Inclusive Practices Outside of the United States: A Scoping Literature Review

Abstract

Regardless of differences in abilities, background, or citizenship, students have a right to effective education alongside their peers who do not have disabilities. For many students with disabilities (especially students with more significant support needs), education is something that happens outside of the general education classroom, which limits their access to the academic and social benefits of inclusive education. Fortunately, advocates, educators, and researchers are working to ensure that more students can access inclusive classrooms. Much is known about these practices within the United States, but less research focuses on inclusion in other nations and regions. This scoping literature review synthesizes intervention research conducted outside of the United States and summarizes data from 15 studies representing seven countries and including 42 total student participants. Key findings highlight important outcomes (e.g., academic engagement and communication) and effective methods (e.g., embedded instruction and peer supports) being used around the world to support students with intellectual and developmental disabilities.

1. Introduction

As a group, public education students include a variety of abilities, needs, and backgrounds [1,2]. This heterogeneous group includes students identified with high incidence disabilities and those with intellectual and developmental disabilities (IDD; e.g., students with intellectual disability, autistic students). The heterogeneity found within students with IDD mirrors the heterogeneity found with all students; however, there are some common traits shared by these students. Students with intellectual disability face challenges in learning, socialization, and adaptive functioning, with the onset of these challenges occurring during childhood [3]. Autism Spectrum Disorder is a developmental disorder characterized by deficits in socialization and restrictive behaviors that impact how an individual functions in their daily life [4]. The deficits and challenges associated with IDD can affect students’ performance and participation in school. Regardless of differences between students, all have a right to learn and access education [1,2].

In addition to collaborative international statements [1], the right to equal access is outlined in educational policy from nations and regions around the globe. For example, the Counsel of the European Union has stated that when students with disabilities are included in “mainstream settings” (e.g., classrooms with other students who do not have disabilities), all students benefit [5] (p. 5). In Sweden, policy requires that students with disabilities receive their education within the existing education system rather than separate settings unless there is a special circumstance or reason [6]. The commitment to inclusive education in policy is also seen in countries outside of Europe. In China, recent policy documents emphasize that “learning in regular classrooms” (i.e., alongside other students) is the preferred placement for student with disabilities [7] (p. 6).

As part of meeting that right, students should have access to the classrooms and schools where students would be educated if they did not have disabilities (e.g., neighborhood schools or general education classrooms) with appropriate supports to meet their needs in those environments [1,2]. When students with disabilities are educated in the same environment as their peers who do not have disabilities, all students benefit [8]. This is also true for students with more significant support needs (e.g., students with IDD), who have improved opportunities for socialization and communication [9,10] and access to the general education curriculum’s academic content [11]. Classrooms which include students with and without disabilities also benefit those peers without disabilities [12].

Researchers and advocates for inclusive education have studied trends in educational placement, attitudes about inclusion, and effective supports and interventions for students with disabilities in the general education classroom. Historically, much of this work has been conducted within the United States of America and does not capture how inclusive education practices are implemented in other countries. Fortunately, increasing attention is being paid to how inclusive education is implemented around the world [13]. One troubling trend seen in some countries (e.g., Indonesia, Albania, Bangladesh, and Rural Ethiopia) is that many students with disabilities do not attend any type of school/education at rates much higher than their peers who do not have disabilities [14]. For example, Mizunoya and colleagues found that 79% of school age children in Indonesia without disabilities attended school, compared to only 30% of students who had disabilities [14]. There is also evidence that when examining placement rates for all students with disabilities, educational placement rates are improving. For example, in European countries, the overall percentage of students included in neighborhood schools/general education classrooms increased for five countries between 2010 and 2016 [15]. However, Plichta and colleagues [15] also identified much smaller changes for students with intellectual disability, suggesting that increases in access to this more inclusive environment are not enjoyed by all students with disabilities.

Researchers have also examined perspectives and attitudes about inclusion in multiple countries. Shurr and colleagues [16] completed a review of research on parent perspectives which included 56% of studies conducted outside of North America. The authors found that research on perspectives about inclusion was increasing, but primarily focused on the experiences of mothers [16]. Other studies have examined pre-service and in-service teacher perspectives related to inclusive education, with a limited number of studies examining the perspectives of students with disabilities and their peers who do not have disabilities [13]. A systematic review of studies examining teachers’ attitudes in Spain found that the majority of included teachers viewed inclusion positively, but that many teachers have mixed feelings or negative feelings regarding inclusion [17]. Positive perceptions and increased research are both indicators of increasing acceptance of inclusive education; however, as pointed out by Magnússon and colleagues [6], much of the ultimate success or failure of inclusive education is related to policy contexts outside of these factors.

When identifying specific interventions to support students with disabilities, there are numerous studies and guidelines to review [18,19,20] as well as specific interventions to identify broad, appropriate supports [21], strategies to include peer supports [22], and instructional strategies like embedded instruction [23]. Although the availability of these strategies is beneficial to students and educators, few of these resources focus on strategies used outside of the United States. In their review of English language and Spanish language articles, which included countries from around the world as well as the United States, only 5% of the English language articles and 2% of the Spanish language articles related specifically to intervention research.

The goal of this study is to synthesize and review intervention research that supports inclusive education for students with IDD outside of the United States. While much of the extant research describes effective practices for inclusive education for these students, it has not focused exclusively on practices used outside of the United States. We designed the following research questions to address this gap:

• What interventions are being used to support students with intellectual and developmental disabilities to access inclusive education outside of the United States?
• What are the characteristics of students and interventionists included in the extant research?
• Which outcomes and interventions have the highest success rates in these studies?

2. Materials and Methods

2.1. Inclusion Criteria

We included studies that met three criteria. First, participants had to include kindergarten through 12th grade students with intellectual or developmental disabilities (e.g., Down Syndrome, intellectual disability, autism spectrum disorder, etc.). Second, the study had to be implemented in a classroom setting outside the United States which included both students with IDD as well as their peers who did not have disabilities. Third, the study had to use a single subject research design (e.g., multiple-baseline across participants to test interventions designed to support students with IDD in an inclusive classroom environment [24].

2.2. Search Strategies

The PICOS search strategy was supplemented with forward and backward searches based on the studies included. Figure 1 shows the phases of our search. To begin our search, we used academic databases (i.e., ERIC, Academic Search Complete, Academic Search Premier, APA PsycInfo, and Education Research Complete) in February of 2023 and included dates from 1972–2023. We used Boolean operators to identify the population (e.g., “developmental disabilit*” OR “developmental del*”), the intervention (e.g., “accomoda*” OR “modifica*” OR “embedded instruct*”), the comparison condition (e.g., “school” OR “class*”), the outcome (e.g., “inclusi*” OR “mainstream*” OR “regular education class*”), and the study methods (e.g., “multiple probe” OR “reversal” OR “withdrawal”). The full search string is presented in Appendix A.

We screened titles and abstracts to exclude search returns that were (a) literature reviews or meta-analyses, (b) studies not published in peer reviewed journals (i.e., dissertations, theses, or book chapters), (c) studies that were not conducted in a school (e.g., community or clinic settings), (d) studies that did not include students with IDD, and (d) studies that were not set in inclusive settings.

To identify additional studies that met the criteria, we conducted backward searches by reviewing the citations within included studies. We also conducted forward searches by entering the titles of included studies into Google Scholar and reviewing the “Cited By” studies to identify other studies that met the criteria. One additional study (i.e., Rakap and colleagues was identified through the forward search procedure [25].

2.3. Coding Procedures

After initial abstract/title screening, 338 articles were eligible for full coding. The second author coded articles using a multi-step process to determine if the remaining articles met the criteria. First, they coded research designs, excluding any study that did not use an experimental (i.e., designs with at least three opportunities to demonstrate an effect) single subject research design and studies that were not conducted in inclusive environments. This resulted in the exclusion of 277 studies. Second, they coded the country where the study was conducted and excluded studies conducted in the United States. After this round of coding, 14 studies from the original search and one study from the forward search met the criteria for inclusion.

The 15 studies were coded by the second and third authors. Specifically, we coded the country where the study was conducted, the study design (e.g., multiple baseline across behaviors), the environment in which the study was implemented (e.g., integrated public school classroom), the classroom activity during implementation (e.g., small group instruction), the specific disabilities of all participants (e.g., intellectual disability), the age and grade of participants, the outcome/dependent variable (e.g., social interaction), the components of interventions used (e.g., reinforcement, embedded instruction, or simultaneous prompting), the role of the person/people implementing interventions (e.g., teachers, paraeducators, or peers who do not have disabilities), the outcomes/dependent variables for interventionists (e.g., use of prompts), and the training components for interventionists (e.g., staff professional development).

In addition to descriptive coding, the first and third authors calculated success estimates [26] for all graphs included in the studies based on the changes in level, trend, and immediacy of effect. Success estimates were calculated for tiers of graphs that had at least three data points during baseline for all participants/settings and three non-concurrent replications with at least three points in intervention for all participants/settings. Determinations of appropriate changes in level/trend were made by comparing baseline data to the first five data points following intervention in the A/B comparison of each tier/participant. Immediacy of effect was determined by comparing the last two datapoints of baseline to the first two datapoints in intervention and looking for change in the appropriate direction (e.g., when the goal was to increase behavior, the datapoints were higher than baseline) and zero overlapping data points. Finally, if there was more than one overlapping data point in comparison groups, the graph was coded as an unsuccessful opportunity.

2.4. Coding Reliability

Coders were provided with a detailed manual which included definitions, examples and non-examples, and directions on how to code each stage. Coders then practiced with the first author until at least 95% agreement was met on a practice article. Reliability coding was completed by the first author for at least 20% of articles at each stage. For title/abstract screening the first author coded 20.4% of the articles (i.e., 140). The agreement was 93.6% with nine disagreements. For all nine disagreements, the articles were included for full review and were later removed for not meeting the criteria. To screen for design criteria and study environment, the first author coded 20.7% of the remaining articles (i.e., 70). The agreement was 95%, and disagreements were resolved by consensus discussion between the coders. After all screening procedures, 61 studies met all criteria (i.e., 47 conducted in the United States, and 14 conducted outside of the United States).

The second and third author coded all 14 articles included in the original search and had 87.4% agreement across all variables. In cases of disagreement, the first author reviewed and discussed the disagreement with the third author to determine consensus. The first and third author coded the 15th article added by the forward search together and came to a consensus on all variables.

The first author and the third author coded graphs from the included studies to determine success estimates. The coders collaborated on the first five articles for training and to ensure estimate rules were being applied correctly. Both coders then independently coded the remaining articles to determine reliability. The coders agreed on 61/64 (i.e., 95.3%) of the success estimates. Disagreements were resolved by discussion and review of the coding rules to reach a consensus.

3. Results

Of the 62 studies that met the criteria for inclusion, 15 (i.e., 24.2%) were conducted outside of the United States and present research from seven different countries from both the Global North and Global South regions [27]. The highest number of studies were conducted in Turkey (i.e., five studies, 33.3%) [25,28,29,30,31]. The second most frequent location was Australia (i.e., three studies, 20%) [32,33,34]. There were two studies (i.e., 13.3%) conducted in both Canada [35,36] and China [37,38]. Finally, there were three countries (i.e., Ireland, Saudi Arabia, and South Korea) [39,40,41] with one study conducted (i.e., 6.7%). When considering the region [27] where studies were conducted, the majority of included studies (i.e., 80%) were conducted in the Global North region, and only three studies (i.e., 20%) [37,38,40] were conducted in the Global South region. All other results presented in this paper are summarized from these 15 studies.

3.1. Student Participants

There were 42 total students included in the 15 studies. Individual student ages were provided for 30 of the 42 students ranging from 4 to 13 years old. A total of 56.7% of the students were under the age of 9, and the remaining 43.3% were between 9 and 13 years old. Student grade level was reported for 30 of the 42 students. Twelve students (40%) were in preschool, two students were in kindergarten (6.7%), nine students were in 2nd grade (30%), three students were reported as in 2nd–3rd grades (10%), four students were in 6th grade (13.3%), one student was reported as 6th–7th grade (3.3%), and three students were reported as 7th–8th grade (10%).

Students in the studies had a range of different diagnoses. Fourteen (33.3%) of the students were autistic, twelve (28.6%) of the students were identified as having an intellectual disability, nine (21.4%) of the students were identified as having multiple disabilities or multiple severe disabilities, four (11.9%) were identified as having developmental disabilities, and, finally, there were two diagnoses/identifications of only one student (2.4%; i.e., fragile X syndrome and attention deficit and hyperactivity disorder).

3.2. Other Participants

The included studies described participants and intervention implementers from a range of roles. Across the studies, school staff included 31 general education teachers, 20 paraeducators, and seven special education teachers. Two studies included researchers as intervention agents with two participants identified as researchers, and 12 participants identified as undergraduate researchers. Finally, one study included 22 peer participants described as typically developing peers in an inclusive kindergarten class.

3.3. Settings and Intervention Formats

One of the primary criteria for inclusion in this review was that at least one setting was inclusive (i.e., an academic location where students with and without disabilities were educated together). Some studies included multiple inclusive settings. Studies most frequently took place in inclusive 2nd and 3rd grade classrooms (12 settings); the second most frequently reported setting was inclusive middle school classrooms (eight settings), and studies reported a total of four settings as inclusive preschools and two as inclusive kindergarten classrooms.

Studies included various formats for intervention delivery, including whole class, small group instruction, and 1:1 support provided in the inclusive environment. Across the studies there were 12 instances of 1:1 supports, nine instances of small groups, and seven instances of whole class formats. Six studies used all three formats, five studies used either 1:1 and whole group or 1:1 and small group, four studies only used 1:1, and one study used only the whole class format.

3.4. Dependent Variables and Success Estimates

Table 1. Dependent Variable Categories and Success Estimates by Study and Participant Type.

Students with Disabilities
Work Completion	Social and Communication	Engagement & Challenging Bx
Number correct 4/4 [25] Percent correct 1/4 [28] Percent correct 4/4 [29] Percent correct 3/3 [30] Work completion * (n/a) [36] Percent correct 1/3 [37] Number correct 3/3 [40]	Interactions 0/8 [32] Engagement with peer * (n/a) [36] Interactions 3/3 [40] Initiations 3/3 [40]	Awake/Alert 1/8 [32] On task 1/3 [33] On schedule 0/3 [33] Compliance * 2/3 [34] On task * 2/3 [34] Latency between tasks 0/3 [35] Challenging Bx 1/3 [35] Morning routine * (n/a) [36] On task 1/3 [38] Replacement Bx 0/3 [41] Appropriate Bx 0/3 [41]
Success Estimate: 16/21 = 76.2%	Success Estimate: 6/14 = 42.9%	Success Estimate: 11/46 = 23.9%
Classroom Staff		Peers without Disabilities
Intervention Implementation	Use of Systematic Prompts	Peer Behavior
Number correct 4/4 [25] Percent correct 4/4 [28] Self-monitor 3/3 [30] Percent correct 3/3 [37] Positive interactions 3/3 [41]	Simultaneous prompts 3/3 [30] Physical prompt 0/3 [33] Verbal prompt 1/3 [33] Gestural prompt 1/3 [33]	On task * 0/3 [34] Positive interactions 3/3 [41]
Success Estimate: 17/17 = 100%	Success Estimate: 5/12 = 41.7%	Success Estimate: 3/6 = 50%

Note: Dependent variables are reported by study and categorized by type of behavior and the participant group completing the behavior. The * indicates that not all participants met design standards for calculation of success estimate.

displays dependent variables and success estimates organized by study category of outcome. A variety of dependent variables were included in the studies. Student behaviors included three different categories. We coded three variables as work completion (e.g., “correct responses”), ten variables as engagement and challenging behavior in class (e.g., “compliance,” “on task intervals,” “problem behavior”), and three variables as social/communication (e.g., “engagement with peers,” “initiations”). Staff behavior included eight variables coded as intervention implementation (e.g., “self-monitor”) and use of systematic prompts (e.g., “verbal prompt”). Peer behavior included two variables (i.e., “on task” and “positive interactions”).

Work completion dependent variables had the highest success estimates for student variables with 76.2% of all opportunities coded as successful. Social and communication variables had the second highest success estimates with 42.9% of all opportunities coded as successful. Engagement and challenging behavior variables had the lowest success estimates, with only 23.9% of opportunities coded as successful. The staff intervention implementation variable was also high, with 100% of opportunities coded as successful, but use of systematic prompts was minimally effective, with 41.7% of opportunities coded as successful. Finally, the peer behavior variable was somewhat successful, with 50% of opportunities coded as successful.

3.5. Independent Variables and Success Estimates

Table 2. Intervention Component and Success Estimates by Study.

	Rf/Praise	Prompting/ Modeling	Alternative Bx	Embedded Instruction	Peer Mediated	Visual/ Antecedent	Conting. Mapping
Ai and colleauges [37]	✓	✓		✓
Aldabas [40]	✓				✓
Brown and Mirenda [35]	✓						✓
Cho Blair and colleauges [41]	✓		✓
Fidan and Tekin-Iftar [28]	✓	✓
Firat and Ergenekon [29]	✓	✓		✓
Foreman and colleauges [32]	✓				✓
Greenberg and colleagues [38]	✓	✓	✓
Guertin and colleagues [36]	✓					✓
Hall and colleagues [33]	✓	✓				✓
Imasaka and colleauges [34]	✓
Kiyak and Tekin-Iftar [30]	✓	✓
O’Reilly and colleagues [39]	✓		✓
Rakap and colleauges [25]	✓			✓
Sazak Pinar [31]	✓		✓
Strategy	Rf/Praise	Prompting/Modeling	Alternative Bx	Embedded Instruction	Peer Mediated	Visual/Antecedent	Conting. Mapping
Number of studies with strategy	15	6	4	3	2	2	1
Success Estimate for Student DV	33/81 = 40.7%	10/20 = 50%	4/20 = 20%	9/11 = 81.8%	10/25 = 40%	1/6 = 16.7%	1/6 = 16.7%

Note: The “✓” symbol indicates that a strategy was used in a given study. Success estimates were calculated using visual analysis of changes in level, trend, and immediacy of effect, and a lack of overlap in 5 data points before and after intervention.

displays intervention components and associated success estimates by study. We coded intervention components into eight different categories. Reinforcement/praise was included in all 15 studies and associated with dependent variables, with 40.7% success estimates. Prompting/modeling was the second most frequently included component (i.e., six studies) and was associated with dependent variables, with 50% success estimates. Providing an alternative behavior was included in four studies and was associated with dependent variables, with 20% success estimates. Embedded instruction and self-mediation were both included components in three studies and were associated with dependent variables, with 81.1% and 77.8% success estimates, respectively. Peer mediation and use of visuals/antecedent changes were both included components in two studies and were associated with dependent variables, with 40% and 16.7% success estimates, respectively. Finally, contingency mapping was included in one study and was associated with dependent variables with 16.7% success estimates.

4. Discussion

All students have a right to receive an effective and inclusive education. This includes students with disabilities, including students with more significant support needs. Researchers have synthesized and reviewed studies that test various interventions for these students in the general education classroom, yet most of these reviews are dominated by studies conducted in the United States. We conducted a scoping literature review focused on studies conducted to support inclusive educational outcomes for students with IDD outside of the United States. Findings showed that most studies included goals related to academic and classroom engagement or the reduction of challenging behavior. Most studies were found to have used multiple intervention components and a variety of grouping strategies. Findings also indicated that publication trends outside of the United States are changing over time.

First, the studies primarily focused on outcomes related to engagement and challenging behavior. The prioritization of these outcomes is encouraging and points to both reducing common barriers to inclusion (i.e., challenging behavior), and ensuring that students have access to academic instruction within the general education classroom. One of the most commonly cited benefits of inclusive education is access to the general education curriculum [9]. The high rate of studies focused on both correct performance of skills and overall engagement in the classroom suggests that extant research is focused on ensuring that students are meaningfully engaged in the classroom, rather than just being present. Similarly, it is encouraging that many of the studies included outcomes related to the reduction of challenging behavior. These outcomes are extremely important, as the presence and persistence of challenging behavior is frequently described and used as a justification to educate students in alternative placements [8]. Finally, several studies also included outcomes related to staff implementation of strategies associated with high success estimates. This points to a high fidelity of implementation within the studies and helps support the validity of the associated student outcomes and suggests that the included strategies are achievable for teachers and other professionals.

Second, all studies included multicomponent interventions using evidence-based strategies to support student outcomes. All studies incorporated the use of praise or other forms of reinforcement which are identified as evidence-based procedures for students with IDD [21]. Similarly, it is encouraging that systematic prompting procedures were commonly used strategies. Interestingly, embedded instruction (which had the highest success estimate of any individual strategy) was only used in three studies. This is somewhat striking given that many of the outcomes were tied to academic engagement, which could be ideal for the use of embedded instruction. A possible explanation could be that embedded instruction was not used in studies that already included some other form of 1:1 instruction (which was the case in 11 of the 12 studies that did not include embedded instruction). The grouping formats used in the included studies also support the idea that meaningful inclusion was being targeted. Twelve of the studies included either small group or whole class formats, which indicates that interventions were not only implemented in isolation, but that an opportunity to generalize outcomes was present.

Finally, the overall trend of intervention research represented in this review is increasing and is representative of countries from around the world. It is encouraging that of the 62 studies which met the criteria for inclusion, nearly a quarter were conducted outside the United States. This is a marked increase in comparison to the review conducted by Dada and colleagues [19], in which 7.7% of studies were conducted outside of the United States. It is also worth noting that while the trends of publication outside of the United States are increasing, the majority of these studies were conducted in the Global North, and the Global South was only minimally represented. Data from the reviewed studies do not specifically point to a reason for this discrepancy; however, recent policy [5] highlights the commitment of countries like China in the Global South to inclusive education. There also is evidence of changes in publishing trends over time. When looking at publication dates over the thirty years represented in this study, 46% of the included studies were published within the last five years (including two of the three studies conducted in the Global South). That number increases to 66% when reviewing the last ten years. This increase in research related to inclusive education is encouraging and aligns with shifts in goals and attitudes surrounding inclusion [17] but should be considered in the broader special and general education policy contexts [6]. It is also encouraging to see that the participants included a large percentage of students with intellectual disability. Given the findings of limited growth in placement rates between 2010 and 2016 for these students [15], the high number of participants could be an indicator of increasing acceptance and placement of these students.

Limitations and Future Directions

The limitations of both this review and the included studies suggest future avenues for research. First, there are limitations with this review. While the goal of this review was to understand and synthesize inclusive research internationally, we only included studies published in English and it is possible that this excluded studies from regions in the Global South (e.g., all of Central and South America). Future research teams could expand on these results by including journals published in other languages and using more diverse databases for their searches. Additionally, by relaxing the constraints on methodology used as a criterion for including studies and adding the names of specific regions and countries to the search strings, future researchers could better capture the work being done that this review missed (e.g., studies in South America or Africa). The use of success estimates is also a limitation of this review. Success estimates are useful for describing the consistency of interventions, but do not speak to the magnitude of intervention effectiveness. Future reviews could include more sophisticated and sensitive measures of effect size for single case research.

Second, the included studies also have limitations associated with single case research design. Single case designs, like those included in this review, are strong in terms of internal validity but suffer from minimal external validity and generalizability. Future researchers could replicate successful interventions from the reviewed studies using more powerful group designs to address this limitation. There is also limited information in the studies included about how skills generalize to other environments. Future researchers could replicate interventions and specifically measure outcomes in untrained generalization settings as well as conduct studies to compare effects of interventions between settings to address this gap.

5. Conclusions

Findings from this review indicate that researchers outside of the United States are conducting more studies related to inclusive education for students with significant support needs. The majority of these studies were conducted in countries from the Global North, with some encouraging new work being conducted in the Global South. The increase in publications in recent years is hopefully indicative of a move towards effectively supporting inclusive education around the world. The reviewed studies demonstrate that outcomes relevant to meaningful inclusion of these students are being targeted using effective and evidence-based strategies. Given the benefits of inclusive education and the commitment of the international community to achieving equitable education for all students, these studies paint a promising picture of how educators, policy makers, and researchers are continuing to make strides towards the goal of inclusive education and ensuring that all students, regardless of individual differences, access their right to an effective education.