How Do They See Themselves? The Self-Concept of Students with Intellectual or Behavioural Disabilities in Special Education

ABSTRACT

Research into the self-concept of students with an intellectual disability (ID) or social, emotional and behavioural difficulties (SEBD) educated in special education is relatively scarce. This, whilst many students with ID or students with SEBD still reside in special rather than inclusive school settings. This study aimed to gain insight into the self-concept of students with either ID or SEBD in special education and examined variables that might relate to the self-concept of these students. A total of 171 students educated in special schools filled out the Self-Perception Profile for Children (SPPC). T-tests utilising bootstrapping and variance analyses were used to analyse the data. The results indicate that girls with ID and boys with SEBD assess their self-concept more negatively than typically developing students. Gender, age and formally assessed classifications of ASD or ADHD yielded different self-concept scores in both groups of students on different dimensions of self-concept. The expected positive illusory bias was not found in this study. The results of the study are discussed in light of the educational setting of the students with ID or students with SEBD. This study demonstrates that interventions are needed to improve the self-concept of students with ID or students with SEBD in special education.

KEYWORDS:

• Special educational needs
• self-concept
• special education
• positive illusory bias
• social comparison theory
• social acceptance

Introduction

Today’s schools are no longer solely regarded as institutions that shape students on an academic level. Besides the clear purpose of teaching knowledge and skills, the role of education in the social-emotional development and well-being of students has been recognised (Biesta, 2020; Florian, 2014), such as the role schools play in the development of students’ self-concept. In line with earlier research (Douma et al., 2022; Harter, 2012), self-concept can be defined as a collection of positive and negative concepts about one’s self, concepts that are continuously re-evaluated through new experiences and are becoming increasingly stable over time (Harter, 2012; Marsh, 1987). It has long been established in self-concept theory that self-concept is a multidimensional or multifaceted construct. Therefore self-concept can encompass various distinct dimensions, such as social, academic or physical self-concept, complemented with a more global self-concept (Marsh & Hattie, 1996; Marsh & Shavelson, 1985).

The importance of self-concept for individuals on numerous developmental outcomes has been widely examined and is well-known. A positive self-concept is found to be positively associated with many favourable outcomes, such as increased prosocial behaviour, academic achievements, school satisfaction, social inclusion, motivation, and better mental and physical health, as well as offering protection from problem behaviours (Avramidis, 2013; Valentine et al., 2004; Ybrandt, 2008). In contrast, a negative self-concept is linked to depression, loneliness, low academic achievements, delinquency, substance abuse, and aggression and poses a risk factor for internalising problem behaviour (Kozina, 2017; Lee & Stone, 2012; Montague et al., 2008; Valentine et al., 2004; Ybrandt, 2008).

The extent to which self-concept contributes to these favourable or disadvantageous outcomes is dependent on the importance assigned to it by an individual (Douma et al., 2022; Harter, 2012) and differentiates between the dimensions of self-concept. Not all dimensions correlate equally with all these outcomes. For academic achievement, literature has shown that different dimensions of self-concept have distinct relationships with academic achievement. Whereas global self-concept and social self-concept show smaller correlations with academic achievement, the correlation between academic achievement and academic self-concept is found to be moderate (Marsh, 1992; Wentzel, 1991). Despite these differential relationships, self-concept is generally considered an important construct in the educational context of students, as it is seen as an important predictor of academic achievement and school success and motivation (Niemiec & Ryan, 2009; Ryan & Deci, 2000).

According to the well-known Social Comparison Theory (SCT) by Festinger (1954), the immediate surroundings of individuals such as peers, play a key role in the development of self-concept. Self-concept is construed by making social comparisons. Festinger argued that to accurately assess our self-concept we have to compare ourselves with others we interact with. By doing so, we can establish a benchmark from which we then can make appropriate inferences about our self-concept. Consequently, peers in the classroom have a great influence on self-concept development, especially on the academic and social self-concept of students (Schaffer, 1996).

The extent to which social comparison takes place is dependent on the developmental age of children. Whereas young children lack the cognitive capabilities to form higher-order self-concepts and form mostly unstable and overly positive self-concepts, a growth in observational and evaluative capabilities in middle childhood enables children to further utilise social comparisons (Harter, 2012; Harter et al., 1988). Ruble (1983) determined that social comparisons first occur in children between the ages of seven and nine years old. By that time children are capable of making both positive and negative evaluations of themselves, with increased accuracy and are able to integrate evaluations to form an overarching assessment of their self-concept on a specific dimension (Harter, 2012). Early adolescence is a developmental stage characterised by uncertainty, increased attention on peers instead of parents, and the focus on discovering one’s self and it is in this period that the most growth in self-concept occurs. The social comparisons that are made by young adolescents are those of extremes, either very favourable or very unfavourable towards themselves. However, in late adolescence, youth have learned more about themselves and are able to come to more balanced comparisons to their peers. During late adolescence, the foundations are laid for a more stable self-concept which carries over into early adulthood (Crain, 1996; Harter, 2012).

To make social comparisons to peers, the school context is subsequently of vital importance for all children and adolescents as it provides the ideal setting for students to engage with each other and by doing so create opportunities for social comparison and the cultivation of the self-concept. However, not all students experience similar educational settings and this holds most true for students with special educational needs (SEN), such as an intellectual disorder (ID) or social, emotional and behavioural difficulties (SEBD).

Although many countries strive for the realisation of inclusive education (United Nations, 2006), there are many students with ID or students with SEBD still educated within special education. One of the countries aiming to make the transition towards inclusive education is the Netherlands. In the Netherlands, there has been a longstanding tradition of a highly segregated educational system, which has led to the founding of special schools for students with a particular disability. In recent years there has been a gradual shift towards combining different special schools, but as this is a transition in progress, there are still many special schools that focus on a specific student population. Among these special schools, there are schools specifically tailored to the education of students with ID and schools focused on students with SEBD. With the passing of several legislations to promote and facilitate inclusive education, the Netherlands was moving towards a decline in the student population in special schools. However, recent years have seen a change in this aspired trend and the student population in special educational settings has been gradually increasing over the last 5 years (Inspectie van het Onderwijs, 2022). It is estimated that 5% of all Dutch students are enrolled in special education and these students are predominantly male, with females accounting for approximately 25% of all students (CBS, 2019).

Given the segregated education setting, the social school contexts of students with a disability in the Netherlands differ greatly from their typically developing (TD) peers attending mainstream education. Applying the SCT, this results in a distinctively different peer group from which social comparisons about the self are made. This, in turn, can influence the development of the self-concept in students with ID or students with SEBD in special schools. This holds particularly true for the self-assessment of academic achievements. Marsh (1987) labelled this phenomenon as the Big-Fish-Little-Pond Effect (BFLPE). The BPLFE refers to the theory that states that students will develop lower academic self-concepts when educated in a high-performance classroom due to negative social comparisons. Comparisons are more apparently made with high-capacity students than with low-capacity students (Bukowsi & Raufelder, 2018). Students with SEN in regular classrooms are therefore more likely to develop negative academic self-concepts since they are most often surrounded by higher-achieving TD peers (Marsh, 1987). Students with SEN, such as ID or SEBD, in special classrooms are educated amongst peers that perform on a more equal academic level compared to regular classrooms, therefore special schools are comprised of very different comparison groups for students with SEN.

Research about the self-concept of students with ID or students with SEBD, which focuses frequently on solely academic or social aspects of self-concept, shows different results for both groups. Several researchers have found that the self-concept of students with ID concerning academic achievement is more negative than that of TD peers in mainstream education (Avramidis, 2013; Bear et al., 1993; Pijl & Frostad, 2010). In addition, the comprehensive review of Maïano et al. (2019) showed that students with ID in special schools reported higher levels of self-concept compared to students in inclusive schools and lower self-concept scores were obtained on global self-concept by students with ID in comparison to TD peers. Research into the self-concept of students with SEBD is relatively scarce in comparison to that of students with ID. Of the limited studies focusing on students with SEBD, several studies suggest that students with SEBD do not differ in self-concept scores from TD students (Hoza et al., 2004) or report even very high scores of self-concept (Gage & Lierheimer, 2012). For students with SEBD, however, a more nuanced picture regarding self-concept emerges from literature. Several studies have found the phenomenon of positive illusory bias (PIB) in samples of students with SEBD. PIB refers to elevated self-reported levels of self-concept that do not seem to reflect reality. In students with SEBD, this leads to an overly positive assessment of self-concept (Gresham et al., 2000; Hoza et al., 2004; Owens et al., 2007). Swinson (2008) demonstrates in his study that students with SEBD in special schools report on average mean self-concept scores, but compared to a normative sample there is a greater number of students with SEBD who report extremely high or extremely low self-concept scores. In contrast, Wei and Marder (2012) found that students with SEBD in mainstream schools assessed their social self-concept more negatively than their typically developing peers.

Studies concerning the self-concept of students with ID or students with SEBD are primarily conducted within mainstream or inclusive school settings, and, even though a large percentage of students with either ID or SEBD is educated in special schools, not many studies have focused on the self-concept of students with SEN in special education. By gaining greater insight into the self-concept of students with ID or students with SEBD in special educational settings, it increases not only our understanding of self-concept as a construct but also sheds light upon self-concept in relation to educational settings. Studies that do focus on students with SEN (i.e. students with ID or SEBD) in special schools mostly confine their focus to the academic and social self-concept (Douma et al., 2022), and often do not specify specific types of SEN within their sample. Therefore, research focusing on the self-concept of both students with ID and students with SEBD educated in special education remains limited. This, whilst it is plausible that students in special education develop their self-concepts differently compared to students with SEN in mainstream or inclusive schools due to classroom composition and comparison group differences. This study, therefore, pursued to answer the following research questions:

How do students with ID or students with SEBD educated in special education assess their self-concept?

1. How are student characteristics gender, age and formally assessed classifications (i.e. DSM classifications) related to the self-concept of students with ID or students with SEBD?

2. Is there a difference in self-concept between students with ID or students with SEBD compared to their typically developing peers?

Method

Design and Setting

A cross-sectional study was set up to gain a greater understanding of the self-concept of students with ID or SEBD educated in different special schools in the Netherlands. The study was part of a larger study, which received approval from the university’s Ethics Committee. The study focused on increasing the autonomy of students with ID or SEBD in primary and secondary special education. These special education schools can be viewed as segregated special school settings for students formally assessed with either ID or SEBD.

Participants

A total of 18 special schools were invited to take part in the study. Of these 18 schools, four primary special schools for students with SEBD and seven secondary special schools for students with ID participated. Because of the wider scope of the study, there were several reasons for schools to be unwilling to participate (i.e. there was deemed to be no time to participate in the study (5), and there were concerns about the feasibility of the study for their students due to the severity of their intellectual impairment (3)).

The participating special schools then selected classrooms with students who met the inclusion criteria. Due to the diversity of the group of participants, different criteria were formulated for students with ID and students with SEBD. For students with SEBD in primary special education, all students enrolled in Grades 5 and 6 were included in the study. Students with ID in secondary special education had to be at least 14 years of age and able to read. Written consent was acquired from parents and students. Students were only included in the study if both parents and students gave consent, regardless of the student’s age.

A total of 171 students took part in the study, of which 71 students with ID (75% boys) and 100 students with SEBD (91% boys). The mean age of the students with ID was 15.9 (SD=1.9) and 12.1 (SD=1.0) for students with SEBD. Of the students with ID, 17 students were formally assessed with AHDH (24%), and 24 students were formally assessed with ASD (34%). In the sample of students with SEBD, there were 31 students with an ADHD classification (47%) and 31 students with an ASD classification (47%).

Instruments

The self-concept of students with ID or students with SEBD in special education was measured using the Dutch version of the Self-Perception Profile for Children (SPPC) by Harter et al. (1988), Veerman et al. (2004). This instrument corresponded most closely to the aims of this study. The instrument was deemed appropriate not only for students with SEBD but also for students with ID. Though the age parameters of the SPPC do not coincide with the chronological age of the students with ID, they are however in line with their developmental age. Furthermore, the items of the SPPC could be read more easily and therefore promote independence in participation for students with ID, while simultaneously providing self-concept scores in line with the aims of the study.

The SPPC is a self-report questionnaire for children aged 8–12 years, designed to assess self-concept in six different dimensions: scholastic competence, social acceptance, athletic competence, physical appearance, behavioural conduct, and global self-worth. The SPPC is comprised of 36 items, six items per subscale. For each item, children have to choose one of two statements (e.g. ‘Some kids often forget what they learn’ but ‘Other kids remember things easily’). After choosing the best fitting statement, children can indicate if the chosen statement is ‘somewhat true’ or ‘very true’ for them. Each item is subsequently scored on a four-point scale with a lower score reflecting a more negative and a higher score a more positive self-concept in each dimension. A total raw score for each dimension of self-concept is calculated and a corresponding percentile score can be found in the norm tables of the SPPC. The SPPC included norm tables for boys and girls separately. In this study, the percentile scores were used to compare the results of students with ID or SEBD to a TD norm group.

The reliability of the Dutch version of the SPPC ranges from .65 to .84 (Veerman et al., 2004). For the current study, the overall reliability coefficients were found to be highly reliable (α = 0.88). The reliability of the subscale scores ranged from .62 (scholastic competence) to .83 (behavioural conduct).

Using school record analysis, student characteristics concerning gender, age and formally assessed type of SEN (i.e. DSM classifications) if available of all students included in the study were gathered.

Procedure

A pen-and-paper version of the SPPC was administered during school hours. Taking into account the differences between both groups of participants, the data-collecting procedure was tailored to the specific support needs of students with ID or SEBD. Students with ID completed the questionnaire in small groups (3–4 students) or individually, whereas students with SEBD filled out the SPPC collectively. Each small group or class first received a standardised introduction to the questionnaire in which the example questions from the manual were discussed. The questionnaire was thereafter administered under the supervision of the principal researcher (first author), a research assistant and the class teacher. This procedure was used to ensure students answered independently and confidentially and to answer questions if necessary.

Data Analysis

To answer the research questions, the descriptive statistics were calculated from the self-concept scores of both groups of students. The mean raw self-concept scores per dimension were rounded and subsequently converted to the corresponding percentile scores according to the norms of the SPPC. The norm scores of the students with ID or SEBD were subsequently compared to the SPPC provided norm group of TD peers.

To determine gender differences within both subgroups and to assess the relationship between formally assessed types of SEN and self-concept scores, independent t-tests were conducted employing bootstrapping with 1,000 bootstraps. The bootstrap method was utilised to control for the violated assumption of normal distribution and sample size. The analyses were conducted using the raw scores. School records yielded mainly formally assessed types of SEN by using DSM classifications of Attention Deficit Hyperactivity Disorder (ADHD) and Autism Spectrum Disorder (ASD). Therefore, these classifications were used for further analyses.

A one-way ANOVA was performed with the raw scores to investigate the age-related differences in self-concept on the six dimensions of self-concept for students with SEBD and an independent t-test for students with ID. Variables for different age groups were computed. Students with ID were divided into two age groups, based on developmental stage. Group 1 consisted of students in middle adolescence and included students aged 16 years or younger (N = 37). Students in their late adolescence aged 17 years and older formed group 2 (N = 34). The sample of students with SEBD was divided into three age groups, based on school grades. Group 1 (N = 26) was aged 10–11 years, group 2 (N = 45) was 12 years of age and group 3 (N = 29) was aged 13–15 years. When appropriate, post hoc comparisons were performed using the Tukey post hoc test. Effect sizes were calculated by means of partial eta squared (ƞ2). Values of 0.01, 0.06 and 0.13 represent a small, moderate, and large effect size respectively, following the guidelines of Cohen (1988).

Results

Self-Concept of Students with ID

The mean raw scores of the six dimensions of self-concept of students with ID were calculated. Table 1 presents an overview of these mean raw scores and is further specified for boys and girls. In addition, the corresponding percentile scores are reported. The percentile scores allow for comparisons of the self-concept scores of students with ID to TD students.

Table 1. Mean Raw Scores and Percentile Scores of the Self-Perception Profile for Children of Students with ID.

	Students with ID
	Boys (N = 53)			Girls (N = 18)
	Mean	SD	Perc.	Mean	SD	Perc.
Scholastic competence	17.68	3.20	56	15.67	2.17	47
Social acceptance	18.43	3.35	47	17.50	3.13	56
Athletic competence	19.04	3.51	53	16.22	4.54	30
Physical appearance	21.28	3.21	51	17.94	3.70	40
Behavioural conduct	18.13	4.79	66	17.22	3.84	36
Global self-worth	21. 38	3.05	57	19.67	2.33	51

Note. Perc. = percentile; the range of the scores of self-concept is 6–24.

The mean raw scores of the total sample of students with ID indicate that students with ID reported the highest self-concept scores on the dimensions physical appearance and global self-worth, whereas the dimensions behavioural conduct and scholastic competence were scored the lowest. The same picture emerges when examining the scores of boys with ID. Girls with ID scored high on global self-worth, but not on physical appearance and scored lowest on scholastic competence. Compared to boys with ID, the mean raw scores of girls with ID were lower across all self-concept dimensions.

To determine whether or not boys and girls with ID differ in their assessment of the six dimensions of self-concept, independent t-tests were conducted. The analyses utilised bootstrapping and were performed with raw scores. The results showed several statistically significant differences between boys and girls with ID. Boys with ID reported significantly more positive self-concept scores on the dimensions scholastic competence, athletic competence, physical appearance, and global self-worth compared to girls with ID. The remaining dimensions showed no significant differences between boys and girls with ID. The results of the independent t-tests utilising the bootstrap method are shown in Table 2.

Table 2. Independent T-test Results Examining Differences in Self-concept Between Boys and Girls with ID.

	Mean difference	SD	t	CI LL	CI UL	Cohen’s d
Scholastic competence	2.013	0.646	2.478*	0.728	3.337	.676
Social acceptance	0.934	0.889	1.037	−0.968	2.744	.283
Athletic competence	2.816	1.176	2.720*	0.625	4.927	.742
Physical appearance	3.339	1.000	3.664*	1.483	5.167	1.000
Behavioural conduct	0.910	1.164	0.729	−1.355	3.149	.199
Global self-worth	1.711	0.724	2.169*	0.196	3.149	.592

Note. Analyses were run with 1,000 bootstraps. CI = confidence interval, LL = lower limit, UL = upper limit, *p < .05.

To determine if students with ID showed age-related differences in self-concept in any of the six dimensions, a one-way ANOVA was performed using raw scores. Group 1 consisted of students aged 16 years or younger (N = 37) and group 2 comprised students aged 17 through 21 years old (N = 34). The results revealed no statistically significant differences between both groups on any of the six dimensions of self-concept, indicating that there are no age-related self-concept differences between students with ID.

In order to calculate differences in self-concept scores of students with ID and a classification of either ADHD or ASD, independent t-tests were performed. The analyses showed no statistically significant results on any of the self-concept dimensions investigated, indicating that having a formally assessed classification of ADHD or ASD does not result in differences in the self-concept of students with ID.

The percentile scores on the six dimensions of self-concept of students with ID were compared to the percentile scores of the norm group, in order to determine if students with ID differ in their self-concept from their TD peers. Comparisons between the percentile scores of students with ID with the norm group, as shown in Table 1, showed differences in self-concept scores. The results showed that boys with ID assessed their self-concept as neutral or more positively compared to TD peers in all dimensions, with the dimension social acceptance being the only dimension scoring under the 50^th percentile with a score of 47. Behavioural conduct was assessed most favourably by boys with ID. Girls with ID were less positive about their self-concept and reported negative scores on three out of the six dimensions compared to their TD peers. Athletic competence, physical appearance and behavioural conduct were assessed most negatively by girls with ID, whereas the other dimensions were scored neutral compared to TD peers.

Self-Concept of Students with SEBD

The mean raw scores of students with SEBD on each of the six dimensions of self-concept were calculated. Similarly to students with ID, students with SEBD reported high self-concept scores on global self-worth. The dimension of athletic competence was also scored highly,

whilst the dimension of behavioural conduct received the lowest mean scores. Table 3 shows an overview of the mean raw scores and percentile scores of students with SEBD on the different self-concept dimensions, specified for boys and girls.

Table 3. Mean Raw Scores and Percentile Scores of the Self-Perception Profile for Children of Students with SEBD.

	Students with SEBD
	Boys (N=91)			Girls (N=9)
	Mean	SD	Perc.	Mean	SD	Perc.
Scholastic competence	16.90	3.52	45	16.33	3.91	47
Social acceptance	17.24	4.40	36	17.67	5.63	56
Athletic competence	19.12	3.84	53	19.56	3.71	76
Physical appearance	19.32	4.19	32	17.67	6.30	40
Behavioural conduct	15.43	4.04	23	18.00	4.95	52
Global self-worth	19.43	3.44	34	17.00	6.67	21

Note. Perc. = percentile; the range of the scores of self-concept is 6–24.

The analysis performed to investigate gender-related differences revealed no statistically significant differences, indicating that boys and girls with SEBD did not differ significantly in their self-concept in any of the six dimensions. The results of the analysis investigating gender-related differences in students with SEBD can be found in Table 4.

Table 4. Independent T-test Results Examining Differences in Self-concept Between Boys and Girls with SEBD.

	Mean difference	SD	t	CI LL	CI UL	Cohen’s d
Scholastic competence	0.568	1.357	0.457	−2.069	3.463	.160
Social acceptance	−0.425	1.949	−0.269	−4.111	3.208	−.094
Athletic competence	−0.435	1.311	−0.325	−2.972	2.038	−.114
Physical appearance	1.652	2.166	0.769	−2.431	5.974	.375
Behavioural conduct	−2.571	1.631	−1.787	−5.341	0.378	−.624
Global self-worth	2.429	2.263	1.078	−1.836	6.650	.638

Note. Analyses were run with 1,000 bootstraps. CI = confidence interval, LL = lower limit, UL = upper limit.

A one-way ANOVA was performed to explore age-related differences in self-concept on the six dimensions of self-concept in students with SEBD. Statistically significant differences were found between the three age groups on the dimensions physical appearance (F(2,97) = 7.451, p = .001; ƞ² = .13) and global self-worth (F(2,97) = 2.955, p = .057; ƞ² = .06). For the dimension physical appearance, the Tukey post hoc test revealed that the raw self-concept scores of group 2 (i.e. students aged 12) were significantly lower compared to both other groups, showing that students with ID aged 12 are more negative about their physical appearance compared to students with ID that are a couple of years older or younger. The post hoc analysis showed that the significance found in the dimension global self-worth was again attributed to group 2 with a lower score compared to group 1. No statistically significant differences were found in the remaining self-concept dimensions, with significance levels ranging from p = .204 to p = .937.

To determine if there were differences in self-concept scores of students with SEBD and a formally assessed classification of either ADHD or ASD, independent t-tests were conducted. As Table 5 indicates, the analyses showed a statistically significant result with a moderate effect size for students with SEBD and ASD in the dimension of athletic competence, with these students assessing their self-concept more positively in that dimension. Students with SEBD and a formally assessed classification of ADHD did not assess their self-concept differently in any of the dimensions in comparison to students with SEBD without an ADHD classification.

Table 5. Results of the Independent T-test Examining the Differences in Self-concept Between Students with SEBD with and without ADHD or ASD.

	Mean difference	SD	t	CI LL	CI UL	Cohen’s d
ADHD
Scholastic competence	0.788	0.922	0.852	−1.018	2.623	.210
Social acceptance	−0.186	1.135	−0.160	−2.369	2.186	−.040
Athletic competence	−0.652	0.915	−0.691	−2.309	1.231	−.170
Physical appearance	−0.128	1.127	−0.113	−2.295	2.245	−.028
Behavioural conduct	1.174	1.109	1.158	−0.789	3.170	.286
Global self-worth	−0.076	0.984	−0.073	−1.964	1.759	−.018
ASD
Scholastic competence	0.910	0.910	0.986	−0.847	2.654	.243
Social acceptance	1.335	1.184	1.161	−1.054	3.684	.286
Athletic competence	2.268	0.904	2.512*	0.495	4.332	.620
Physical appearance	1.879	1.117	1.696	−0.424	4.256	.418
Behavioural conduct	−1.441	1.017	−1.429	−3.284	0.400	−.352
Global self-worth	0.959	1.037	0.935	−1.145	3.004	.231

Note. Analyses were run with 1,000 bootstraps. CI = confidence interval, LL = lower limit, UL = upper limit, *p < .05.

When compared to TD students, boys with SEBD assessed their self-concept more negatively than girls in almost all of the dimensions, except for athletic competence. Boys with SEBD were most negative about their behavioural conduct in comparison to TD peers, with a 23^rd percentile score in this dimension. Similar to boys, girls with SEBD assessed athletic competence highest and with a 76^th percentile score, this assessment is more positive than in students without SEBD. The self-concept score on global self-worth of girls with SEBD was very low in comparison to TD peers, with a percentile score of 21 (see Table 3).

Discussion

This study aimed to gain a greater insight into the self-concept of students with either ID or SEBD. The results of this study indicate that the self-concept of students with ID is positive or neutral on most dimensions in comparison to their TD peers. Remarkably, the self-concept of students with ID differs greatly depending on gender. Though boys with ID show relatively similar scores on all different dimensions of self-concept and these scores are comparable to those of their TD peers, girls with ID are distinctly more negative in their self-concept assessments than boys. Girls with ID are very negative in their assessments of their athletic competence, physical appearance and behavioural conduct compared to their peers without ID, although they score more similar to their peers on the remaining self-concept dimensions. Though this study found gender differences in students with ID on self-concept scores, literature about gender-related differences in self-concept is inconclusive (Avramidis, 2013; Douma et al., 2022; O’Byrne & Muldoon, 2017).

Students with SEBD, however, are more critical of themselves than their TD peers, especially boys with SEBD. Whereas boys with ID are quite similar in their self-concept assessments in comparison to TD peers, boys with SEBD have a very negative self-concept in the majority of dimensions, especially the dimension concerning their behaviour. These evaluations are significantly lower than those of students without a disability. Though girls with SEBD are more positive on some dimensions compared to boys with SEBD, they are also very negative about their physical appearance and both boys and girls with SEBD assess their global self-worth poorly. This latter finding in particular is cause for concern since a positive global self-concept, as mentioned earlier, is detrimental to individuals in various areas of development, such as academic performance and social inclusion.

The findings for students with ID and students with SEBD are in line with several other studies that found students with ID have a more positive self-concept and students with SEBD have a more negative self-concept (Avramidis, 2013; Douma et al., 2022). Both students with ID and students with SEBD do not report overly positive self-concept scores. Surprisingly, the self-concept scores of especially students with SEBD do not demonstrate the presence of the expected PIB, except for the elevated scores on athletic competence reported by girls with SEBD, which could be indicative of the PIB.

An explanation for the specific higher and lower assessed dimensions of self-concept for both groups of students can be found in Cooley’s Looking Glass-Self Theory (Cooley, 1902). Cooley argued that the self-concept is not only formed by social comparisons with significant others but also by the evaluations and interpretations we make about how we are perceived by those significant others. Applying this theory to the results of this study, students with SEBD often perceive negative evaluations of their behaviour by their environment. They are, therefore, more likely to develop a negative self-concept concerning their behavioural conduct. For students with ID, there is much positive emphasis from their environment on their personal, intra-individual development, which may result in more positive self-concept scores, such as for global self-worth and scholastic competence (Feng & Sass, 2013).

This study also tried to shed light on the personal characteristics of students that might be related to their self-concept. Firstly, age-related differences within each group of students were researched. For students with ID, age does not seem to be an influencing variable, though it is for students with SEBD. Specifically, students with SEBD aged 12 differ in their self-concept in the dimensions physical appearance and global self-worth. Upon closer examination, the group of girls with SEBD in the whole sample was overrepresented in the group of twelve-year-olds. This might explain the age-related results since girls tend to be more negative about their physical appearances, and subsequently, their global self-worth, since physical appearance is the strongest predictor of global self-worth (Baudson et al., 2016; Klomsten et al., 2004).

Secondly, formal assessments of disabilities, specifically ADHD and ASD, were included in the analysis in order to determine whether students with such a formal assessment see themselves differently. According to the results of this study, being formally assessed with ADHD or ASS does not result in differences in self-concept in students with ID. However, students with SEBD and a formal assessment of ASD assessed their athletic competence significantly lower than students with SEBD without ASD. This result is in line with other research indicating students with ASD report lower scores on the self-concept dimension athletic competence (Capps et al., 1995; Williamson et al., 2008). This result might also be expected since numerous studies have shown that a majority of students with ASD have problems developing motor skills, needed for athletic performances (Fournier et al., 2010). Even though one of the characteristics of students with ASD is difficulty in social interactions, surprisingly no differences were found in social self-concept between students with ID or SEBD with or without an ASD classification (American Psychiatric Association, 2013).

This study intended to provide an overview of the self-concept of students with either ID or SEBD in special education. Even though this study has provided a deeper understanding of the self-concept of students with ID or students with SEBD in special education, it was not without some practical and methodological limitations. Firstly, there were some limitations inherent to research conducted within special education settings. Boys are generally overrepresented in special education and this well-known issue was also encountered in this study. In the sample of students with ID, 75% of the participants were boys, a ratio that is reflective of the national figures of the dispersion of boys and girls in special education in the Netherlands. However, the distribution of gender among the students with SEBD was more uneven. In order to take this skewness in the data into account, analyses utilising a bootstrapping method were conducted.

Furthermore, there are no instruments available to measure self-concept in students with a disability that can be used and are appropriate for students with different types of disabilities or have norm scores for specific populations. Though there are questionnaires available specifically for students with ID, they are not suitable for students with SEBD.

The last limitation lies in the choice to omit using multi-level analyses. Multilevel models recognise the different levels within the dataset and can therefore make adjusted inferences on different levels. However, this study has chosen to include only variables on the student level and not to include variables on other levels. Therefore, a multilevel approach was not deemed fit for this study.

Conclusion and Implications

Self-concept is an important factor in human development and this study has demonstrated that it is worthwhile to further investigate the self-concept of students with ID or students with SEBD in special education. Although there is a worldwide transition towards inclusive education, many students with ID and students with SEBD still reside in segregated (special) school settings and these students must not be overlooked in scientific research.

Research into the self-concept of students with ID or students with SEBD is relatively scarce, in particular for the latter group of students. In addition, the relationship between self-concept, age and specific classifications of ADHD or ASD are often not included in self-concept research. By including these relationships in the current study, this study aimed to contribute to furthering the understanding of the self-concept of students with ID or students with SEBD.

With regard to the self-concept of these students, this study shows not all self-concept results can be described as advantageous for students with ID or students with SEBD. For two groups in particular the results deem further investigation, namely girls with ID and boys with SEBD. Both groups report self-concept scores well below average across several dimensions. Future research should therefore be aimed towards improving the self-concept of all students with special educational needs, with an added focus on these two groups of students. Furthermore, considering school experiences play a fundamental role in the development of self-concept, as established by the SCT (Festinger, 1954), it is necessary to investigate whether different educational settings have different implications on the self-concept of students with ID or students with SEBD. In addition to this recommendation, future research should explore the relationship between the social school environment and self-concept. Students compare themselves to their peers to form their self-concept and therefore the relationship between students with ID or students with SEBD and their comparison group of peers might be linked to the development of self-concept. This specific direction of future research should certainly extend to inclusive education settings, considering the inclusion of students with SEN in general education is currently globally one of the main educational developments (UNESCO, 2016) and provides students with ID or students with SEBD with a more heterogenous comparison group of peers, and, subsequently, a broader range of opportunities for social interactions and social comparisons.

This study yielded also some practical implications. To increase the self-concept of students who might be at risk of developing a negative self-concept, it is recommended educators in schools be aware of potential differences in self-concept between boys and girls and implement interventions for self-concept when students are at a young age, particularly for students with SEBD, since age was found to be an influencing variable for this group. Furthermore, the findings and existing literature (Cooley, 1902) highlight that it is important for the development of a positive self-concept to create a supportive school and classroom environment for students for them to be encouraged and stimulated to develop a positive self-concept, which then will give them the best chance for achieving positive life outcomes.

CRediT Author Statement

Ivonne Douma: Conceptualisation, Methodology, Investigation, Validation, Data curation, Formal analysis, Writing – original draft, preparation, and Writing – review & editing.

Anke de Boer: Conceptualisation, Methodology, Writing – review & editing.

Alexander Minnaert: Conceptualisation, Methodology, Writing – review & editing.

Hans Grietens: Conceptualisation, Methodology, Writing – review & editing.

Disclosure Statement

No potential conflict of interest was reported by the author(s).