Exploring in-service teachers’ attitudes, teacher self-efficacy, and influential factors for implementing inclusive practices in Hong Kong classroom

Abstract

This study examines in-service teachers’ attitudes and teacher self-efficacy in implementing inclusive practices in Hong Kong classrooms. The predictors of teachers’ attitudes and teacher self-efficacy for inclusive practices were also investigated. Participants of the study comprised 1110 primary and secondary in-service teachers. Results reveal that Hong Kong in-service teachers had slightly positive attitudes towards inclusion. Teacher self-efficacy for inclusive practices was moderate and had a positive impact on teachers’ attitudes. Prior experience in teaching students with disabilities, confidence in teaching students with disabilities, level of training, years of teaching experience, self-efficacy in inclusive instruction and collaboration were significant predictors of teachers’ attitudes towards inclusive education. Knowledge of legislation and policies, confidence in teaching students with disabilities and teachers’ beliefs and their feelings were determined as the significant predictors of teacher self-efficacy. Recommendations and the limitations of the findings are discussed in light of these results.

Keywords:

• Inclusive education
• special education
• in-service teachers
• attitudes
• self-efficacy

Introduction

All children have the right to an education, and in this regard, international treaties have supported the implementation of inclusive education. General Comment No. 4 on the United Nations Convention on the Rights of Persons with Disabilities (CRPD) Article 24: Right to Inclusive Education in 2016 emphasises the rights of people with disabilities, such as the right to inclusive education at all levels (United Nations, 2016). This General Comment defines inclusive education as ‘central to achieving high-quality education for all learners, including those with disabilities, and for the development of inclusive, peaceful and fair societies’ (United Nations, 2016, p. 1), and clarifies the differences between segregation, integration and inclusion. Although it is internationally accepted that inclusive education provides all students with equal opportunity and access at all education levels in developed and developing countries, the knowledge and the practice of inclusive education vary substantially by nation (Forlin & Loreman, 2014). In March 2007, Hong Kong signed the CRPD, and in 2008, it ratified the Convention. Since that time, Hong Kong has taken important steps in the development of legislation and policies that foster the implementation of inclusive education. In this study, Hong Kong in-service teachers’ attitudes and teacher self-efficacy regarding the implementation of inclusive practices, as well as the factors that influence these elements, are examined. Moreover, this study investigates whether teachers’ attitudes and teacher self-efficacy are related. Teachers’ attitudes towards inclusive education and teacher self-efficacy influence how they implement instructional practices and their willingness to collaborate with professionals and their peers. Understanding teachers’ attitudes and teacher self-efficacy can help identify potential barriers and facilitators, as well as provide effective support for including students with disabilities in their classrooms. Overall, this study aims to contribute to the knowledge about teachers’ attitudes and teacher self-efficacy for inclusive education practices in the Hong Kong classroom setting.

Inclusive education in Hong Kong

In Hong Kong, following international practices that focused on the implementation of inclusive education reforms, the inclusion of students with disabilities began in the 1970s and became mandatory in 1997 (Education Bureau, 2014). In 1996, the Hong Kong government legislated the Disability Discrimination Ordinance (DDO) to promote inclusive policies within community systems, fostering diversity amongst people. Furthermore, the Equal Opportunities Commission issued the Code of Practice on Education to provide educators with practical advice regarding modifications and accommodations for students with disabilities in accordance with the DDO’s rules (Equal Opportunities Commission, 2001).

The Hong Kong Government adopts a dual-track mode in providing education to students with disabilities in primary and secondary schools (Education Bureau, 2014). Students with more severe (e.g. intelligent quotients fall below 50) or multiple disabilities are educated in special schools for intensive support upon the referral of professionals and with the parents’ consent. Students with other disabilities are encouraged to attend regular schools with typically developing peers (Education Bureau, 2014), and their parents retain the right to place them in special schools if their intelligence quotients are below 70. Students with other disabilities, as defined by the Education Bureau, include specific learning difficulties, intellectual disability, autism spectrum disorder, attention-deficit/hyperactivity disorder, physical disability, visual impairment, hearing impairment, speech and language impairment, or mental illness (Education Bureau, 2021b).

In 1997, the Hong Kong government promoted a whole school approach (WSA) that aims to improve the quality and standards of education for all students with diverse learning needs in regular schools (Education Bureau, 2014). To enhance the preparation of teachers for implementing inclusive education, the Hong Kong government has, since 2007, expressed its commitment to providing intensive professional learning programmes to in-service teachers. These programmes primarily focus on instructing teachers how to identify and meet the needs of all students, as well as how to implement WSA and inclusive teaching practices to support students in the classroom. Consisting of three levels – Basic, Advanced and Thematic – they range from a 30-hour theory-based introductory course to a 72-hour course with theory-based training and supervised practicums. Each school is required to have 80%, 20% and 25% of its teachers complete the Basic, Advanced and Thematic programmes, respectively (Education Bureau, 2021a).

Professional learning programmes are particularly important for in-service teachers who have not received training in teaching students with disabilities. Since the 2004/2005 academic year, qualified teachers in Hong Kong have been required to hold a bachelor’s degree. Currently, pre-service teachers are required to complete a module on inclusive education as part of their initial teacher training course (Legislative Council Secretariat, 2019). Hence, newly qualified teachers have a foundation of skills and knowledge related to teaching students with disabilities before they enter the workforce. Some teachers with more than 20 years of teaching experience may only hold a teacher’s certificate. Thus, professional learning programmes can help bridge the gap in training for teachers who did not receive inclusive education training.

Theoretical framework

Teachers require specific skills and knowledge to effectively implement inclusive practices that support all students. However, if individuals solely hold the beliefs in their capability to perform a specific behaviour, it is not sufficient to generate the intentions to engage in that behaviour (Eagly & Chaiken, 1993). Their attitudes have an impact on behaviour through the influence on intentions (Ajzen & Fishbein, 2005), which reflect the individuals’ motivation to perform the behaviour (Eagly & Chaiken, 1993). According to Ajzen (1991), the Theory of Planned Behaviour (TPB) explains that an individual’s actual behaviour can be most accurately predicted by their intention, which depends on three interrelated variables: attitude, subjective norms, and perceived behaviour control. Attitude refers to one’s positive or negative beliefs about the behaviour. Subjective norms refer to the views of important people or social pressure. Perceived behavioural control refers to beliefs regarding the difficulty of a specific task and to what extent individuals feel they can control it (Ajzen, 1991).

In this study, within this construct, perceived behaviour control is considered compatible with perceived self-efficacy, as described by Bandura (Opoku et al., 2021). Perceived behaviour control and perceived self-efficacy are related, yet they possess distinct conceptual meanings. Perceived behaviour control concerns the extent of control one has over behaviour (Ajzen, 1991). Based on the TPB, individuals are more likely to engage in a behaviour when they perceive that they have a high level of control over it. Self-efficacy primarily focuses more on individuals’ beliefs in their ability to execute the behaviour or specific tasks (Bandura, 1997). Self-efficacy beliefs can be influenced by factors including level of training, past experiences, and knowledge (Chow, 2024). Thus, there is a difference between the concepts of perceived behaviour control and perceived self-efficacy. However, both concepts share a common emphasis on one’s perceptions and beliefs about their capability to perform specific behaviours. Although the primary focus of this study is not on teachers’ intentions or their actual behaviour, understanding the underlying factors that shape teachers’ attitudes and teacher self-efficacy is crucial for promoting the implementation of inclusive practices.

Teachers’ attitudes towards inclusive education

Teachers play a crucial role in the implementation of inclusive education. Teachers’ attitudes are key to promoting the successful inclusion of students with disabilities in the classroom. Previous studies (e.g. Dorji et al., 2021) stressed that teachers’ positive attitudes towards inclusive education can encourage the implementation of inclusive teaching strategies and the execution of its reforms. However, teachers’ unfavourable attitudes or lack of support for underlying concepts of inclusion would challenge the implementation and operation of an inclusive education policy (Yada et al., 2022). Research (e.g. Yada & Savolainen, 2017) has shown that teachers in Asian countries had neutral or less-than-favourable attitudes towards inclusion. In a study conducted in Hong Kong about the educators’ attitudes towards inclusive education, their overall attitudes were found to be slightly negative (Sharma & Chow, 2008). Individuals’ explicit attitudes (i.e. positive, neutral, or negative) can be influenced by the cognitive component, which includes the knowledge and beliefs about it (Gawronski & Bodenhausen, 2007). This means that a person’s explicit attitudes can be shaped by their beliefs and understanding about a particular topic or concept. In the context of educators’ attitudes towards inclusive education, their certain beliefs or limited knowledge of inclusive education could negatively affect their attitudes towards it.

Several factors have been found to influence teachers’ attitudes towards inclusion (De Boer et al., 2011). Avramidis and Norwich (2002) found that the nature and severity of students’ disabilities influenced teachers’ attitudes towards inclusion, and that elements of the education environment, such as physical and human support, were correlated with teachers’ attitudes. In addition, teachers’ attitudes tend to be influenced by their teacher-related variables, including gender, age, type of school, and years of teaching experience. For example, teachers who have more years of teaching experience tend to have negative attitudes towards inclusion (MacFarlane & Woolfson, 2013). Other research also revealed that younger teachers with fewer years of teaching experience would likely include students with disabilities in the classroom than those teachers with more years of experience (Savolainen et al., 2012). Teachers’ gender has a mixed influence on teachers’ attitudes, as expressed in the literature. Several studies reported that female teachers were found to have more positive attitudes than male teachers (Boyle et al., 2013), but other studies obtained an opposite result (Dorji et al., 2021). Furthermore, much research has emphasised the value and significance of training in the development of teachers’ positive attitudes towards inclusion (Avramidis & Kalyva, 2007; Boyle et al., 2013). Teachers who received training regarding teaching students with disabilities in the classroom have more favourable attitudes towards inclusion than those who did not (MacFarlane & Woolfson, 2013).

Teacher self-efficacy for inclusive practices

Teacher self-efficacy has emerged as one of the most important factors for successfully implementing inclusive practices (Kiel et al., 2020). Researchers (e.g. Yada & Savolainen, 2017) worldwide are paying close attention to research on teacher self-efficacy in relation to inclusive education. Self-efficacy, which refers to individuals’ beliefs in their ability to plan and carry out a course of action to achieve specific goals, significantly impacts human accomplishment in various settings (Bandura, 1997). That is, people who have a strong sense of self-efficacy are more likely to persevere and reach their goals. In education, teacher self-efficacy refers to teachers’ level of confidence in their abilities to teach their subject to even the most challenging students (Holzberger et al., 2013). Moreover, teacher self-efficacy is associated with students’ motivation, achievement (Herman et al., 2018) and classroom management (Reinke et al., 2013). Teachers who have higher teacher self-efficacy in managing students’ behaviour in the classroom are more likely to provide successful teaching practices and achieve better students’ learning outcomes. Likewise, teacher self-efficacy impacts how teachers teach and motivate their students (Skaalvik & Skaalvik, 2007). Literature has revealed that teachers with higher sense of teacher self-efficacy tend to engage more with students and insist on helping them for longer periods (Woolfolk Hoy et al., 2009).

Past studies have investigated the factors that influence teacher self-efficacy for inclusive practices. A number of predictors were found to significantly impact teacher self-efficacy. Regarding the type of school or teaching grade, studies (e.g. San Martin et al., 2021) have shown that primary school teachers have higher self-efficacy in teaching students with disabilities than those who teach in secondary schools. Secondary school teachers reported less motivation, willingness, and readiness to manage students’ behaviour than their colleagues who taught in lower grades (Baker, 2005). Teachers who instruct younger children in primary schools have higher teacher self-efficacy for classroom management and student engagement (Klassen & Chiu, 2010). Chao et al. (2017) surveyed 347 primary and secondary school teachers in Hong Kong, and the results indicated that school type significantly influenced teacher self-efficacy in inclusive education. Other predictors that influence teacher self-efficacy for implementing inclusive practices include confidence in teaching students with disabilities (Chao et al., 2016) and knowledge of legislation and policies (Forlin et al., 2014). It is believed that teachers who have greater knowledge of the reasons behind legislation and of their own responsibilities in this regard are more likely to do so.

Relationship between attitudes towards inclusive education and self-efficacy for inclusive practices

The TPB suggest that both attitudes and perceived behavioural control (compatible with perceived self-efficacy) are important predictors of behaviour (Ajzen, 1991). It proposes that behaviour is most likely to occur when individuals have positive attitudes towards the behaviour and when they perceive a high level of self-efficacy for performing it. Attitudes and efficacy beliefs can be interconnected and influence each other. Numerous studies (e.g. San Martin et al., 2021; Savolainen et al., 2022; Yada et al., 2022) have documented that the relationship between teacher self-efficacy and attitudes towards inclusive education is positively associated. Research has shown that teacher self-efficacy significantly impacts their attitudes towards inclusive education and is considered essential in successful inclusive classrooms (Forlin et al., 2014). Savolainen et al. (2012) examined Finnish and South African in-service teachers and revealed that teacher self-efficacy, particularly self-efficacy in collaboration, is positively associated with their attitudes towards inclusive education.

In addition, it is expected that teachers with higher level of self-efficacy are anticipated to correspond to more positive attitudes, while lower levels of self-efficacy are expected to be associated with less positive attitudes (Wray et al., 2022). Savolainen et al. (2022) indicated that increased teacher self-efficacy is likely to result in more positive attitudes towards inclusive education. Other studies suggested that teachers’ attitudes are positively correlated with new skills and teacher training (MacFarlane & Woolfson, 2013) which in turn may increase and build teacher self-efficacy.

Aims of the study

Teachers are expected to meet the learning needs of all students. They adapt teaching practices, accommodate learning materials, and modify the curriculum to help students achieve their full learning potential (Forlin, 2010). The inclusion of students with disabilities relies heavily on teachers, specifically their attitudes towards and teacher self-efficacy in implementing inclusive practices (Leatherman & Niemeyer, 2005). However, several studies (e.g. Li & Cheung, 2021) have focused on pre-service teachers’ attitudes and self-efficacy in the implementation of inclusive education in Hong Kong. Few studies have explored in-service teachers’ attitudes and self-efficacy in including students with disabilities in the classroom in an Asian context. Besides, the implementation of inclusive practices can vary between Western and Eastern countries. Western educational systems may prioritise student-centred learning and catering students with diverse learning needs, whereas Eastern educational systems may prioritise academic achievement and discipline. Hong Kong has a history as a former British colony. Its special position in Asia may have influenced its approach to inclusive education. Thus, this study particularly focuses on primary and secondary in-service teachers in Hong Kong and investigates teachers’ attitudes and teacher self-efficacy. Moreover, this study examines whether teacher self-efficacy is related to their attitudes towards implementing inclusive practices. The predictors of teachers’ attitudes and teacher self-efficacy in an Asian country were of special interest because most of the related research has been conducted in Western countries. This research adds to the current body of literature and offers a comprehensive picture of the perception of in-service teachers in inclusive education in Hong Kong. The current study aims to address the following questions:

1. What are teachers’ attitudes towards inclusive education and self-efficacy for inclusive practices in Hong Kong?

2. What factors predict teachers’ attitudes towards inclusive education and self-efficacy for inclusive practices?

3. Is there a correlation between teachers’ self-efficacy for inclusive practices and their attitudes towards inclusive education?

Methods

Survey research was used to collect data from a large group of in-service teachers. Convenience and snowball sampling methods were employed to acquire the data. First, the in-service teachers attending professional learning programmes about catering for students with disabilities were recruited from a university in Hong Kong through a convenience sampling method. The Chinese version questionnaires were administered to the teachers on the first day of the professional learning programmes to minimise the impact of the course content on the teachers. Then, the participating teachers were encouraged to invite their colleagues to participate in the study through the snowball sampling method.

Participants

The data were obtained from 1110 primary and secondary in-service teachers in Hong Kong. Approximately, half of the participants were primary school teachers (n = 529, 47.7%) and half were secondary school teachers (n = 581, 52.3%). One-third of the participants were male (n = 351, 31.6%) and two-thirds were female (n = 759, 68.4%). Amongst all the teachers, 28 teachers (2.5%) were under 25 years old, 169 teachers (15.3%) were between 25 and 30 years old, 200 teachers (18%) were between 31 and 35 years old, 167 teachers (17.7%) were between 36 and 40 years old and 516 teachers (46.5%) were older than 40 years old. Detailed demographic information is shown in Table 1.

Table 1. Demographic information of in-service teachers.

Variable		Frequency	Percentage
Qualification	Less than bachelor’s degree	9	0.8
	Bachelor’s degree	537	48.4
	Master’s degree	560	50.5
	Doctor of Education	1	0.1
	Doctor of Philosophy	3	0.3
Years of teaching experience	Less than 5 years	178	16.0
	6–10 years	203	18.3
	11–20 years	393	35.4
	More than 21 years	336	30.3
Teaching experience with students with disabilities	No opportunity to teach these students yet	10	0.9
	Low	166	15.0
	Average	520	46.8
	High	414	37.3
Level of confidence in teaching students with disabilities	No opportunity to teach these students yet	8	0.7
	Low	140	12.6
	Average	773	69.6
	High	189	17.0
Level of training in teaching students with disabilities	None	194	17.5
	Low (less than 30 hours)	342	30.8
	Moderate (30–90 hours)	246	22.2
	High (more than 90 hours)	328	29.5
Level of knowledge of the local legislation/policy	None	161	14.5
	Poor	436	39.3
	Average	448	40.4
	Good	51	4.6
	Very good	14	1.3

For teaching experience and level of confidence in teaching students with disabilities, low referred to teachers who perceived they had limited experience or confidence, while high referred to teachers who perceived they had extensive experience or confidence. The average category referred to teachers who perceived they had some but not yet extensive experience or confidence in teaching students with disabilities. For knowledge of local legislation and policy, the four categories were used to reflect teachers’ perceived level of knowledge about inclusive legislation and policy. Poor indicates limited understanding, average indicates basic understanding, good indicates solid understanding, and very good indicates extensive understanding. The data showed that more than half of the participants perceived that they had some experience and confidence in teaching students with disabilities. Most participants perceived that they had limited or basic knowledge about legislation and policy related to inclusive education.

Data collection instruments

In this study, two survey instruments and a section with demographic information were utilised to collect data. Attitudes towards Inclusion Scale (AIS: Sharma & Jacobs, 2016) was employed to measure in-service teachers’ attitudes towards inclusive education. The scale initially consisted of 10 items, but two items were removed at the suggestion of the authors (Sharma & Jacobs, 2016) because they did not fit well with the construct of the scale. This instrument had two factors – attitudes in beliefs and attitudes in feelings – and comprises eight positively worded items using a seven-point Likert scale ranging from strongly disagree to strongly agree, with an undecided option in the middle of the response choices. Participants may interpret the midpoint of a Likert scale differently (Nadler et al., 2015). An undecided option allows respondents to choose a response that indicates they are unsure or have no strong opinion, which can help distinguish it from the midpoint on the scale (Raaijmakers et al., 2000). The higher scale scores represent the more positive attitudes of teachers towards inclusive education. Exploratory factor analysis was conducted with the current sample to identify whether the items on the AIS scale were loaded on the expected factors. The analysis showed that eight items of AIS in this sample loaded onto two expected factors with eigenvalues greater than one. The Kaiser–Meyer–Olkin measure was 0.86, and Bartlett’s test of sphericity was significant (p < .001). Cronbach’s alphas of the overall attitudes scale and the two factors for the current sample were 0.87, 0.86, and 0.87, respectively, proving that the content is adequately measured and validated. An example item is ‘I am happy to have students who need assistance with their daily activities included in my classrooms’.

Teacher Efficacy for Inclusive Practices (TEIP: Sharma et al., 2012) scale was used to determine teacher self-efficacy for inclusive practices in Hong Kong. The scale comprised three subscales: efficacy in inclusive instructions, efficacy in managing behaviour, and efficacy in collaboration. The scale involved 18 items using a six-point Likert scale ranging from strongly disagree to strongly agree. The higher scale scores imply teachers’ higher sense of self-efficacy to teach in inclusive classrooms. Exploratory factor analysis was employed to confirm the underlying construct of the scale in the current sample. Eighteen items of TEIP loaded onto three factors with eigenvalues greater than one. The Kaiser–Meyer–Olkin measure was 0.93, and Bartlett’s test of sphericity was significant (p < .001). The alpha coefficient for the overall TEIP was 0.92 and for the three subscales ranged from 0.81 to 0.91. The TEIP in this sample demonstrated good content validity and reliability. An example item is ‘I am confident in informing others who know little about laws and policies relating to the inclusion of students with disabilities’.

The last part of the questionnaire collected participants’ demographic information, including gender, age, highest educational qualification, years of teaching experience, type of school of teaching, experience of interacting with people with disabilities, experience in teaching students with disabilities, level of confidence in teaching students with disabilities, training in teaching students with disabilities and level of knowledge of the local legislation or policies concerning inclusive education. Each question provided teachers with two to six options. Teachers selected the option that best reflected their response to the questions. For instance, when asked about their gender, teachers had two options: male or female. Regarding their confidence in teaching students with disabilities, teachers were provided with four options: ‘no opportunity to teach these students yet’, ‘low’, ‘average’, and ‘high’. Similarly, when asked about their knowledge of the local legislation, the question offered five options: ‘none’, ‘poor’, ‘average’, ‘good’, and ‘very good’.

Pilot testing

The AIS, TEIP, and demographic information sections were initially written in English. They were translated into traditional Chinese and then back translated into English to ensure the validity of the translation. A pilot sample similar to the population was selected to test the questionnaire (Mertens, 2015). Thirty teachers were invited to pilot the questionnaire. They were asked to comment on the terms or words used in the Chinese version of the questionnaire, the clarity of the questions and the format, layout and style of the questionnaire. Several items were modified in response to their suggestions.

Ethical approval

The study was approved by the Ethics Committee of both the University in Australia, where the author is affiliated, and the University in Hong Kong, where the professional learning programme was conducted. Participants were informed about the purpose of the study, and they were given an explanation sheet on the research aims. The voluntary nature of the participation and the anonymity in the study was emphasised to the teachers. Participants were informed that they could not withdraw their participation once they submitted the anonymous questionnaire. The consent was implied when the completed questionnaires were returned.

Data analysis

Data analysis in this study was conducted using the Statistical Package for Social Sciences version 26 (SPSS Inc., Chicago, IL). First, Cronbach’s alpha was used to assess the overall scales and subscales’ reliability. Then, the mean scores of all scales and subscales were calculated to determine the level of teachers’ attitudes towards and teacher self-efficacy in inclusive education. Third, correlations analysis was conducted to understand the relationship between teachers’ attitudes and teacher self-efficacy. Lastly, regression analyses were used to identify the predictors of teachers’ attitudes and teacher self-efficacy for inclusive practices.

Results

Teachers’ attitudes towards inclusive education

An examination of the AIS revealed that the overall mean of teachers’ attitudes towards inclusive education was higher than the neutral range of the scale (i.e. 3.58–4.42) (see Table 2). The result indicates that teachers have slightly positive attitudes towards inclusive education. Of the survey respondents, 37.7% of participants had a positive attitude towards inclusive education, while 34.9% were undecided and 27.4% had a negative attitude towards it. Regarding the subscales, the mean score of teachers’ attitudes about their beliefs in the benefit of implementing inclusive education fell just within the neutral range of the scale (i.e. 3.58–4.42). Their attitudes about the feelings in teaching in an inclusive classroom scored the highest between the two subscales.

Table 2. Overall scale and subscales of teachers’ attitudes scores.

		95% confidence interval
Variable	Mean	Lower CI	Upper CI	Standard deviation
AIS full	4.54	4.48	4.60	1.04
AIS (beliefs)	4.04	3.96	4.11	1.32
AIS (feelings)	5.05	4.98	5.11	1.08

Attitude mean responses range from 1 (strongly disagree) to 7 (strongly agree).

Teacher self-efficacy for inclusive practices

The overall mean score of teacher self-efficacy was moderate (Table 3). Most teachers (71.8%) perceived that they were capable to teach students with disabilities, and more than one-quarter of teachers (28.8%) perceived that they were unable to teach students with disabilities in the classroom. Regarding the subscales, Hong Kong teachers had the highest self-efficacy in managing students’ behaviour amongst the three subscales, and were the least confident in collaboration.

Table 3. Overall scale and subscales of teachers’ self-efficacy scores.

		95% confidence interval
Variable	Mean	Lower CI	Upper CI	Standard deviation
TEIP full	4.27	4.30	4.33	0.63
TEIP (inclusive instruction)	4.33	4.30	4.37	0.64
TEIP (managing behaviour)	4.36	4.32	4.41	0.80
TEIP (collaboration)	4.11	4.07	4.15	0.73

Self-efficacy mean responses range from 1 (strongly disagree) to 6 (strongly agree).

Correlation between teachers’ attitudes towards inclusive education and their self-efficacy for inclusive practices

As shown in Table 4, teachers’ overall attitudes towards inclusive education correlated significantly with their overall self-efficacy for inclusive practices. All the subscales of teachers’ attitudes correlated significantly with subscales of teacher self-efficacy. Amongst the three subscales of teacher self-efficacy, teacher self-efficacy in collaboration correlated most with their overall attitudes and teachers’ feelings, which correlated most strongly with teachers’ overall self-efficacy.

Table 4. Pearson’s correlations amongst the AIS and TEIPs’ overall scores and sub-scale scores.

		1	2	3	4	5	6
1	AIS full	–
2	AIS (beliefs)	0.89**	–
3	AIS (feelings)	0.83**	0.48**	–
4	TEIP full	0.46**	0.34**	0.48**	–
5	TEIP (inclusive instruction)	0.39**	0.27**	0.41**	0.85**	–
6	TEIP (managing behaviour)	0.36**	0.24**	0.40**	0.88**	0.61**	–
7	TEIP (collaboration)	0.46**	0.36**	0.44**	0.87**	0.63**	0.63**

** p < .01.

Predictors of teachers’ attitudes towards inclusive education

A stepwise multiple regression analysis was used to predict the factors of teachers’ attitudes towards inclusive education. Demographic characteristics (i.e. prior experience in teaching students with disabilities, knowledge of legislation and policies, confidence in teaching students with disabilities, level of training in teaching students with disabilities, years of teaching experience, gender and school type) and the subscales of teacher self-efficacy (i.e. inclusive instruction, managing behaviour, and collaboration) were tested as predictors of teachers’ attitudes. The results (Table 5) indicated that the stepwise regression was significant. Prior experience in teaching students with disabilities, confidence in teaching students with disabilities, level of training in teaching students with disabilities, years of teaching experience, efficacy in inclusive instruction, and efficacy in collaboration were determined to significantly predict teachers’ attitudes towards inclusive education.

Table 5. Stepwise regression analysis for AIS predicting teachers’ attitudes towards inclusive education.

Predictor variables	B	SE B	β
Prior experience in teaching students with disabilities	–0.126	0.042	–0.088**
Confidence in teaching students with disabilities	0.254	0.058	0.139***
Level of training in teaching students with disabilities	0.114	0.027	0.119***
Years of teaching experience	–0.068	0.026	–0.068**
Efficacy in inclusive instruction	0.226	0.055	0.139***
Efficacy in collaboration	0.426	0.049	0.300***
R^2	0.260
Model statistics	F(6, 1103) = 64.426

** p < .01.

*** p < .001.

Predictors of teacher self-efficacy for inclusive practices

Similarly, a stepwise regression analysis was employed to identify the predictors of teacher self-efficacy. The demographic characteristics and the subscales of attitudes (i.e. beliefs and feelings) were tested as predictors of teacher self-efficacy. The results (Table 6) showed that the stepwise regression was significant. Knowledge of legislation and policies, confidence in teaching students with disabilities, teachers’ beliefs and their feelings significantly predicted teacher self-efficacy for inclusive practices.

Table 6. Stepwise regression analysis for TEIP predicting teachers’ self-efficacy for inclusive practices.

Predictor variables	B	SE B	β
Knowledge of legislation and policies	0.084	0.019	0.112***
Confidence in teaching students with disabilities	0.316	0.029	0.287***
Attitude (beliefs)	0.042	0.013	0.089**
Attitude (feelings)	0.192	0.017	0.331***
R^2	0.348
Model statistics	F(4, 1105) = 145.984***

** p < .01.

*** p < .001.

Discussion

Implementing inclusive education has been a long journey in the Hong Kong education system. The influence of teachers’ attitudes and teacher self-efficacy on implementing inclusion has drawn significant attention in previous research (De Boer et al., 2011). This study examined teachers’ attitudes towards inclusive education and teacher self-efficacy for inclusive practices in the classroom, the relationship between them and the factors to predict their attitudes and self-efficacy in inclusive education.

The results show that, generally, teachers held slightly positive attitudes towards inclusive education in Hong Kong, suggesting that they supported inclusive education. In comparison, studies conducted in other Asian countries – for example, Japan (Yada & Savolainen (2017) and Chile (San Martin et al., 2021) – revealed that teachers neither supported nor opposed the implementation of inclusive education. In addition, teachers’ attitudes in this study were positive in terms of their feelings about teaching students with disabilities. However, when asked about their beliefs in inclusive education, the result indicated that teachers were undecided; thus, there was a gap between teachers’ feelings and beliefs regarding inclusive education. While teachers have favourable feelings in teaching students with disabilities in the classroom, they are less supportive of the philosophy of inclusive education.

The overall teacher self-efficacy for inclusive practices was moderate, suggesting that teachers believed they were confident in using inclusive teaching strategies to some degree. Although teachers have limited choice and control in inclusive education, as it is a human right for students with disabilities, this study’s results imply that teachers felt self-efficacious about implementing inclusive practices for these students. The level of teacher self-efficacy was highest in managing students’ behaviour in inclusive classrooms. One potential explanation for this finding could be that the Hong Kong education system places a strong emphasis on academic achievement (Forlin, 2010), which may require teachers to maintain a high level of student discipline to foster a successful learning environment for academic purposes. This emphasis on discipline may contribute to teachers’ perception in managing students’ behaviour. On the other hand, teachers had the least self-efficacy in collaboration. One possible explanation for this finding might be the lack of time and training in collaboration for inclusive practices. In Hong Kong, most in-service teachers spend a tremendous amount of time in extracurricular activities and administrative work, leading to less time collaborating with colleagues, professionals, and parents. The findings imply that teachers were not confident in working with others, which might be related to a lack of collaboration skills. As is more often the case, since most teachers have a heavy workload in Hong Kong, they lack sufficient time to discuss or work with parents, colleagues, and professionals about the learning programmes of students with disabilities. Policymakers should provide support and allocate more resources to teachers, thereby lessening their workload, to improve this situation.

The predictors of teachers’ attitudes and self-efficacy were analysed. Several predictors were identified in the current study. Regarding teachers’ attitudes, teachers’ prior experience in teaching students with disabilities had a small but significant negative effect on teachers’ attitudes. Teachers with more teaching experience with students with disabilities tended to hold negative attitudes towards inclusive education. This result contradicted with previous findings (Avramidis & Kalyva, 2007) that showed a positive change in teachers’ attitudes when their previous experience in teaching students with disabilities increased. In the current sample, teachers’ attitudes shifted negatively over time may be because they struggled to include students with disabilities in their classrooms without adequate resources and support in implementing inclusive education. The results of the current study also indicated that teachers with more years of teaching experience in general had a small and negative effect on their attitudes towards inclusive education. This result is consistent with previous studies (e.g. Savolainen et al., 2012; Yada & Savolainen, 2017). Teachers with more years of teaching experience could be those who finished their teacher education before inclusion became a usual practice in education (MacFarlane & Woolfson, 2013). Inclusive education only recently became a part of teacher education programmes, teachers with less experience may possess a greater level of familiarity with and acceptance of inclusion, and therefore they may be more likely to adopt inclusive practices (Yada et al., 2018). In Hong Kong, inclusive education reforms became mandatory in 1997, and teacher training on inclusive education was launched in 2007 to prepare teachers for teaching students with disabilities. As a result, teachers with more teaching experience may not be as prepared as those with less experience. Another important finding was that teacher self-efficacy in inclusive instruction and in collaboration were the predictors of teachers’ attitudes. Mainly, teacher self-efficacy in collaboration was the most potent predictor of teachers’ attitudes towards inclusive education. This result supports the findings of Yada and Savolainen (2017), which indicated that teachers with higher self-efficacy in collaboration had more positive attitudes towards inclusive education. Previous research (e.g. Savolainen et al., 2012) has suggested that the collaboration skills of teachers must be improved and must be the focus of teacher training programmes. This can explain the importance of training to prepare teachers to implement inclusive education. The level of training was another significant predictor of teachers’ attitudes in this study. Teachers who received more training in catering for students with disabilities expressed more favourable attitudes towards inclusion. This finding is consistent with those of previous studies (e.g. Avramidis & Kalyva, 2007; MacFarlane & Woolfson, 2013), which concluded that teachers with more training were more accepting of students with disabilities than those who had little or no such training. The knowledge about teaching students with disabilities through formal training was considered an essential factor to improving teachers’ attitudes towards inclusive education.

In terms of teachers’ confidence level, the confidence in teaching students with disabilities was a significant predictor of teachers’ attitudes and teacher self-efficacy. Confidence differs from self-efficacy: ‘Confidence is a nondescript term that refers to strength of belief but does not necessarily specify what the certainty is about … Perceived self-efficacy refers to belief in one’s agentive capabilities, that one can produce given levels of attainment’ (Bandura, 1997, p. 382). In this study, confidence in teaching refers to teachers’ beliefs in their competence to teach and support students with disabilities. Perceived teacher self-efficacy refers to teachers’ beliefs in their capacity to provide meaningful education and progress for students with disabilities. The findings of this study indicate that teachers who perceived more confidence in their ability to teach students with disabilities tend to have more positive attitudes and a higher sense of teacher self-efficacy for inclusive practices. The findings of this study are consistent with those of an earlier study conducted in Hong Kong, in which Chao et al. (2016) reported that confidence in teaching students with disabilities was the predictors of teacher self-efficacy with respect to managing behaviour, collaboration, and inclusive instruction.

In identifying the predictors of teacher self-efficacy, teachers’ attitudes in feelings and beliefs were the significant predictors of self-efficacy for inclusive practices. Teachers’ feelings were found to be the strongest predictor of teachers’ overall self-efficacy. Teachers who hold more positive feelings towards inclusive education are likely to have a higher sense of teacher self-efficacy for inclusive practices. Meanwhile, teachers’ beliefs in inclusive education had a small but significant effect on teachers’ overall self-efficacy for inclusive practices. Another important predictor of teacher self-efficacy for inclusive practice is the knowledge of legislation and policies, which is in line with previous research conducted in Hong Kong (e.g. Chao et al., 2016, 2017; Forlin et al., 2014). Teachers with more knowledge of legislation and policies tend to have higher teacher self-efficacy, demonstrating the need to include the learning content of local laws and policies within the Hong Kong context in the training course for teachers in order to understand inclusive practices, which is aligned and consistent with the legislation.

To determine the relationship between teachers’ attitudes towards inclusive education and teacher self-efficacy for inclusive practices, the result reveals a statistically significant, positive relationship between them. The finding of this study with the Hong Kong sample is consistent with that of previous research conducted in various nations (e.g. Yada & Savolainen, 2017). The result implies that the relationship between teachers’ attitudes and teacher self-efficacy is universal (Yada et al., 2022), even though Hong Kong has its unique position in Asia. In this study, teachers’ overall attitudes were particularly correlated with self-efficacy in collaboration. This finding suggests that teachers who perceived they were more capable in collaboration had more positive attitudes towards inclusive education. In addition, teachers’ overall self-efficacy was especially correlated with teachers’ feelings towards inclusion. This result indicates that teachers who felt more positive in an inclusive classroom had a higher level of teacher self-efficacy for inclusive practices.

The findings of this study show that there is a positive correlation between teachers’ attitudes and teacher self-efficacy, indicating that these factors are closely related. In the TPB, attitudes and perceived behaviour control, which is compatible with teacher self-efficacy in this study, influence intentions and the subsequent actual behaviour. The positive relationship between teachers’ attitudes and teacher self-efficacy in inclusive education can contribute to a positive impact on teachers’ intentions, which, in turn, increases the likelihood of them implementing inclusive practices in the classrooms. Moreover, the results indicate that teachers’ attitudes and teacher self-efficacy are mutually predictive, suggesting that teachers’ attitudes can influence teacher self-efficacy, as well as the reverse, indicating a bidirectional relationship between these factors. In other words, positive attitudes can enhance teachers’ belief in their own abilities, while increased teacher self-efficacy can lead to more positive attitudes towards inclusive education.

Implications

For inclusion to succeed, teachers must be aware of, and motivated with regard to, the philosophy of inclusive education and acquire the necessary knowledge and skills to meet the diverse learning needs of all students. The need for teacher training on catering for students with disabilities and diverse learning needs is stressed in the literature (Chao et al., 2017). However, not all teachers can receive special education training in Hong Kong (Forlin, 2010). Offering teachers from various programmes the opportunity to gain knowledge of legislation and policies, collaboration experience and knowledge about hands-on, concrete, and practical training in applying appropriate teaching strategies and inclusive instructions can be one way to improve Hong Kong teacher self-efficacy and attitudes towards inclusive education. When attempting to positively change teachers’ attitudes towards inclusive education through teacher training, policymakers and teacher education providers should place greater emphasis on course content that focuses on legislation and policies. This can enhance teacher self-efficacy and contribute to the successful implementation of inclusive education.

Limitations

This study holds certain limitations. It was conducted using self-administrated questionnaires, which may lead to participants’ dishonest answers due to social bias in the community. Although anonymous questionnaires were used, social desirability bias amongst teachers may still exist in Hong Kong, where the government places a strong emphasis on its inclusive education policy. Thus, teachers may feel pressure to support inclusive education even if they have concerns about it. Furthermore, participants’ differences in interpreting and understanding the questions influence the results of the study. The dataset of this study does not represent the entire population of in-service teachers in Hong Kong, although the participants were from various districts across the Hong Kong region. Generalisation of the results could be challenging due to the convenience and snowball sampling methods were used to invite participants and their teacher colleagues to participate in this study. Consequently, a random sampling method may be applied to collect data in future research. Furthermore, even though the attitude and self-efficacy scales provide good reliability, the translated Chinese version may not relay the same meaning as the original English version, especially due to the cultural differences between the two countries and backgrounds. Moreover, this quantitative study cannot provide a whole picture of teacher self-efficacy and attitudes towards inclusive education. The instruments were well written and perform well in terms of psychometrics, but they do not give explanations as to why teachers have such attitudes and levels of self-efficacy about inclusive education. Mixed method or qualitative research can be used in future investigation to further understand teachers’ perception of the implementation of inclusive education.

Conclusions

In Hong Kong, inclusive education has been implemented for nearly three decades. The successful implementation of inclusive education continues to pose challenges. The findings of this study help to understand teachers’ perspectives on inclusive education and provide several insights and implications for policymakers, teacher education providers, and administrators on how to improve the implementation of inclusive education. The Hong Kong government must focus more on providing training opportunities to enhance teachers’ collaborative skills and their knowledge of legislation and policies of inclusive education to bridge the gap between inclusive policy and the implementation of inclusive education.

Acknowledgements

The author would like to thank all the teachers who participated in the study.

Disclosure statement

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