Assessing Inclusive School Leadership Practices in Malaysia: Instrument Adaptation and Validation

ABSTRACT

School leadership practices that promote inclusion for students with special educational needs have received significant attention in the literature. However, few studies have examined the psychometric properties of scales to evaluate principals’ inclusive leadership practices in the Asian, specifically Malaysian, context. This study examined the dimensionality, reliability, and validity of a scale testing inclusive leadership in Malaysian schools. A total of 604 responses were collected from teachers working at 63 public primary and secondary schools in Kuala Lumpur, Malaysia. The study was conducted across two phases. In phase 1, the construct validity of the scale for Malaysian principals’ inclusive leadership practices was established using exploratory factor analysis. In phase 2, a confirmatory factor analysis was conducted to assess the dimensionality, reliability, convergent, and discriminant validity of the scale. Results supported a two-factor, 29-item scale. The scale may be utilised as a reliable and valid instrument for examining principals’ inclusive school leadership practices in the Malaysian context.

KEYWORDS:

• Inclusive school leadership
• inclusive education
• instrument validation
• psychometric analysis
• Malaysia

Introduction

The Salamanca Statement (UNESCO, 1994) is arguably the most significant international document in the field of special education. It endorsed the idea of inclusive education, which became a major global influence in subsequent years. More than 92 nations signed on to advocating equal educational opportunity and access for all students, including those with special educational needs (SENs) (UNESCO, 2020). After this, many education systems began to implement policies and programmes that enabled schools to serve all children, particularly those with SENs (Adams et al., 2020), as full inclusion is only possible if mainstream schools are capable of educating all children in their communities (Tan & Adams, 2021; UNESCO, 2020).

The Malaysian Education Act 1996 (1998) mandates children’s right to equal education without discrimination. Schools are mandated to accept students with SENs, provide inclusive educational opportunities and offer support and facilities to meet their needs (Adams et al., 2017, 2020; Jelas & Ali, 2014). The Malaysia Education Blueprint 2013–2025 (Ministry of Education, 2012) further emphasises the national commitment to an inclusive education model. Among the key transformation and changes required in the Blueprint was the commitment that, by 2025, 75% of children with special needs will be enrolled in inclusive programmes, and high-quality education will be provided to every child with SENs. At all levels of the educational system, leadership is a crucial factor in supporting inclusive school development (Adams et al., 2020; DeMatthews et al., 2020; UNESCO, 2020).

School leadership practices that promote inclusion for students with SENs have been the subject of many investigations in the literature (see Adams, 2020, Billingsley et al., 2019; DeMatthews et al., 2020). In schools, principals are instrumental in driving the culture and focusing on inclusion as well as promoting equality and equity, which are critical in promoting learning for all students and particularly for students with SENs (Swaffield & Major, 2019; UNESCO, 2017). A range of instruments with various dimensions have been used for measuring school leadership in inclusive education, depicting the inconsistencies in determining the psychometric properties of the construct (e.g. Billingsley et al., 2019; DeMatthews et al., 2021a; Swaffield & Major, 2019). However, few such studies have been conducted in the Asian, specifically Malaysian, context (Adams et al., 2023). This research aims to close this gap.

This study validated an instrument for use in measuring principals’ inclusive school leadership practices in the context of Malaysia. The research question that guided the study was as follows: what are the dimensions of inclusive school leadership practices in Malaysia? The study offers future directions for assessing school principals’ inclusive leadership practices in the given context.

Sociocultural Theory

A sociocultural perspective highlights the mutual influences between individuals and their immediate surroundings. This includes individual learning in interaction with others within learning context (Rogoff, 2003). Vygotsky concentrated on the inter-relationships that occur between individual learning and its relation to language, environmental arrangements, aspects of curriculum over time (Beneke et al., 2020; Vygotsky, 1978). For example, through school principals’ interactions with teachers and students, particularly students with SENs in their schools, principals learn and construct meanings of inclusive education. Further, principals’ knowledge, abilities and attitudes establish and sustain inclusive practices in their schools (Lyons, 2016). Cultural adaptation in the validation of an instrument through a sociocultural lens is important to ensure the validity and relevance of the instrument across diverse contexts. Therefore, applying sociocultural theory for this study is a crucial step to ensure the validity of the instrument in the context of Malaysia.

Inclusive School Leadership

Interest in the area of leadership and its effectiveness is widespread among both researchers and practicing leaders (Adams et al., 2023; Gómez-Leal et al., 2022; Sebastian et al., 2019). However, the interpretation of the term ‘leadership’ by researchers and practitioners varies greatly. Here, we consider it to be the process ‘whereby an individual person influences a group of individuals to achieve a common goal’ (Sharma & Jain, 2013, p. 310). The process of leadership involves the management of people’s emotions, thoughts, and actions to achieve organisational goals (Oskarsdottir et al., 2020). A leader can persuade others by applying leadership knowledge and skills, directing the organisation in a cohesive and coherent manner in which everyone has a chance to grow to fulfil their full potential (Uljens, 2018).

A growing body of research and educational theories offer immense support for the notion that principals’ leadership contributes to improved school performance (Sebastian et al., 2019) and student learning (Adams, 2018, Gómez-Leal et al., 2022; Adams et al., 2017; Sebastian et al., 2019) in their schools. From a wide range of global evidence (Day et al., 2016; Leithwood et al., 2006), four broad categories emerge that indicate successful school leadership practices; (1) setting direction, (2) developing people, (3) restructuring the organization, and (4) managing instructional programmes. Researchers and educators have proposed a range of other conceptual models to explain how leadership can impact student learning (Hallinger & Wang, 2015; Leithwood et al., 2020; Robinson et al., 2008).

School leadership has been shown to successfully enable inclusive education for students with SENs. Numerous studies have shown the important role of leadership in establishing successful inclusive schools (DeMatthews et al., 2020; DeMatthews, 2020; Hoppey & McLeskey, 2013; Kozleski, 2019; Waldron et al., 2011). These researchers show that leadership is crucial for the development of inclusive schools that seek to cater to the academic, social, and emotional needs of all students, including those with SENs. Principals must have certain information, abilities, and attitudes to collaborate effectively with important stakeholders and establish and sustain inclusive practices in their schools (Lyons, 2016). In the same manner, strengthening a principal’s capacity to enhance organisational performance, school climate, and teacher cooperation can produce an indirect impact on student achievement (Orphanos & Orr, 2014). Principals who are equipped with this knowledge and these competencies are better able to improve education delivery, comprehend student needs, encourage school-family collaboration, and eliminate segregated programmes (DeMatthews et al., 2020).

Oskarsdottir et al. (2020) developed a model that was specifically focused on inclusive school leadership practices. In this model, inclusive school leadership is made up of a combination of three core functions of leadership: (1) setting direction and building a vision, (2) human development, and (3) organizational development. Due to their central focus on valuing all learners in the learning community, Oskarsdottir et al. (2020) consider that these three broad areas of the inclusive school leadership model can influence and support leaders’ core functions in the school community. Similarly, Billingsley et al. (2019) suggest that inclusive leadership in schools includes four major functions: (1) creating a school-wide vision for inclusive education, (2) supporting professional learning communities, (3) redesigning schools for inclusive education, and (4) sharing leadership with others. These four functions were identified in an extensive review of recent research related to the role of the school principal in addressing the needs of students with SENs (Billingsley et al., 2019).

In this paper, we defined inclusive school leadership with reference to the commitment of principals to promoting equity and fairness in building the school community and promoting the full participation of all students. According to this definition, which is in line with the articulation of other researchers, principals should ensure the full participation and contribution of the staff to create and sustain the learning community and support everyone to reach their maximum potential. This is supported by DeMatthews (2015), for whom principals are responsible for establishing a welcoming environment for all children, selecting faculty and staff that uphold inclusive beliefs and practices, formalising structures that allow teachers to use data for tracking student development and fix problems and allocating resources in the most effective and flexible way possible, as well as Oskarsdottir et al. (2019).

The School As an Inclusive Community

The idea of the inclusive school is rooted in the concept of inclusive education, although some consider it to be an ideological concept (Allan, 2014). In inclusive education, regardless of the challenges, students are enrolled in age-appropriate general education classes, where they can ideally receive excellent instruction and support commensurate with their abilities to help them succeed in the core curriculum (Oskarsdottir et al., 2019). Leaders of inclusive schools need to ensure that their schools and classrooms operate on the understanding that students with SENs are as fundamentally competent as their abled peers. Overarching this belief, parameters need to be established by principals who commit to developing inclusive communities by emphasising non-discrimination and preventing exclusion on the basis of disability (Billingsley et al., 2019; Kefallinou et al., 2020). The underpinning of inclusive leadership therefore resides in enhancing involvement and representation among all teachers, administrators, and students, as well as in the entirety of the school (Moya et al., 2020).

As an inclusive community, a school respects the right of all learners to quality education (Kefallinou et al., 2020). It focuses on increasing participation for all learners, creating systems that value all individuals equally, and promoting equity, compassion, human rights, and respect (DeMatthews et al., 2020). In a school community, the principal provides leadership in assessing curricular materials, staffing, and resource allocation, reconfiguring resources as needed (Causton & Theoharis, 2014; Lyons, 2016). Likewise, the principal shows commitment to inclusivity though discussions about the meaning of diversity, understanding of and beliefs related to dis/ability, and the values associated with the inclusive approach (Causton-Theoharis et al., 2011; Lyons, 2016). Principals also establish a clear direction for building and sustaining the learning community (Hoppey & McLeskey, 2013; Oskarsdottir et al., 2019). School leaders who promote inclusivity will re-examine priorities and make informed decisions based on the evidence (Billingsley et al., 2018a; DeMatthews et al., 2020; Robinson et al., 2015). All of these attributes resonate with consistent and interrelated leadership practices that are evident in a school that forms an inclusive community.

Management of Teaching and Learning Process

Segregated schools have a curriculum, teaching and evaluation methods, facilities, and specialised teachers and staff, based on the type of disabilities experienced by their students (Kassah et al., 2018; Nag et al., 2021; Robiyansah et al., 2020). Unlike a segregated setup, inclusive schools ensure that students with SENs are educated in mainstream classes together with their abled peers (Kefallinou et al., 2020; Robiyansah et al., 2020; Sakiz, 2017). School principals and teachers collaborate to ensure that all students learn successfully, including students with SENs. Teachers, therefore, need to be equipped with the requisite knowledge, skills, and attitudes to teach students with SENs in mainstream classrooms (Adams et al., 2021, Moberg et al., 2020; Saloviita, 2020).

To manage teaching and learning among students with SENs in inclusive schools, principals must provide a shared vision and support systems that will promote inclusive practices (UNESCO, 1994). Strong leadership, driven by a clear vision of inclusion, plays a major role in envisioning, evaluating, and strengthening family – school – community partnerships (Kefallinou et al., 2020). Likewise, increasing instructional awareness, holding discussions on social inclusion, and fostering participation within and beyond the school community are essential components in making teachers and the school community sufficiently competent and motivated to sustain an inclusive effort (Gross et al., 2015; Poulos et al., 2014).

Further, inclusive education focuses on the preparation of high-quality individualised educational programs (IEPs) to meet the learning needs of students with SENs in mainstream classroom settings (Sakiz, 2017; Timothy & Agbenyega, 2018). In developing IEPs, a team of educators and classroom assistants sets goals and formulates the classroom instructions that students with SENs will receive (Sakiz, 2017; Sharma et al., 2016). Then, parents are notified of their child’s learning progress (Adams et al., 2016; Muega, 2016). However, the adoption of this type of personalised approach to promote teaching and learning requires principals to ensure adequate infrastructure and resources to facilitate the learning process of each individual student in the classroom (Kefallinou et al., 2020; Rowe et al., 2012).

The preceding review of the literature indicates that two components, namely, (i) the school as an inclusive community and (ii) the management of teaching-learning processes be crucial components in the measure of a principal’s inclusive school leadership practices. Hence, in this study, principals; inclusive school leadership practices are broadly defined as the principal’s commitment to promoting equity and fairness for building the school community and providing a shared vision of inclusion in education.

Methodology

Data Collection

Data for this study were drawn from a survey on inclusive school leadership practices carried out in Malaysia. A total of 604 responses were collected from teachers at 63 public primary and secondary schools in Kuala Lumpur, with a response rate of 73%. To ensure adherence to ethical standards, permission to collect data was sought from the Ministry of Education of Malaysia before contacting the teachers for data collection. All participation in this study was voluntary, and the anonymity and confidentiality of respondents were guaranteed during the process of data collection.

Participant Characteristics

Of the 604 respondents, 369 (61.1%) were teaching at primary schools and 235 (38.9%) were at secondary schools. The majority of respondents, 495 (82%), were female, and 109 (18%) were male teachers. Most (538) teachers had a bachelor’s degree (89.1%), followed by 51 with a master’s degree (8.4%) and 15 with a diploma certificate (2.5%). All had experience teaching students with SENs. Their experience ranged across less than 5 years (119 teachers), 5 to 10 years (206), 11 to 15 years (178), 16 to 20 years (48), and more than 20 years (53). Table 1 presents the demographic breakdown of the participants.

Table 1. Demographic profile of respondents.

Variable	Category	Number	%
School type	Primary school	369	61.1
	Secondary school	235	38.9
Gender	Male	109	18.0
	Female	495	82.0
Educational qualification	Diploma	15	2.5
	Bachelor’s degree	538	89.1
	Master’s degree	51	8.4
Teaching experience (SEN)	<5 years	119	19.7
	5–10 years	206	34.1
	11–15 years	178	29.5
	16–20 years	48	7.9
	>20 years	53	8.8

Instrument

The instrument used in this study was adapted from the inclusive leadership in schools (LEI-Q) survey developed by Moya et al. (2020). This instrument consists of 40 items and is used to capture the perception of teachers regarding their principal’s inclusive school leadership practices. These items test two constructs (1) The school as an inclusive community, and (2) Management of the teaching-learning processes; they were to be rated across four response options, from (Not implemented) through 2 (Partially implemented), 3 (Substantially implemented), and 4 (Fully implemented). The reliability of the two scales were α = 0.971 and α = 0.975, respectively.

We first reviewed the two constructs and the 40 items in terms of the suitability, clarity, and interpretation for the Malaysian context. The review panel consisted of three lecturers in educational management and leadership from two local public universities, three researchers of the current study, and one doctorate student who was studying inclusive school leadership in primary schools, one school head teacher, and one teacher. The panel agreed that the two constructs were multidimensional, and its 32 items were found relevant to the Malaysian principal’s inclusive school leadership practices. However, minor adjustments were made to the terminologies in the items for better alignment with the local context. Notably, 8 items were excluded from the original 40-item scale developed by Moya et al. (2020). The decision to remove these items was not arbitrary; rather, it emanated from a rigorous consideration of their applicability within the Malaysian cultural and educational context. For example, items such as ‘enables the different members of the educational community to participate in the evaluation of management tasks’, ‘establishes sanctions for the use of symbols and actions that promote exclusion’ and ‘establishes mechanisms to promote the participation of students in the regulation of conflicts that arise in the school environment’ were found to be less relevant due to specific cultural nuances and contextual differences.

Specifically, within the Malaysian cultural context, how people respect authority and follow clear structures is important. In our culture, openly judging how leaders manage tasks might be seen as questioning their authority. Also, giving punishments for symbolic actions may clash with the prevailing cultural emphasis on harmony. Likewise, allowing students to take part in solving conflicts, while a valuable practice in other cultural settings, might be seen differently here where teachers usually handle such matters without student involvement. Thus, the adaptation of this scale was guided by a commitment to ensuring the scale’s sensitivity to the unique characteristics of Malaysian schools, thus contributing to the validity and reliability of our assessment.

The final questionnaire consisted of two parts. The first part collected data on the teachers’ demographic information, including school type, gender, educational qualification, and teaching experience in SEN. The second part consisted of the 32 items based on two sub-constructs: “(1) The school as an inclusive community, in 12 items, and (2) Management of teaching-learning processes, in 20 items. A 5-point Likert-type scale was used, ranging from 1 (strongly disagree) to 5 (strongly agree). A sample item from the first dimension follows: ‘My principal promotes initiatives that foster the participation of school community in the educational process’, and a sample item from the second dimension follows: ‘My principal encourages teachers to communicate situations of discrimination that occurs in the school’.

Data Analysis

A variety of procedures can be used for tool validation, but the current investigation engaged the three stages of (i) item development, (ii) scale development, and (iii) scale evaluation, following the suggestion of Boateng et al. (2018). For the statistical analysis, this study employed both an exploratory factor analysis (EFA) and a confirmatory factor analysis (CFA). However, following the strong recommendation not to conduct both EFA and CFA on the same data (Ramayah et al., 2018), the sample was split into two data sets to enable one to be used for EFA and the other for CFA. Given that there were 32 items in the questionnaire, a minimum sample of 200 is sufficient for an EFA, based on sample recommendations ranging from 3 to 20 responses per variable by Mundfrom et al. (2005). The random split feature in Statistical Package for Social Sciences (SPSS) 25 was employed to select the sample from the dataset. The remaining sample of 404 responses was adequate for conducting a CFA (Kline, 2015; Muthén & Muthén, 2002) and was so used.

Data Cleaning and Preparation

Data cleaning and preparation were the first actions conducted in examining the dataset. In this stage, suspicious and inconsistent responses were assessed, using the straight lining method (Hair et al., 2010). No critical issues were identified at this stage. However, in seven cases, teachers responded that they had not taught SEN classes. These cases were omitted, leaving a final sample size of 604. The dataset was further explored using minimum and maximum values, means, and standard deviation, and its distribution was examined through checking univariate skewness and kurtosis. The results of these analyses are displayed in Table 2.

Table 2. Descriptive statistics for the questionnaire items.

Item code	Sample	Range	Minimum	Maximum	Mean	SD	Skewness	Kurtosis
IL_SIC1	604	4.00	1.00	5.00	3.972	0.782	−0.410	−0.136
IL_SIC2	604	4.00	1.00	5.00	3.859	0.828	−0.508	0.206
IL_SIC3	604	4.00	1.00	5.00	3.917	0.790	−0.319	−0.181
IL_SIC4	604	3.00	2.00	5.00	3.964	0.797	−0.289	−0.592
IL_SIC5	604	4.00	1.00	5.00	3.932	0.811	−0.363	−0.235
IL_SIC6	604	4.00	1.00	5.00	3.826	0.843	−0.461	0.050
IL_SIC7	604	4.00	1.00	5.00	3.906	0.854	−0.524	0.046
IL_SIC8	604	4.00	1.00	5.00	3.727	0.879	−0.379	−0.092
IL_SIC9	604	4.00	1.00	5.00	3.538	0.916	−0.256	−0.263
IL_SIC10	604	4.00	1.00	5.00	3.849	0.844	−0.374	−0.190
IL_SIC11	604	4.00	1.00	5.00	3.896	0.831	−0.481	0.112
IL_SIC12	604	4.00	1.00	5.00	3.868	0.814	−0.364	−0.063
IL_MTL1	604	4.00	1.00	5.00	3.856	0.784	−0.260	−0.276
IL_MTL2	604	4.00	1.00	5.00	3.778	0.839	−0.341	−0.079
IL_MTL3	604	4.00	1.00	5.00	3.858	0.828	−0.326	−0.283
IL_MTL4	604	4.00	1.00	5.00	3.742	0.854	−0.376	0.046
IL_MTL5	604	4.00	1.00	5.00	3.796	0.866	−0.456	0.142
IL_MTL6	604	4.00	1.00	5.00	3.901	0.817	−0.384	−0.173
IL_MTL7	604	4.00	1.00	5.00	3.632	0.872	−0.403	0.061
IL_MTL8	604	4.00	1.00	5.00	3.828	0.808	−0.414	0.086
IL_MTL9	604	4.00	1.00	5.00	3.874	0.801	−0.451	0.269
IL_MTL10	604	4.00	1.00	5.00	3.871	0.808	−0.479	0.273
IL_MTL11	604	4.00	1.00	5.00	3.853	0.818	−0.287	−0.385
IL_MTL12	604	4.00	1.00	5.00	3.786	0.863	−0.582	0.399
IL_MTL13	604	4.00	1.00	5.00	3.843	0.850	−0.473	0.021
IL_MTL14	604	4.00	1.00	5.00	3.861	0.829	−0.314	−0.227
IL_MTL15	604	4.00	1.00	5.00	3.906	0.798	−0.341	−0.163
IL_MTL16	604	4.00	1.00	5.00	3.886	0.782	−0.277	−0.281
IL_MTL17	604	4.00	1.00	5.00	3.922	0.828	−0.417	−0.200
IL_MTL18	604	4.00	1.00	5.00	3.914	0.808	−0.334	−0.341
IL_MTL19	604	4.00	1.00	5.00	3.919	0.778	−0.346	−0.168
IL_MTL20	604	4.00	1.00	5.00	3.886	0.793	−0.354	−0.182

As noted in Table 2, the responses are generally evenly spread out across the scale, as indicated by the range and minimum and maximum values. With the exception of the item IL_SIC4, the items were assigned both maximum and minimum scores by the respondents. The means for the items seem to be slightly on the higher end of the scale (between 3.538 and 3.972) while the standard deviations (SDs) indicate no extraordinary deviations of data from the mean (SDs between 0.778 and 0.916). For the distribution patterns of the data, the recommendations of Curran et al. (1996) were used, such that skewness in the range of −2 to +2 and kurtosis in the range of −7 to +7 are considered not to cause adverse effects on the results of subsequent analyses. The statistics for skewness and kurtosis are thus within the acceptable range.

Findings

Exploratory Factory Analysis (EFA)

This study used specific parameters for performing and assessing EFA. Maximum likelihood with varimax rotation were employed as the method for factor extraction, following the prevailing recommendations (Costello & Osborne, 2005; Osborne, 2014). Sample adequacy was assessed using the Kaiser – Meyer – Olkin (KMO) measure, in which a value greater than or equal to 0.90 was expected (Pallant, 2007). Finally, factors were identified based both on eigenvalues greater than one and on observation of the scree plot (Cattell, 1966).

Examination of the SPSS output for the EFA indicates that the sample was statistically adequate for the analysis, as the KMO value was 0.96. Further, using the scree plot shown in Figure 1 and eigenvalues, the items developed for measuring inclusive school leadership in the current study could be grouped into two factors.

Figure 1. Scree plot.

Table 3 shows the rotated factor matrix, in which factor loadings greater than or equal to.50 are displayed. As seen in the table, the majority of items are clearly loaded onto one of the factors. Six items are cross-loaded; these are put under the factor with the higher loading, as indicated in italics. Accordingly, for the results of the EFA, factor 1 (school inclusive community) has 12 items, while factor 2 (managing teaching and learning) consists of 20 items. The two factors explained 77.52% of the variance in inclusive school leadership. As such, this is considered as a high-potential factor structure for measuring inclusive school leadership. Consequently, CFA proceeded with these factors and their associated items to examine their validity and reliability.

Table 3. Rotated factor matrix for two factors.

Item code	Item description	Factor
		1	2
IL_SIC1	My principal promotes initiatives that foster the participation of school community in the educational process		0.774
IL_SIC2	My principal establishes plans of action to promote student inclusivity, developed in collaboration with the school community		0.805
IL_SIC3	My principal promotes continuous collaboration to strengthen the school-industry relationship		0.811
IL_SIC4	My principal ensures adequate infrastructure and resources to cater for all members of the school community		0.764
IL_SIC5	My principal promotes collaboration with other schools to learn and share experiences		0.823
IL_SIC6	My principal initiates discussions on social inclusion topics among the school community		0.767
IL_SIC7	My principal promotes participation in programs on inclusive education organized by other schools or organizations in the community		0.792
IL_SIC8	My principal organizes programs to inform the school community on the importance and benefits of inclusion		0.800
IL_SIC9	My principal initiates educational activities outside the school to increase awareness on inclusion practices		0.636
IL_SIC10	My principal promotes social inclusion activities that foster harmony within the school community		0.732
IL_SIC11	My principal collects information on the needs of teachers, students, and other school staff	0.512	0.733
IL_SIC12	My principal promotes a shared vision on inclusive education among the school community	0.516	0.743
IL_MTL1	My principal encourages students to express their opinion freely (e.g. their educational process, learning, etc.)	0.720	0.501
IL_MTL2	My principal encourages action-research projects among teachers in order to guide inclusion processes	0.714
IL_MTL3	My principal encourages activities (seminars, courses, workshops etc.) to promote inclusivity in the school	0.719
IL_MTL4	My principal establishes protocols to address conflicts among the school community (e.g. dialogs)	0.708
IL_MTL5	My principal develops educational programs to prevent discrimination among students	0.666	0.515
IL_MTL6	My principal provides opportunities for all members of the school community to participate in decision-making	0.683	0.553
IL_MTL7	My principal conducts teacher orientation on the inclusive education policy	0.603	0.544
IL_MTL8	My principal conducts activities that promote mutual understanding among the students	0.735
IL_MTL9	My principal facilitates collaboration among teachers to improve their inclusive teaching strategies	0.784
IL_MTL10	My principal encourages teachers’ voices and views on student inclusivity	0.793
IL_MTL11	My principal promotes teachers’ reflection on inclusive educational practices in the school	0.783
IL_MTL12	My principal encourages teachers to communicate situations of discrimination that occurs in the school	0.666
IL_MTL13	My principal encourages teachers to have high expectations on all students	0.771
IL_MTL14	My principal ensures the planning of lessons is done with an emphasis on promoting student inclusion	0.803
IL_MTL15	My principal promotes a flexible curriculum for students with different learning needs	0.792
IL_MTL16	My principal promotes an evaluation of curricular materials to ensure the inclusion of all students	0.797
IL_MTL17	My principal ensures that all students are represented in the curricula of the school	0.821
IL_MTL18	My principal promotes activities that enhance empathy among students in the school	0.861
IL_MTL19	My principal ensures that teachers set appropriate criteria to assess students with different learning needs	0.838
IL_MTL20	My principal promotes the evaluation of teaching practices to foster the inclusion of diverse students	0.833

Extraction method: maximum likelihood, rotation method: varimax rotation converged in three iterations, variance explained: 77.52%

Confirmatory Factor Analysis (CFA)

As noted, CFA procedures were conducted for the 404 samples with the use of AMOS 23. For CFA, three types of indices (residual-based, parsimonious, and incremental) along with composite reliability (CR) and average variance extracted (AVE) were used in assessing the results. Following recommendations from researchers of structural equation modelling (see Schermelleh-Engel et al., 2003; Thakkar, 2020), and chi-square (x²) was not used as a criterion for accepting or rejecting the model. Table 4 summarises the criteria used in this study to evaluate the model fit and to assess validity and reliability.

Table 4. Criteria for model fit indices.

Type of fit	Index	Criteria	Interpretation
Residual-based	RMSEA	≤0.05 (Browne & Cudeck, 1993)	good
		≤0.08 (Maccallum et al., 1996)	fair
		≤0.1 (Maccallum et al., 1996)	mediocre
		>0.1 (Maccallum et al., 1996)	poor
	SRMR	≤0.05 (Byrne, 2016)	good
		≤0.08 (Byrne, 2016)	reasonable
Parsimonious	CMIN/DF	≤5 (Wheaton et al., 1977)	reasonable
		≤3 (Byrne, 2016)	reasonable
Incremental	CFI	≥0.90 (Byrne, 2016)	cut-off
Reliability	CR	0.7 ≤ CR < 0.9 0.9 ≤ CR ≤ 0.95 >0.95 (Hair et al., 2017)	satisfactory acceptable undesirable
Validity	AVE	≥0.50 (Hair et al., 2017)	cut-off

For the initial CFA model (model A, Figure 2), despite significant factor loadings (that is, loadings ≥ .40) for the items, none of the fit indices were within the pre-defined cut-off values. Thus, model A did not have acceptable fit to the data per the indices in Table 1.

Figure 2. CFA model A.

Consequently, the model was modified by drawing co-variances between errors. Moreover, only one item (IL_MTL3) was deleted during the modification. Figure 3 (model B) shows the model after the minimum necessary modifications had been made.

Figure 3. CFA model B.

From the fit indices (RMSEA = 0.099; SRMR = 0.037; CMIN/DF = 4.965; CFI = 0.909), the model has reasonable fit to data. Subsequently, the convergent and discriminant validity of the model was assessed. Convergent validity was assessed using AVE (≥0.50) and CR (≥0.70), while discriminant validity was confirmed by ensuring that correlation between the factors is lower than the average variance explained (square root of AVE) due to the factors.

Table 5 shows the CR, AVE, and correlations for the constructs of model B. From the results presented in Table 5, model B passes the threshold for convergent validity, as both CR and AVE are above the minimum cut-off values. However, the model did not achieve discriminant validity, as the correlation between the factors (0.881) was greater than the variance explained by the factors (shown in the √AVE column, Table 4). Thus, model B did not pass the assessment of validity. Subsequently, modifications were obtained by deleting items (item IL_MTL5 and IL_MTL7), which produced some amount of cross-loading in the initial stages of the factor exploration.

Table 5. Assessment of validity: model B.

				Factor Correlations
Constructs	CR	AVE	√AVE	SIC
SIC	0.976	0.772	0.879
MTL	0.985	0.770	0.877	0.881

Figure 4 shows the CFA model after the necessary modifications have been made, as stated above. Based on the obtained model fit indices (RMSEA = 0.099; SRMR = 0.037; CMIN/DF = 4.963 CFI = 0.915), the model provides reasonable fit to data. Subsequently, the convergent and discriminant validity of the model were assessed, the results of which are shown in Table 6.

Figure 4. CFA model C.

Table 6. Assessment of validity: model C.

				Factor Correlations
Constructs	CR	AVE	√AVE	SIC
SIC	0.976	0.772	0.879
MTL	0.983	0.775	0.880	0.874

As seen in Table 6, the CR and AVE for model C were both above thresholds for convergent validity. Further, the square root for AVE (√AVE) for each factor is greater than the correlation between the factors (0.874). Thus, model C achieved discriminate validity as well. From this finding, it is evident that school inclusive community (SIC) and managing teaching and learning (MTL) are distinct factors that can be used together to measure inclusive school leadership in the given context.

Discussion

This study tested the dimensionality and assessed the reliability and validity of an instrument to measure inclusive leadership in schools in the context of Malaysia. EFA produced two factors for use in measuring inclusive school leadership. It was found that the modified CFA model (model C) achieved reasonable fit and both convergent and discriminant validity. The findings confirmed that SIC and MTL are acceptable dimensions for use in measuring inclusive school leadership in the Malaysian context.

The identification of inclusive school leadership factors in this study offers crucial insights for school leaders seeking to implement inclusive education, with a particular emphasis on understanding the cultural context. One of the primary aspects of inclusive school leadership is transforming the school into an inclusive community (Moya et al., 2020). This study corroborates the core leadership practices identified by D. DeMatthews (2020) for creating effective inclusive schools. These are (i) creating a culture of change-oriented collaboration, (ii) planning and evaluating, (iii) building capacity, and (iv) developing/revising plans. The items in this dimension also provide for explicit leadership practices that school leaders can enact for reforming schools into inclusive communities.

In the Malaysian context, where hierarchical structures and respect for authority are significant cultural factors, the role of principals in creating and promoting inclusive schools becomes more complex. Our instrument, validated for the Malaysian context, becomes a valuable tool for assessing daily inclusive leadership practices, prompting a reconsideration of principals’ roles in implementing inclusive education through a whole-school approach. Principals’ engagement in creating and promoting inclusive schools is essential for the successful inclusion of students with special needs in mainstream classes, aligning with inclusive educational policies and practices (Khaleel et al., 2021).

Inclusive school leadership also encompasses the management of teaching and learning (MTL), serving as the core aspect of instructional leadership. Notably, Lambrecht et al. (2022) found that instructional leadership has a direct on the implementation of an Individualized Education Program (IEP). Additionally, principals are required to utilise available autonomy accorded to them to adjust the curriculum and assessment methods, addressing the diverse needs of students with SENs. Additionally, principals must influence teaching and learning through the development of a mechanism that is supportive for all students. Thus, a key responsibility that principals have is supporting the development and utilisation of inclusive pedagogy in mainstream classes (Florian & Beaton, 2018). This study raises the possibility that school principals could engage in MTL to a greater degree to increase the level of achievement of all students, particularly among students with SENs.

Finally, the instrument adopted was found to be a valid measure of principals’ inclusive school leadership practices in the Malaysian context. it had a two-factor structure, composed of 29 items with 12 items for SIC and 17 items for MTL. The item analysis in this study showed that the reliabilities of the two factors were acceptable, where the CRs for SIC and MTL were 0.976 and 0.983, respectively. These findings not only validate the instrument but also underscore its relevance and applicability within a unique cultural context, providing a lens through which to understand and enhance inclusive school leadership practices in Malaysia.

Limitations and Recommendations for Future Research

Although the scale for principals’ inclusive school leadership practices was validated in the current study, our sample only included teachers in public primary and secondary schools in Kuala Lumpur. Future studies could further validate the scale with teacher samples in other states in Malaysia, especially in the Borneo States, where schools are faced with added challenges, such inadequate educational resources or facilities and poor teacher quality.

Conclusion and Implications

As Malaysia remains committed to an inclusive education model aiming to have 75% of children with special needs enrolled in inclusive programmes by 2025, with a focus on providing high-quality education for every child (Ministry of Education, 2012), school principals are key figures in transacting this progress. Thus, the need for assessing a principal’s inclusive leadership practices has become more significant, necessitating the development of a reliable measuring tool. Hence, the primary objective of this study was to validate an instrument that could be used to assess a principal’s inclusive school leadership practices in the Malaysian context. The results of the validation process revealed that the measurement of a principal’s inclusive school leadership practices is composed of two dimensions (i) SIC and (ii) MTL. Building upon this adapted and validated instrument, this scale holds the potential to serve as a reliable and valid tool for examining principals’ inclusive school leadership practices not only in other Malaysian states but also in various Asian contexts.

Acknowledgments

This research is part of the Fundamental Research Grant Scheme FRGS/1/2020/SSI0/UM/02/5 financed by the Ministry of Higher Education, Malaysia.

Disclosure Statement

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