentage of all nonoverlapping data) methods for calculating effect-size within meta-analyses of single-case research designs that may advance possibilities for using such analyses to identify evidence-based practices from single-case research design studies (Shadish et al., 2008). As these statistical analyses continue to be adapted and refined, visual analysis remains the dominant and most widely accepted approach to the analysis of single-case design data (Horner & Kratochwill, 2012). It is for this reason that this review did not utilize any measures of effect-size within our analysis of the evidence base for shared reading.
The decision rules for determining levels of evidence as they appear in Hudson and Test (2011) with the addition of the “5-3-20” rule were used to identify the level of evidence supporting shared reading as a literacy intervention for students with ESN (Horner & Kratochwill, 2012; Test et al., 2009). These decision rules include four requirements to establish a strong base of support for a practice: (a) identification of five high-quality studies that demonstrate a functional relation, (b) conducted by at least three independent research teams, (c) at least 20 participants across qualifying studies, and (d) an absence of contradictory evidence on the effect of the intervention (Horner & Kratochwill, 2012; Test et al., 2009).
Results
RQ1: Common Characteristics of the Research Design and Intervention
Pertinent characteristics of all studies were recorded including study design, interventionist, participant characteristics, intervention location, and ratio. A detailed description of these characteristics is found within the Study characteristics subheading and summarized in Table 1. Details about shared reading intervention characteristics were also recorded, including types of systematic instruction used, accessibility features, book type, and characteristics. A detailed description of the intervention used in each study is found within the Shared reading intervention characteristics subheading and summarized in Table 2.
Shared Reading Study Characteristics.
Shared Reading Intervention Characteristics.
Note. UDL = Universal Design for Learning; (+) = functional relation; (/) = positive effect; (0) = neutral/no effect; (−) = negative effect.
Study characteristics
Design
All studies were single-case designs. No group design studies met all inclusion criteria. Several single-case designs were utilized, including multiple baseline across participants (n = 8); multiple probe across participants (n = 15); multiple probe across materials, conditions, or skills (n = 6); multiple baseline across materials, conditions, or skills (n = 2); and withdrawal (n = 1). The following sections on study design are summarized in Table 1.
Interventionist and intervention ratio
A university researcher implemented all or part of the shared reading intervention for students in 19 (59%) of the reviewed studies, making a university researcher the most common interventionist (e.g., Alison et al., 2017; Kim et al., 2018; Mims et al., 2012). As seen in Table 1, two studies (i.e., Rivera et al., 2017; Wood et al., 2019) included both the researcher and a special education teacher as the interventionist across different phases of the study and one study involved a researcher and parents as interventionists (i.e., Golloher, 2018). Within three studies, a university researcher delivered part of the shared reading intervention, with peers involved in the intervention delivery in at least one phase of the study (i.e., Hudson & Browder, 2014; Shurr, 2012; Shurr & Taber-Doughty, 2017). Special education teachers were the sole interventionist in seven studies (e.g., Browder et al., 2007; Mucchetti, 2013; Ruppar et al., 2017). Two studies (i.e., Courtade et al., 2013; Wood et al., 2015) involved general and special education teachers in the development and/or implementation of the shared reading intervention, and one study (i.e., Roberts & Leko, 2013) involved both special education teachers and paraeducators as interventionists. No studies were identified in which the only interventionist was a general education teacher. A paraeducator was the sole implementer in one study (i.e., Spooner et al., 2009). Parents implemented two shared reading studies with their own children (i.e., Koppenhaver, Erickson, McLellan, et al., 2001; Koppenhaver, Erickson, & Skotko, 2001). The most common intervention ratio was 1:1 (n = 26; e.g., Browder et al., 2008; Spooner et al., 2014; Wood et al., 2019). Only six studies included more than one participant at a time (e.g., Browder et al., 2007; Courtade et al., 2013; Shurr & Kromer, 2018).
Location
Special education classrooms were the most common location for shared reading (n = 12; e.g., Mims, 2009; Rivera et al., 2017; Roberts et al., 2019). Three studies were conducted both in special education classrooms and other school areas, such as an office or hallway (i.e., Mims et al., 2009; Shurr & Kromer, 2018; Spooner et al., 2014). Four studies implemented shared reading in the participant’s school building (i.e., cafeteria, office, hallway) but not in a classroom (e.g., Mims et al., 2012; Shurr, 2012; Spooner et al., 2015). Five studies occurred in both special and general education classrooms, (e.g., Hudson & Browder, 2014; Kemp-Inman, 2016; Roberts & Leko, 2013). Courtade et al. (2013) and Ruppar et al. (2017) conducted the only studies that were implemented entirely in general education classrooms. One study was implemented in a clinic (i.e., Kim et al., 2018), and Edmister and Wegner (2015) deliver shared reading in both home and clinic settings. Three studies occurred in participant homes or daycare facilities (i.e., Golloher, 2018; Koppenhaver, Erickson, McLellan, et al., 2001; Koppenhaver, Erickson, & Skotko, 2001).
Participants
The average number of student participants was 3.41, with a range of 1–6. The total number of student participants across all studies was 109, with ages ranging from 3 to 21. Twenty-one of the 32 included studies reported the race or ethnicity of participants. This resulted in 35 participants (32%) of an unknown race or ethnicity. Of the participants for whom race or ethnicity was reported, 46 were White (42%), 13 were Latino/a or Hispanic (12%), 9 were Black (8%), 4 were multi-racial (4%), and 2 were Asian (2%). Elementary-aged students represent 65.6% of participants (n = 21; e.g., Cheek, 2016; Spooner et al., 2009; Wood et al., 2015), with 12.5% in middle school settings (n = 4; e.g., Browder et al., 2007; Mims et al., 2012; Shurr & Taber-Doughty, 2013), and 18.6% in high school settings (n = 6; e.g., Knight et al., 2018; Roberts et al., 2019; Shurr, 2012). The study conducted by Finke et al. (2017) did not indicate whether participants were considered elementary or middle school students.
Information on the disabilities of participants was found in all studies except Wood et al. (2019) who indicated all participants attended a self-contained special education class for students who qualify to take their state’s alternate assessment. Participant disability labels across the 32 studies were as follows: autism (n = 38, 34.9%), intellectual disability (n = 31, 28.4%), intellectual disability and multiple disabilities (n = 19, 17.4%), autism and intellectual disability (n = 10, 9.2%), other health impairment (n = 4, 3.7%), unspecified (n = 3, 2.8%), multiple disabilities (n = 2, 1.8%), and developmental delay (n = 2, 1.8%).
Shared reading intervention characteristics
The following sections describe some of the most common features of shared reading that were identified through the review of the literature. Table 2 provides a visual summary of the shared reading intervention characteristics described in the following sections.
Systematic instruction
Systematic instruction in the form of a system of least prompts or time delay was used in 20 studies. Nineteen studies implemented a system of least prompts. Each research team modified the system of least prompts, some providing multiple types of verbal or gestural cues before moving to more intrusive prompts. For example, Finke et al. (2017), described a five-step least to most prompting hierarchy. Other studies utilized a modified system of least prompts in which students were provided with a re-reading of shorter and shorter portions of text to support their identification of the answer to a comprehension question (Alison et al., 2017). Some research teams utilized a traditional least to most prompting hierarchy by providing the natural cue before supporting with a gesture then verbal, then model and finally physical prompt to participants (e.g., Golloher, 2018; Roberts et al., 2019).
Time delay strategies were commonly used in shared reading (n = 12). Ruppar et al. (2017) used constant time delay prompting to support a student with ESN to complete 12 vocabulary, engagement, and comprehension tasks during each shared reading session. In this study, the interventionist delivered the cue, waited up to 7 s for the student to respond, then provided the controlling prompt or praise. Other studies used time delay to introduce a system of least prompts, described in the previous section (e.g., Mims et al., 2009; Roberts et al., 2019; Wood et al., 2015). For example, the interventionist in Roberts et al. (2019), provided 5 s after the initial comprehension cue for the student with ESN to respond, before providing increased level of prompting until the student responded correctly.
Book type
The most common types of books used in shared reading were fiction books such as storybooks (e.g., Browder et al., 2011; Mims, 2009; Spooner et al., 2015) and novels (e.g., Hudson & Browder, 2014; Ruppar et al., 2017). Twenty studies used only fictional books (e.g., Alison et al., 2017; Edmister & Wegner, 2015; Mucchetti, 2013). Nine studies used only nonfiction books, such as biographies (e.g., Mims et al., 2012), textbooks (e.g., Wood et al., 2015), and newspapers (e.g., Shurr & Taber-Doughty, 2017). Three studies used both fiction and nonfiction books (i.e., Courtade et al., 2013; Roberts & Leko, 2013; Shurr & Kromer, 2018).
Materials
The following sections describe shared reading materials and their frequency including adaptations, nonadapted texts, task analyses, and technology supports.
Adaptations
Adaptations were frequently added to or available outside books to increase accessibility during shared reading (n = 29). Addition of picture or video supports to books, graphic organizers, prereading materials, or answer options were the most common adaptations (n = 27). One of these 27 studies provided both picture and video to support vocabulary acquisition during shared reading of science books (Rivera et al., 2017). The text within books was rewritten in a simplified form or reduced in 18 studies to make content and vocabulary more accessible to students (e.g., Alison et al., 2017; Mims et al., 2009; Roberts & Leko, 2013). Object supports are any physical items that are presented along with books to represent targeted content. Object supports were used in five studies (e.g., Browder et al., 2018, 2011; Golloher, 2018). A repetitive text line is text that appears on multiple pages of a book and often includes a key element of the story (Courtade et al., 2013). Twelve studies utilized books with repetitive text lines originally included or added (e.g., Browder et al., 2011; Rivera et al., 2017; Spooner et al., 2009).
Nonadapted texts
The use of nonadapted age-appropriate books in shared reading has been explored in nine studies. Picture Plus Discussion (PPD) within shared reading was the independent variable of choice in four studies (i.e., Shurr, 2012; Shurr & Kromer, 2018; Shurr & Taber-Doughty, 2013, 2017). PPD provides participants with a pre- and post-reading discussion of book content supported by a book-specific picture strip. The texts used were nonmodified short (200 words or less) passages.
Task analysis
Task analyses were common supports for systematic instruction within shared reading (n = 19). Task analyses used in the included studies outlined specific steps interventionists were intended to provide throughout each shared reading session (e.g., Browder et al., 2007; Courtade et al., 2013; Roberts & Leko, 2013). Some task analyses included the exact wording of the intervention to decrease variability in implementation across sessions (e.g., Edmister & Wegner, 2015; Ruppar et al., 2017; Spooner et al., 2009).
Technology
Technology, including iPads and speech-generating AAC devices were used in shared reading (n = 11); for example, Knight et al., 2018; Spooner et al., 2015; Wood et al., 2019). Some research teams used iPads as the means for presenting an adapted storybook, engagement, vocabulary, and question prompts (n = 8). Typically, AAC devices were preprogrammed with response options, a repetitive text line button, and key story elements each student could interact with throughout the shared reading intervention. Although iPads were the most commonly used technology investigated in shared reading, other technology, such as DynaVox, Cheap Talk, and BIGmack AAC devices were used to support students to engage in shared reading (e.g., Edmister & Wegner, 2015; Koppenhaver, Erickson, McLellan, et al., 2001; Koppenhaver, Erickson, & Skotko, 2001).
RQ2: Literacy Skills Supported by Shared Reading
The National Reading Panel identified five critical components of reading instruction: phonics, phonemic awareness, fluency, vocabulary, and reading comprehension (National Institute of Child Health and Human Development, 2000). Inclusion criteria for the review required that studies have a dependent variable that measured one of the five components of literacy identified by the National Reading Panel or emergent literacy skills, such as engaging with text. Table 2 displays the components of literacy instruction targeted for instruction in each included study and the impact of the shared reading intervention on each targeted literacy component. Studies frequently (n = 15) measured multiple components of literacy with a range of two to three components. Seventeen studies measured only one literacy component.
Comprehension was the most commonly targeted literacy skill in shared reading intervention studies (n = 24). A functional relation between shared reading and an increase in student comprehension was present in 22 studies. The remaining two studies showed a generally positive effect on comprehension outcomes. Comprehension was most commonly measured by student response to “wh” questions directly related to the text or through story retell. Emergent literacy skills were supported by shared reading in 18 studies.
Emergent literacy included conversational turn-taking using AAC devices during shared reading (Edmister & Wegner, 2015; Finke et al., 2017), pointing to the title or turning pages in the shared reading book (Spooner et al., 2015), and engagement in the shared reading book as measured by eye gaze (Browder et al., 2011). A functional relation between a shared reading intervention and emergent literacy skill improvement was established in 13 studies. Three studies showed a generally positive effect and two studies showed a neutral effect.
Vocabulary knowledge was the dependent variable in six studies, with a functional relation or positive effect established between shared reading and vocabulary acquisition in all studies. The remaining two studies showed a generally positive effect. Browder et al. (2007) were the only research team to measure phonics skills as an outcome of a shared reading intervention. This study established a functional relation between the intervention and improved phonics skills. Shared reading was not used in any studies to teach phonemic awareness or fluency skills. Finally, no study reported negative effects of shared reading on any literacy skill.
RQ3: Level of Evidence Analysis
Table 3 displays the results of quality indicator coding for all studies. Three studies met all 20 quality indicators defined by Hudson and Test (2011) and were identified as high-quality: Cheek (2016), Knight et al. (2018), and Wood et al. (2015). These three high-quality studies were conducted by three independent research teams with a total of 10 participants; importantly, no negative effects of shared reading were identified in the 32 included studies. Yet, the benchmark of five high-quality studies with at least 20 participants was not met to establish a strong base of support. Therefore, shared reading remains a practice with moderate evidence to promote comprehension (e.g., Knight et al., 2018; Mims et al., 2012; Wood et al., 2015), vocabulary (e.g., Alison et al., 2017; Koppenhaver, Erickson, McLellan, et al., 2001; Ruppar et al., 2017), and emergent literacy skills (e.g., Cheek, 2016; Golloher, 2018; Spooner et al., 2015).
Quality Indicator Ratings.
Note. DV = dependent variable; N = no; Y = yes.
The most common reason a study did not meet the criteria of high quality was indicator 19, which requires an explicit statement of how the intervention is practical and cost-effective. It was uncommon for a study to provide information on the cost of materials for implementation. The second most commonly missed indicator was 20, which requires the study to occur in a natural setting with a typical intervention agent (e.g., parent, teacher, paraeducator, or peer). Nineteen of the 32 shared reading interventions in this sample were implemented in part or whole by a researcher. These studies were not eligible to be considered high quality. A study was considered of acceptable quality if all quality indicators were met except indicators 2 and 11 and at least one of the social validity indicators (17–20) were met. Twenty-eight studies were of acceptable quality. Due to having only two baseline data points for one participant, Roberts and Leko (2013) did not meet the requirements of acceptable quality, as this is a threat to internal validity (Indicator 15).
Discussion
This systematic literature review utilized study characteristics and quality indicator coding to allow for a deeper understanding of shared reading interventions, identification of the literacy skills that are supported, and a description of the strength of the evidence base for shared reading for students with ESN. The number of studies that investigated shared reading has risen substantially since the last literature review in 2011, which contained six studies (Hudson & Test, 2011). However, due to a lack of reporting on cost feasibility and the presence of a researcher as the interventionist, there continues to be only a moderate base of evidence for the use of shared reading to support text comprehension, vocabulary, and emergent literacy skill acquisition for students with ESN. Although there were not enough high-quality studies or participants in those high-quality studies to designate shared reading as a practice with strong research support, all 32 studies showed a positive effect of shared reading on at least one literacy skill.
Shared reading interventions were most commonly implemented by a researcher or special education teacher within special education settings in a one-to-one ratio. Although the intervention was most often investigated in elementary settings, 12 studies implemented shared reading to support student literacy in middle and high school settings. Systematic instructional practices were an integral part of shared reading interventions. A system of least prompts and constant time-delay were the most common systematic instructional practices embedded within shared reading. Shared reading was shown to be effective in supporting student comprehension of fiction and nonfiction texts. The use of adapted texts with simplified wording, added pictures, repetitive text lines, and task analyses were common within shared reading. There is also a growing research base demonstrating shared reading can be used to support students with ESN to access nonadapted grade-level texts. The use of technology to present e-books is also a strategy for shared reading implementation with growing support.
Limitations
There are limitations that should be considered when interpreting the results of this review. First, there are many terms used to reference shared reading. Although the most commonly used terms were incorporated in the search, it is possible that studies using other terms for shared reading were overlooked. In addition, five shared reading studies were located during the hand search process. These studies referenced at least one of the shared reading search terms explicitly. This indicates the databases used may not have access to all relevant journals. Another potential limitation is the quality indicators used. There are multiple tools available for assessing the quality and rigor in single-case design research, and the results of the assessment can vary based on the tool used (Zimmerman et al., 2018). These results must be interpreted with the understanding that they may be different if study quality had been assessed with different tools such as the What Works Clearinghouse or Single-Case Analysis and Design Framework (Zimmerman et al., 2018). However, because no consensus has been reached as to which quality indicators should be used, the Horner et al. (2005) indicators are considered valid.
Implications for Research
The findings from this review indicate teachers can integrate shared reading with adapted or nonadapted texts, systematic prompting, and planned engagement points into the comprehensive literacy instructional programs of students with ESN. There are three key areas where more investigation is needed. These areas include the following: (a) investigation of shared reading to support students with ESN to access grade-level texts within general education settings, (b) integration of individualized shared reading supports into small and whole group instruction, and (c) research on shared reading that is implemented by school staff or peers rather than researchers. In addition to these implications, it is important that future research on shared reading includes explicit mention of the cost of interventions to facilitate a deeper understanding of the resources necessary for implementation.
National legislation indicates all students should access education in the least restrictive environment (Individuals with Disabilities Education Act [IDEA], 2004) and receive the support necessary to make progress toward grade-level standards (Every Student Succeeds Act [ESSA], 2015). In addition, students with ESN have been shown to make higher levels of progress in literacy skill development (de Graaf & van Hove, 2015; Dessemontet et al., 2012) and are 10 times more likely to receive literacy instruction when included in general education classrooms (Ruppar et al., 2018). For these reasons, research on evidence-based practices for students with ESN that are validated in special education settings should be investigated for effective use or adaptation in general education settings (Toews & Kurth, 2019).
In the nearly 20 years of research and 32 reviewed research studies, only two studies (Courtade et al., 2013; Ruppar et al., 2017) were conducted entirely in general education settings. This is problematic because a lack of research in general education settings leaves the field without an example of how students with ESN can be supported in fully inclusive instruction. Future research should focus on measuring literacy skill outcomes of shared reading that is delivered entirely in general education settings. In addition, future research should investigate the use of typical grade-level texts, including the processes used for identifying the type and intensity of adaptation or external supports necessary for students to access these texts in the general education setting. The emerging base of research on the use of shared reading using nonadapted books (e.g., Shurr & Taber-Doughty, 2017; Wood et al., 2015) should be expanded to investigate the use of grade-level fiction and nonfiction texts within general education settings. The use of nonadapted texts with external image, graphic organizer, or peer supports has the potential to save valuable teacher time in the adaptation process. In addition, the use of technology to implement shared reading virtually could be explored. Virtual delivery of shared reading could improve student access to literacy instruction in situations where distance learning is necessary and during independent learning time in classrooms.
Whole-class instructional groupings are common in general education classrooms. For example, Toews et al. (2020) observed general education classrooms that include students with ESN and found teachers lectured or led whole group discussion in 37.5% of observation intervals within elementary and middle school academic instruction. Thirty out of 32 studies in this review utilized a one-to-one student to interventionist ratio. Consequently, research using shared reading in general education settings in typical student grouping arrangements (e.g., small and whole group) is needed to establish validity in these arrangements.
Additional areas for research include embedding individualized supports and systematic instruction into shared reading within small or whole group book reading in general education classrooms. Wood et al. (2015) investigated the use of shared reading of nonadapted grade-level social studies books with a graphic organizer to support the generation of and response to comprehension questions about the book within special education classrooms. Generalization probes were conducted in general education classrooms in which small groups of peers, including a student with ESN, worked together to complete the same shared reading intervention. Additional research that looks more closely at peer-implemented shared reading supports within-group assignments in subjects such as science and social studies would also benefit the field.
The use of natural interventionists, such as teachers, paraeducators, related service providers, and peers within educational research is important to the generalizability of interventions to natural school settings. Nineteen of 32 studies in this review were implemented by researchers. Future research should support teachers and other natural supports in general education settings to implement shared reading. The current review identified only two studies that incorporated general education teachers as an interventionist. This is consistent with the findings of a recent review of literature on supporting the academic improvement of students with ESN in general education settings conducted by Kuntz and Carter (2019), who found general educators were infrequently involved in intervention research occurring within general education classrooms. It is important that future research focuses on supporting general educators to implement shared reading instruction in general education settings because they are the most natural instructor in that setting (Kuntz & Carter). Future research is also needed to investigate the professional development, collaborative planning time, and planning structures that may be necessary for general and special educators to design and implement shared reading in general education settings for students with ESN.
Implications for Practice
These findings indicate that teachers can confidently utilize shared reading as an instructional practice within their comprehensive literacy instructional programs. Therefore, it is important that shared reading, including the systematic instructional practices, book adaptation or presentation strategies, and technology supports that have been identified as effective, are integrated into preparation programs for general and special education teachers. University method courses should include instruction on evidence-based practices, such as systematic instruction with the provision of examples, readings, and assignments that highlight the use of those practices embedded within shared reading. It is particularly important for assignments and fieldwork opportunities to require implementation of practices such as shared reading with the required planning and material preparation to facilitate teacher candidate skill development. A repetitive loop of professional development, practice, and practice-based feedback are essential to ensuring teachers learn to implement strategies with fidelity (Bransford et al., 2000). Without practice and feedback, teachers may internalize ineffective adaptations to instructional strategies or refrain from using them at all (Ericsson et al., 1993).
Footnotes
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
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Editor-in-Charge: Fred Spooner