Abstract
Keywords
Toews and Kurth (2019) provide convincing arguments to support their call to action. Their appeal is for research addressing comprehensive literacy instruction for students with extensive support needs that is implemented in general education classrooms. To date, there are no studies addressing this need. We do have an emerging body of research evaluating the effectiveness of multi-component early literacy interventions designed for this population of students (e.g., Allor, Mathes, Roberts, Jones, & Champlin, 2010; Bradford, Shippen, Alberto, Houchins, & Flores, 2006; Browder, Ahlgrim-Delzell, Flowers, & Baker, 2012; Lemons, Mrachko, Kostewicz, & Paterra, 2012; Whalon, Otaiba, & Delano, 2009). However, as Toews and Kurth (2019) point out, these studies were conducted in separate settings. We also have a group of literacy studies that were implemented in general education settings (e.g., Collins, Evans, Creech-Galloway, Karl, & Miller, 2007; Hudson & Browder, 2014; Ruppar, Afracan, Yang, & Pickett, 2017). However, this research is limited to investigations of components of literacy instruction, rather than comprehensive courses of instruction that address the literacy skill areas identified by the National Reading Panel (NRP) Report (National Institute of Child Health and Human Development [NICHD], 2000) as building blocks for beginning reading—phonological awareness, phonics, comprehension, vocabulary, and fluency. I propose that through the unification of these separately developing lines of research, we can accomplish the desired outcome of Toews and Kurth’s (2019) call to action. To that end, investigations incorporating both comprehensive, research-based literacy instruction and inclusive instructional contexts could be designed as replication-extension studies (Bonett, 2012; Coyne, Cook, & Therrien, 2016; Doabler et al., 2016). These conceptual replication studies would give researchers the freedom to explore various inclusive contextual models for implementing comprehensive early literacy instruction while maintaining intervention fidelity.
An Example
During the 2016-2017 school year, Elizabeth Kozleski and I, with members of our research teams, conducted such an investigation. Our study, funded by the Institute of Education Sciences, was designed to replicate a comprehensive emergent literacy program developed for students with extensive support needs and to extend that research to small, integrated group contexts in general education settings (Hunt et al., 2019). We chose Early Literacy Skills Builder (ELSB; Browder, Ahlgrim-Delzell, Courtade, Gibbs, & Flowers, 2008; Browder et al., 2012) as our literacy intervention. ELSB was selected because it draws on the NRP Report (NICHD, 2000) to include both code and meaning-focused skills while using instructional methods and systematic prompting procedures found to be effective with students with severe disabilities, including students who are nonverbal responders (Browder, Ahlgrim-Delzell, Spooner, Mims, & Baker, 2009). In addition to lessons addressing phonological awareness, phonics, vocabulary, and comprehension, it includes a shared story-reading component using literature from the students’ grade level to teach the children to interact with books in addition to developing comprehension and vocabulary skills.
Although the results of the studies conducted by Browder and her colleagues (Browder et al., 2008; Browder et al., 2012) to investigate the effects of ELSB instruction on the emergent reading performance of students with extensive support needs are promising, the instructional settings for those studies were limited to special education classrooms. By conducting efficacy trials in separate settings only, practitioners and policy makers may conclude that special education classrooms are the only places where students can receive instruction that is systematic and comprehensive and which cannot easily be distributed and embedded in the context of ongoing literacy instruction in general education classrooms.
The purpose of our conceptual replication study was to evaluate the effects of ELSB instruction on the early literacy performance of students with extensive support needs when it was implemented in general education classrooms in small group contexts with peers participating in the lessons as “reading buddies.” The randomized controlled trial (RCT) was conducted in 16 schools in three states. Eighty students with extensive support needs participated in the study (4-8 students per school site). The students were eligible to take alternate assessments, enrolled in Grades K-4, read below the first grade-level, had picture discrimination skills, and were able to sit for short periods of time for instruction. Twenty-four students who were nonverbal responders participated in the study. The eligible students at each school site were matched into pairs for each teacher based on their disability diagnosis, verbal status, gender, and grade. One student from each matched pair was randomly assigned to the intervention group (ELSB instruction), and the other student was assigned to the control group (“business-as-usual” instruction; BAU).
Instruction for both groups of students was implemented in general education classrooms by special educators during the classroom’s literacy block. The majority of the students were mainstreamed into general education classrooms for the literacy lessons, although a small number of the students (4%) were full-time members of the class. The daily lessons were 30 to 40 minutes in length. All 80 students were assigned a general education peer as a reading partner. On the average, 66% of the students in each class volunteered to be reading buddies. The opportunity for the general education students to serve in this role was rotated through the list of volunteers. Reading buddies participated with their partner in the literacy lessons. They did not take a “tutor” role; however, they did serve as proficient models of the targeted reading skills, thereby providing repeated opportunities for observational learning by the focal students (Bandura, 1977; Bandura, 1986).
ELSB lessons were implemented in small, integrated group contexts (typically, two students with extensive support needs and two reading buddies). The majority of BAU instruction was also implemented in small, integrated groups; although in some classrooms, BAU instruction was delivered through whole-class lessons with adapted materials and support from special educators and peers or a combination of whole class and small group lessons. These integrated group contexts for students in both groups capitalized on the potential for observational learning through peer modeling of targeted literacy skills. Special education teachers and paraprofessionals partnered to teach the lessons on a rotating basis. General education teachers managed the calendar that identified the reading buddy for each day and collaborated with the special education teacher to determine the way in which ELSB and BAU instruction would be physically integrated into the classroom space and ongoing classroom activities.
ELSB small group lessons to teach phonological awareness, phonics, vocabulary, and comprehension focused on stories from a book about a frog named Moe. A frog puppet was used during the lessons to motivate the students to attend and participate. Systematic prompting procedures (constant time delay), correction, and reinforcement procedures were used. The direct instruction method of model, lead, and test was embedded within the lesson framework. Lessons were repeated as needed for students to begin making independent responses. These lessons alternated each day with the shared story-reading component of ELSB. During shared reading lessons, students were taught the conventions of reading and text awareness, comprehension, and vocabulary skills. Adapted grade-level books were used, with general education teachers recommending books that were popular among students at that grade level and books currently included in the classroom’s literacy activities.
The BAU lessons were designed by the special education teachers to meet the participating students’ literacy goals. An analysis of the characteristics of that instruction is currently underway. Preliminary analyses reveal that the majority of the lessons were “teacher designed” and addressed individual components of early literacy instruction (e.g., letter names and sounds or listening comprehension taught in a shared reading context), rather than the full range of early literacy skills.
Replicating with Fidelity
Efforts were made to promote high levels of fidelity of ELSB implementation for this replication study. Research team members were certified in ELSB implementation by members of the research team who developed and evaluated the ELSB program. Following their certification, they conducted a 2-day ELSB training for participating special education teachers during the summer, with a “booster” session scheduled close to the start of the school year. Finally, after the onset of the study in early September, research team members (“school site coordinators”) observed literacy lessons implemented each week across school sites and provided corrective feedback if needed. Their observations were guided by a detailed fidelity checklist documenting the level of accuracy of implementation of each instructional step. These efforts produced an average fidelity score across school sites of 95.1%. Interobserver agreement measures, collected during 22.4% of the fidelity sessions, showed 96.2% agreement between the independent observer and the primary fidelity data collector.
Promising Results
Literacy data were gathered in five waves scheduled across the school year. The two literacy measures were the Gates–MacGinitie Reading Test (4th ed.), Pre-Reading Level (GMRT; MacGinitie, MacGinitie, Maria, & Dreyer, 2000) and the Nonverbal Literacy Assessment (NVLA; Baker, Spooner, Flowers, Ahlgrim-Delzell, & Browder, 2010). Data analysis procedures included hierarchical linear modeling (Raudenbush & Bryk, 2002; Singer & Willett, 2003) conducted separately for the GMRT score and NVLA total score and subtest scores. The models estimated overall group difference across time and group difference in change over time. In addition, independent-sample t tests were performed to identify specific measurement times where the group differences were significant.
The results of the study were promising. They showed that assessment scores for students in both the ELSB group and the BAU group showed a significant increase in each of the two literacy measures over the school year when the students’ literacy instruction was implemented in general education classrooms. However, we did find that scores for students receiving ELSB instruction showed greater gains as compared to those of students in the BAU control group. The one exception to this was the finding for the NVLA subtest, Vocabulary/Word Study. For this subtest, scores for both groups showed a significant increase over time; however, there was no difference between the groups. This finding suggests that both ELSB and BAU instruction promoted vocabulary development as measure by the NVLA. The moderate effect sizes for our study were promising, particularly given the targeted population of students. Effect sizes at 9 months were .49 for the total GMRT score and .55 for the total NVLA score. Effect sizes for the NVLA subtests ranged from .48 to .69, with the largest effect size found for measures of comprehension. These effect sizes were stronger than those reported by the Browder et al. (2012) study (i.e., .30-.49), although it is difficult to make comparisons between the studies because of the difference in control group interventions.
The Role of Social Validity Analyses in Replication-Extension Research in Inclusive Settings
Social validity studies serve an important role in understanding, from stakeholders’ perspectives, the benefits associated with the intervention under investigation—as well as the cultural, ecological, and logistical obstacles that may affect the adoption, sustainability, and scaling up of the intervention (Kazdin, 1977; Lindo & Elleman, 2010; Wolf, 1978). To that end, we conducted a qualitative study—as a partner to our RCT—to examine the social validity of the intervention’s goals, procedures, and outcomes from the perspective of participating general and special education teachers, principals, and reading buddies (Kazdin, 1977; Wolf, 1978). Focus groups and individual interviews were conducted to solicit stakeholders’ perceptions of, their opinions about, and their experiences with the delivery of comprehensive early literacy instruction to the students with extensive support needs in general education classrooms. For example, we asked all interview participants to share their perspectives on the importance of teaching early literacy skills to students with extensive support needs and the importance of implementing that instruction in general education classrooms. We explored teachers’ perceptions of their ability to implement the ELSB instructional procedures easily and effectively. We asked teachers and principals to discuss their perceptions of the logistical and ecological factors that facilitated or impeded the inclusion of the students with extensive support needs in general education classrooms for literacy instruction. Finally, we asked all participants to describe the degree to which they thought that focal students, their reading buddies, and adults in the general education classrooms benefited from their experiences.
The “intervention” for our RCT included not only a comprehensive early literacy program for students with extensive support needs (i.e., ELSB), but also a model for delivering that instruction in general education classrooms (i.e., a small group context with peers participating in the instruction as reading buddies). Participants’ perspectives on the importance and acceptability of the intervention goals, procedures, and outcomes provided critical guidance for revising and refining the intervention to increase its adoptability and sustainability. When principals and teachers perceive that the intervention is important, teachers have confidence in their ability to implement systematic comprehensive literacy instruction in integrated, small group contexts, and peers find their participation as reading buddies enjoyable and worthwhile, the probability that the intervention will be adopted and sustained is much more likely. Stakeholder feedback provides critical guidance on how to improve the intervention to address obstacles and create a better fit with classroom and school culture and ecology.
Final Thoughts: Placement Makes a Difference
If inclusive models for delivering comprehensive literacy instruction are to be adopted, sustained, and scaled up, the issue of educational placement must be considered.
The majority of students with extensive support needs in the United States are placed in separate classrooms for the majority of the school day (National Center for Education Statistics, 2018). In our study, only four of the 80 students were members of general education classrooms and educated for the majority of the day in inclusive settings. Therefore, the reality is that models for delivering comprehensive literacy instruction in general education settings that can be widely replicated have to be built upon a mainstreaming version of inclusive education—and placement makes a difference.
When the special education teachers in our study were engaged in delivering comprehensive literacy instruction to students in general education classrooms, the education of the remaining students on the teachers’ caseload could not be disrupted. Teachers and principals put plans into place so that educational activities continued under the supervision of qualified personnel. However, scheduling and staffing logistics and the teachers’ concerns about “leaving their class behind” were challenges that focus group teachers and principals candidly discussed. They are also the realities that must be considered as the field responds to Toews and Kurth’s (2019) call to action.
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: Stacy K. Dymond