Teaching and Teacher Education 59 (2016) 338e346

Contents lists available at ScienceDirect

Teaching and Teacher Education

journal homepage: www.elsevier.com/locate/tate

What are the inclusive teaching tasks that require the highest
self-efficacy?
Francisco T.T. Lai a, b, Eria P.Y. Li c, Mingxia Ji c, Wikki W.K. Wong c, Sing Kai Lo b, c, d, *
a
Department of Psychological Studies, The Education University of Hong Kong, Hong Kong
b
Assessment Research Centre, The Education University of Hong Kong, Hong Kong
c
Graduate School, The Education University of Hong Kong, Hong Kong
d
Centre for Special Educational Needs and Inclusive Education, The Education University of Hong Kong, Hong Kong

h i g h l i g h t s

There is insufficient information on the most urgent needs for training in inclusive education.

A survey on primary school teachers was conducted to examine their efficacy for inclusive tasks.

The tasks were ranked by the extent of self-efficacy using a Rasch rating scale model.

Tasks requiring notably higher efficacy could potentially imply the need for additional training.

Findings can also be taken into considerations when designing curriculum for teacher training.

a r t i c l e i n f o a b s t r a c t

Article history: Inclusive teaching tasks have consistently been found challenging for teachers, but it is unclear how they
Received 17 November 2015 are ranked in terms of the extent of self-efficacy required. This study aimed at deriving such a hierarchy.
Received in revised form A survey was conducted on 107 primary school teachers in Hong Kong using the Teacher Efficacy for
27 June 2016
Inclusive Practices scale. A Rasch rating scale model was applied to empirically examine the hierarchical
Accepted 6 July 2016
structure. Good person reliability (0.89) and model fit (MNSQ 0.6e1.4) were achieved. Managing physical
aggression was found at the top of the hierarchy; this and other results could facilitate the identification
of training needs.
Keywords:
Inclusive education
© 2016 Elsevier Ltd. All rights reserved.
Teacher efficacy
Rasch rating scale model
Special educational needs

1. Introduction with developmental difficulties (Counts, Nigg, Stawicki, Rappley, &

Implementation of inclusive practices in mainstream class-
rooms has always been full of challenges and obstacles for teachers
(Lacey & Scull, 2015; Pivik, McComas, & Laflamme, 2002; Shah, Das,
Desai, & Tiwari, 2016). Teachers of students with special educa-
tional needs (SEN) in inclusive settings are required to juggle
numerous difficult tasks in their daily practice (Foote & Collins,
2011; Westwood, 2013). For example, one of the most frustrating
missions is to manage the occasional aberrant behavior of students
Von Eye, 2005; Winstanley, Eagle, & Robbins, 2006). Frequent
follow-up consultations with professionals and regular meetings
with parents are also required to tackle the problems. All these
jointly imply an urgent need for additional knowledge, experience
and skills. Therefore, in recent years, an emphasis on inclusive
education training for teachers in the hope of enhancing their
capability to face the challenges of inclusive education has been
strengthened (Florian & Rouse, 2009; Lancaster & Bain, 2007;
Sharma, Forlin, & Loreman, 2008).

1.1. Three common aspects of inclusive teaching

* Corresponding author. The Education University of Hong Kong, 10 Lo Ping Road,

Tai Po, New Territories, Hong Kong.
Regarding the design of curricula for this kind of training, it is
E-mail addresses: francisco@s.eduhk.hk (F.T.T. Lai), eriali@eduhk.hk (E.P.Y. Li), commonly believed that teachers would require skills in at least
mxji@eduhk.hk (M. Ji), wikki@s.eduhk.hk (W.W.K. Wong), skl@eduhk.hk (S.K. Lo). three aspects as proposed and discussed by Sharma, Loreman, and

http://dx.doi.org/10.1016/j.tate.2016.07.006
0742-051X/© 2016 Elsevier Ltd. All rights reserved.
 F.T.T. Lai et al. / Teaching and Teacher Education 59 (2016) 338e346
Forlin (2012). Firstly, appropriate and clear instructions to be
communicated to students are important. For instance, educational
goals for students with SEN are in most cases different from those
of mainstream students. It is demanding to assign learning tasks
and assessments tailor-made for them while simultaneously
teaching all students as a single class. Although guidelines are
available (Broderick, Mehta-Parekh, & Reid, 2005), many teachers
need additional training and years of experience to be self-
efficacious and fully capable of performing this task.
Secondly, inclusive education always requires collaborative
effort to succeed (Ainscow & Sandill, 2010; Engelbrecht, Oswald, &
Forlin, 2006). Trust and support from parents, information and
administrative assistance from seniors in school, and professional
advice from experts need to be solicited and gathered for the best
results. This has been proven to be difficult given the heavy
workload of teachers. For example, in the Hong Kong aided school
systems, educational psychologists, who typically provide consul-
tation and service to several schools simultaneously, only visit
schools twice a month, sometimes even less. This imposes chal-
lenges for teachers to initiate collaboration and seek their profes-
sional advice.
Thirdly, as illustrated by the aforementioned example, man-
aging the behavior of students is of utmost importance. Disruptive
behavior is common among students with developmental diffi-
culties (Counts et al., 2005; Winstanley et al., 2006). Intervention
strategies have been introduced to improve their behavioral issues,
such as Social Story ™ and Treatment and Education of Autistic and
Related Communication Handicapped Children (TEACCH) (Crozier
& Tincani, 2005; Virues-Ortega, Julio, & Pastor-Barriuso, 2013).
However, given the limited time and resources as well as the tight
teaching schedule and syllabus, teachers seldom have opportu-
nities to properly implement these strategies, and have to leave the
job to counselors and other professionals. This, in turn, leads to
even fewer opportunities for them to master the skills of handling
the students.
Although these task domains have been well categorized and
tested with good validity, and have facilitated well-organized
design of inclusive education training curricula (Forlin, Sharma, &
Loreman, 2013; Park, Dimitrov, Das, & Gichuru, 2014; Sharma
et al., 2012), inclusive education training should be more focused
and specialized on tasks that represent areas where teachers feel
less efficacious. This is because, if teachers are found to be highly
efficacious in certain tasks, they might have already obtained a
certain level of relevant experience and skills. In this case, resources
invested in the provision of such training to teachers would better
be allocated to other areas where they feel less efficacious.
Therefore, the main objective of this study was to identify a
sequence of inclusive teaching tasks along the continuum of
teachers’ efficacy; this will enable us to recommend those requiring
the highest level of efficacy, on which more focus and training
emphasis should be placed. Furthermore, it would be insightful to
investigate how the tasks of each domain require different levels of
efficacy, and to analyze how the teacher’s background might in-
fluence his/her particular efficacy for individual tasks, given a
similar overall efficacy level.
1.2. Inclusive teaching efficacy
To guide our reasoning on inclusive teaching efficacy, Bandura’s
conceptual framework on self-efficacy has been adopted (Malinen
et al., 2013). Inclusive teaching efficacy is defined as the judge-
ment of one’s capability to implement the required teaching
practices in an inclusive education setting (Bandura, 2006; Gibbs
et al., 2007). Efficacy is theorized as being constructed from
mastery experiences, vicarious experiences, social persuasion, and
339
somatic and emotional states (Bandura, 1977). When comparing
the required teacher efficacy levels between tasks and across socio-
demographic factors, adoption of this conceptualization of efficacy
will facilitate the interpretation of the results.
Research on teaching efficacy specific to the context of inclusive
education is relatively scarce in spite of the emerging trend of in-
clusive education. A significant proportion of this research in-
vestigates the effects of different specific training programs on
inclusive teaching efficacy (Forlin, Loreman, & Sharma, 2014;
Sharma & Sokal, 2015; Sokal, Woloshyn, & Funk-Unrau, 2013). For
example, Sokal et al. (2013) suggested that student teachers going
through both practicum and coursework achieved higher efficacy
scores than those who did only coursework, in support of the
conceptualization of mastery experiences as a crucial source of self-
efficacy (Malinen et al., 2013). Forlin et al. (2014) evaluated im-
provements in efficacy on using inclusive instructions after a
system-wide professional training program in Hong Kong. There
were other studies examining socio-demographic determinants of
efficacy. Gender (Shaukat, Sharma, & Furlonger, 2013), experience
with students with SEN (Malinen et al., 2013), previous special
education training (Levi, Einav, Raskind, Ziv, & Margalit, 2013),
school climate (O’Toole & Burke, 2013), cultural background
(Savolainen, Engelbrecht, Nel, & Malinen, 2012) and democratic
beliefs (Almog & Shechtman, 2007) were identified as having sig-
nificant influence on inclusive teaching efficacy. Despite the
importance of the insights generated from these studies, little
research has investigated the associations with efficacy by item
level (i.e., specific tasks), or comparing each inclusive teaching task
in terms of their required self-efficacy. The current study intends to
fill this gap.
1.3. Research hypothesis
Based on an overview of the literature on inclusive education
practices, it is hypothesized that teaching tasks regarding physical
aggression and disruptive behavior of students with SEN will be
seen as most significant in terms of a hierarchy of required self-
efficacy (i.e., requiring the most efficacy to carry out), followed by
tasks that involve collaborations with parents in the classroom.
1.3.1. Physical aggression and disruptive behavior
It is common that students with certain developmental diffi-
culties show occasional aberrant behavior or physical aggression
(Bearss, Johnson, Handen, Smith, & Scahill, 2013; Eikeseth,
Klintwall, Jahr, & Karlsson, 2012; Kaat, Lecavalier, & Aman, 2014;
Logan et al., 2015). In primary school settings, although such
behavior probably causes no physical harm to teachers, it inevitably
affects the emotional stability of the teacher, and would lead to less
satisfying teaching quality and thus worse learning outcomes
(Klusmann, Kunter, Trautwein, Lüdtke, & Baumert, 2008). Teaching
and taking care of students with behavioral issues is difficult, as
reflected by the abundance of behavioral intervention strategies
teachers need to master and implement in order to properly
manage behavior. Also, student behavioral issues have been re-
ported to have caused serious stress for teachers in Hong Kong
(Pang, 2011).
Facing the prevalent behavioral issues of students with SEN,
behavioral interventions, such as Social Story ™ (Karkhaneh et al.,
2010) and TEACCH (Virues-Ortega et al., 2013) have been intro-
duced. However, in spite of the evidence of their effectiveness, it
takes time for noticeable improvements to be observed (Brophy,
2003; Crozier & Sileo, 2005; Karkhaneh et al., 2010; Scattone,
Wilczynski, Edwards, & Rabian, 2002; Virues-Ortega et al., 2013).
Also, these strategies might have to be used in after-class hours or
when the student is taken out of class. During class time, given the
340 F.T.T. Lai et al. / Teaching and Teacher Education 59 (2016) 338e346

tight schedule and the presence of other students, some other
strategies have been recommended, for example, whole-classroom
interventions (Oliver, Wehby, & Reschly, 2010), and antecedent
strategies (Kern & Clemens, 2007). In most cases, teachers need to
implement both individualized programs and whole class inter-
vention strategies to achieve an ideal result. This implies that once
behavioral issues arise, teaching practice would require more tasks
to be accomplished and more skills to be mastered to implement
them. In other words, tasks related to managing disruptive
behavior consist of many more sub-tasks.
Heavy stress due to challenging behavior has been observed
among teachers in Hong Kong (Pang, 2011). This comparative study
found that Hong Kong teachers were adversely affected by stu-
dents’ challenging behavior, and reported more negative beliefs in
working with these students than teachers in Italy, the USA, the
Netherlands, Surinam, South Africa and Russia. It was speculated
that a lack of relevant knowledge and skills to manage the behavior
might explain the stress in the face of such challenging behavior.
This suggests the plausibility of the reasoning discussed above.
Therefore, it is hypothesized that tasks related to disruptive or
aggressive behavior in students with SEN would require higher
self-efficacy than other tasks.
1.3.2. Collaboration with parents
It is also hypothesized that collaboration with parents in
teaching students with SEN ranks second to handling physical
aggression and disruptive behavior in terms of the required self-
efficacy. This is because parents of students with SEN are usually
highly protective of their children, and some even feel the need to
act as experts and advocate for their children (Falkmer, Anderson,
Joosten, & Falkmer, 2015; Wong, 2002). Moreover, they generally
do not have sufficient confidence in caretakers other than them-
selves. Thus, there are often complaints from parents about
teachers’ lack of knowledge and skills in handling SEN students, as
well as poor communication between teachers and parents and
among the teaching staff (Falkmer et al., 2015). In order to effec-
tively collaborate with parents, teachers need to gain their confi-
dence and trust by showing a considerable degree of expertise in
taking care of children with SEN.
1.4. Broader aim and significance of the study
By testing these research hypotheses and deriving a hierarchy of
inclusive teaching tasks in terms of required self-efficacy, it is our
broader aim that this study will facilitate future development and
refinement of inclusive education training. The identification of
teachers’ most difficult challenges could arouse attention, opinions
and debate which could, in turn, lead to substantial improvements
in the inclusive education environment for teachers, and eventually
students and other stakeholders as well.
the city’s situation. Second, stress experienced by teachers has been
largely discussed as a major issue of the Hong Kong educational
system (Chan, 2006; Jin, Yeung, Tang, & Low, 2008). It seems more
than legitimate to be exploring the situation of the population
whose needs are more urgent, since inclusive education has been
proven to be one source of serious stress for teachers (Engelbrecht,
Swart, & Eloff, 2006; Talmor, Reiter, & Feigin, 2005).
2.1.2. Sampling and data collection
Convenience sampling was adopted in this study. The target
participants were in-service teachers who were teaching, or would
be teaching, students with autism spectrum disorder (ASD) in an
inclusive classroom. Data collection was conducted at the begin-
ning of two workshops on teaching students with ASD in July 2013
and July 2014, respectively. In addition, a public lecture for school
teachers on pharmaceutical treatments for students with ASD and
Attention Deficit Hyperactivity Disorder (ADHD) was held in
February 2015. Self-administered questionnaires were distributed.
A total of 161 valid responses were received from the three sites of
data collection. After excluding duplicates (those who attended two
or more of the three events), 107 primary school teachers were
adopted as our final sample for analysis. Table 1 illustrates the
demographic characteristics of the sample. The study was approved
by the Human Research Ethics Committee of the authors’ home
institution.
2.2. Adopted instrument
To fulfill the objectives of the study, a scale covering a suffi-
ciently large variety of tasks in inclusive education was needed.
Among those available, the Teacher Efficacy for Inclusive Practices
(TEIP) scale was most suitable. Developed by Sharma et al. (2012),
the scale covers three domains of inclusive practices, namely effi-
cacy in using inclusive instruction, efficacy in collaboration, and
efficacy in dealing with disruptive behavior.
The TEIP scale consists of a total of 18 items, each of which
represents one specific task for teachers in an inclusive educational
environment. A six-point Likert rating scale was used: ‘strongly
disagree ¼ 1’, ‘disagree ¼ 2’, ‘somewhat disagree ¼ 3’, ‘somewhat
agree ¼ 4’, ‘agree ¼ 5’ and ‘strongly agree ¼ 6’. The scale had good
reliability, as the Cronbach’s alpha was found to be as high as 0.89
by the developers. In addition, factor analyses by previous studies
revealed a three-factor structure with each of the factors containing
six items across study populations (Park et al., 2014; Sharma et al.,
Table 1
Basic demographic characteristics of participants.
Male (n ¼ 20) Female (n ¼ 86) Total

n % n % n %
2. Method Age
20-29 5 25.0% 20 23.3% 25 23.4%
2.1. Participants 30-39 5 25.0% 34 39.5% 39 36.4%
40-49 8 40.0% 27 31.4% 35 32.7%
 50 2 10.0% 5 5.8% 7 6.5%
2.1.1. Adopting the Hong Kong population Qualification
With regard to the study population, we conducted the analyses Certificate/Diploma 2 9.5% 8 9.4% 10 9.3%
with a sample of Hong Kong primary school teachers based on two Bachelor’s degree 12 57.1% 51 60.0% 63 58.9%
major reasons. First, many relevant studies have been conducted in Master’s degree 7 33.3% 26 30.6% 33 30.8%
Job Position
the Hong Kong educational context, for example, Forlin et al. (2014)
SENCO 2 10.0% 9 10.5% 11 10.3%
examined teachers’ perceptions of inclusive education after pro- Teacher (10 years) 8 40.0% 42 48.8% 50 46.7%
fessional training, while Tait and Mundia (2013) compared teach- Teacher (2e9 years) 7 35.0% 28 32.6% 35 32.7%
er’s efficacy in inclusive practices in Hong Kong with that of Teacher (1 year) 3 15.0% 7 8.1% 10 9.3%
teachers in Brunei. Our results could be given more weight if we Note. SENCO ¼ special educational needs coordinator.
base our analysis on the already unearthed knowledge specific to
 F.T.T. Lai et al. / Teaching and Teacher Education 59 (2016) 338e346
2012). Most recently, it was nevertheless also suggested by Park
et al. (2014) with preliminary evidence that the scale is essen-
tially unidimensional, and that the existence of a single general
latent trait of teacher efficacy on inclusive practices is plausible.
Due to limitations such as adoption of an undergraduate student
(instead of in-service teacher) sample and a relatively small sample
size, further definitive evidence is warranted for confirmation of
this potential unidimensional factorial structure.
2.3. Data analysis
2.3.1. Adopting the Rasch rating scale model framework
Unlike previous studies which adopted the TEIP scale, the Rasch
rating scale model (Andrich, 1978) instead of the traditional clas-
sical test theory (CTT) framework was applied in this current study.
The Rasch model is well known for possessing several advantages
in deriving the item hierarchy of scales (Rasch, 1966).
First, the item measures (in logits) and hierarchy generated from
the Rasch model were sample-free, which implied the reliability of
such measures and thus hierarchical ordering across samples with
different levels of efficacy, or in other words, another group of
participating teachers with higher or lower efficacy levels would
give roughly the same estimates of item difficulty (in logits). Sec-
ond, the Rasch model converted responses into item and person
measures of an interval nature (in logits). This could avoid unre-
alistically assuming interval Likert-rating scores in item compari-
son by statistical testing, e.g. differential item functioning (DIF).
Third, the model enables us to observe the distribution of item
measures relative to the person measures because they were con-
verted into the same unit (logits). We could then imply whether the
scale was lacking items of relatively higher endorsibility or other-
wise, with regard to the specific sample. The Rasch analyses were
conducted using Winsteps 3.81.0 (Linacre, 2014).
2.3.2. Dimensionality of teacher efficacy
Further to the findings by Park et al. (2014) of an evident single
latent trait of self-efficacy of implementing inclusive teaching tasks,
a residual principal component analysis (RPCA) was conducted to
confirm the unidimensionality of the TEIP scale. We followed the
guidelines suggested by Linacre (2009), that for the existence of a
secondary dimension (besides the current Rasch model) to be
evident, the eigenvalue of the first contrast in the RPCA, proportion
of raw variance explained by the model as well as the quotient from
dividing the raw variance explained by the item measures by that
explained by the first contrast should be carefully examined. If any
of these three criteria were not met, evidence for secondary di-
mensions would be insufficient and unidimensionality could be
confirmed.
2.3.3. Inclusive teaching task hierarchy
With scale unidimensionality, deriving the item measures from
the Rasch analysis enabled us to locate each item (or teaching task)
of the TEIP scale along the measured latent trait, i.e. self-efficacy for
inclusive practice. Some items (i.e. the corresponding inclusive
tasks) were used to detect lower levels of self-efficacy, while some
were used to detect higher levels. Model fitness was also assessed
based on the infit and outfit mean square (MNSQ) statistics for each
item. An effect size of MNSQ ranging from 0.6 to 1.4 was considered
acceptable in terms of the fitness of the particular item (Bond & Fox,
2007). In addition, the Wright map was generated for a graphical
illustration of the distribution of item measures relative to the
participants’ levels of efficacy.
2.3.4. Differential item functioning (DIF) analyses
A DIF analysis was conducted on several demographic variables
341
to investigate by what patterns the item measures differed across
groups, given the same overall efficacy level. Thee stratification
variables that were examined included gender, age, education, and
job position. The Mantel-Haenszel chi-square statistic for polyto-
mous models was used to test the statistical significance of DIF for
each item.
3. Results
3.1. Sample characteristics
Table 1 describes the characteristics of the sample of primary
school teachers recruited to participate in the study. Notable
gender imbalance was observed, with only 20 (18.9%) male par-
ticipants, which is probably consistent with the overall demog-
raphy of the primary education industry. Age was distributed more
evenly with the most common range being 30e39 (36.4%) and the
least being  50 (6.5%). Nearly 90% of the participants had obtained
a bachelor’s degree or above. Almost half of the sample was
teachers with more than ten years of experience, with one-tenth
being SEN coordinators (SENCO). For supplementary information
on the TEIP scores, the Appendix lists the total and domain scores of
TEIP by demographic information. Bivariate associations between
the scores and demographic factors were tested with one-way
analysis of variance (ANOVA) and were tabulated.
One-way ANOVA was conducted to detect any differences in the
scores on the TEIP scale among the three sites, but no significant
differences were identified. In addition, DIF analysis was conducted
with the between-group mean square statistic (with corresponding
t value) for each item carefully examined to test for significantly
different items across sub-samples (Linacre, 2009). Again, no sig-
nificant DIF items were found. Hence, it seems plausible that par-
ticipants from different sub-samples demonstrated similar self-
efficacy for various inclusive teaching tasks.
3.2. Unidimensionality of teacher efficacy
A total of 44.0% of the raw variance in the data was explained by
the estimated Rasch rating scale model parameters. Residual
principal component analysis indicated that within the 56.0% un-
explained variance, the first contrast explained 7.6% variance with
an eigenvalue of 2.4. Therefore, unidimensionality seemed to hold
because the eigenvalue, indicative of the strength of a potential
secondary dimension, was less than three.
3.3. Inclusive teaching task hierarchy
The reliability of both person and item measures were good,
with person reliability being 0.89, Cronbach’s alpha being 0.90 and
item reliability being 0.94. This implied that while person measures
were consistent across different sets of items, item measures were
also consistent across different samples of participants. As shown in
Table 2, item measures generated from the model indicated that
item 16 (“I am confident in informing others who know little about
laws and policies relating to the inclusion of students with dis-
abilities”) was located at the highest level of self-efficacy, while
item 3 (“I can make parents feel comfortable coming to school”)
was at the lowest level. No apparent pattern of ranking of item
measures by domain was observed since the three domains were
fairly evenly spread across different levels of efficacy. Only item 1 (‘I
can make my expectations clear about student behavior’) was
found to be significantly unfit with a notable effect size according to
the criteria of Bond and Fox (2007). This was probably because the
phrasing of the item was not sufficiently specific to the inclusive
education context.
342 F.T.T. Lai et al. / Teaching and Teacher Education 59 (2016) 338e346

Table 2
Item hierarchy of teaching efficacy in inclusive practices (TEIP) scale with goodness-of-fit statistics.

Item Domain Measure SE Infit Outfit

MNSQ z MNSQ z

16 I am confident in informing others who know little about laws and policies CLB 1.41 0.15 1.09 0.7 1.11 0.8
relating to the inclusion of students with disabilities.
17 I am confident when dealing with students who are physically aggressive. MAB 1.41 0.15 1.19 1.3 1.2 1.4
9 I am confident in my ability to get parents involved in school activities of their CLB 0.78 0.16 1.12 0.8 1.17 1.2
children with disabilities.
7 I am confident in my ability to prevent disruptive behavior in the classroom MAB 0.65 0.17 0.89 0.7 0.9 0.7
before it occurs.
10 I am confident in designing learning tasks so that the individual needs of INS 0.51 0.17 0.56 3.5 þ 0.6 3.2 þ
students with disabilities are accommodated.
5 I can accurately gauge student comprehension of what I have taught. INS 0.22 0.17 0.84 1.1 0.83 1.1
8 I can control disruptive behavior in the classroom. MAB 0.19 0.17 0.76 1.8 0.72 2.1 þ
15 I can use a variety of assessment strategies (e.g. portfolio assessment, modified INS 0.16 0.17 0.86 0.9 0.86 1.0
tests, performance-based assessment, etc.).
12 I can collaborate with other professionals (e.g. itinerant teachers or speech CLB 0.01 0.17 1.32 2 1.3 1.9
pathologists) in designing educational plans for students with disabilities.
4 I can assist families in helping their children do well in school. CLB 0.17 0.18 0.88 0.8 0.9 0.6
6 I can provide appropriate challenges for very capable students. INS 0.17 0.18 0.89 0.7 0.88 0.8
2 I am able to calm a student who is disruptive or noisy. MAB 0.23 0.18 1.05 0.4 1.03 0.3
18 I am able to provide an alternate explanation or example when students are INS 0.42 0.18 1.08 0.6 1.09 0.6
confused.
13 I am able to work jointly with other professionals and staff (e.g. aides, other CLB 0.51 0.18 0.95 0.3 0.94 0.3
teachers) to teach students with disabilities in the classroom.
11 I am able to get children to follow classroom rules. MAB 0.54 0.18 0.89 0.7 0.9 0.7
14 I am confident in my ability to get students to work together in pairs or in small INS 0.67 0.18 1.13 0.9 1.15 1
groups.
1 I can make my expectations clear about student behavior. MAB 1.13 0.18 1.8 * 4.7 þ 1.66 * 3.9 þ
3 I can make parents feel comfortable coming to school. CLB 1.51 0.19 0.74 2 0.69 2.4 þ

Note. * denotes items that are beyond the recommended acceptable MNSQ range of 0.6e1.4; þ denotes high z score level reaching statistical significance. INS ¼ efficacy in
using inclusive instruction; CLB ¼ efficacy in collaboration; MAB ¼ efficacy in managing behavior; SE ¼ standard error; MNSQ ¼ mean square statistics.

The Wright map in Fig. 1 shows a lack of less endorsable items at
the higher end of the measured construct, i.e. self-efficacy. Highly
efficacious teachers were therefore not as well classified by efficacy
level as their less efficacious counterparts were.
3.4. Differential item functioning (DIF)
Table 3 demonstrates significant results from the DIF analyses
for the aforementioned grouping variables. The table also shows
that in each pairwise comparison, the group with an item measure
for that particular item was larger than the comparison group.
While DIF for all grouping variables yielded significant results, only
items 2, 7, 10, 11, 14, 15, 17 and 18 were found to be significant DIF
items. Given a similar overall level of efficacy, several significantly
different performances on different tasks across groups of partici-
pants were detected:
3.4.1. Gender and age of teachers
Males tended to be less efficacious in assigning accommodating
learning tasks (item 10), while young teachers (20e29 years) were
less efficacious in preventing disruptive behavior (item 7) than
older teachers (30e49 years). This was also the case for using
various assessments (20e29 years versus 40e49 years).

3.4.2. Qualifications of teachers

Teachers with a master’s degree were less efficacious in dealing
with physical aggression (item 17) than those with lower qualifi-
cations. However, they were more efficacious in assigning accom-
modating learning tasks (item 10) than teachers with a diploma/
certificate.

3.4.3. Teaching experience and position of teachers

The less experience the teacher possessed (2e9 years compared
with more than 10 years), the less efficacious they were in getting
students to follow classroom rules (item 11), whereas SENCOs were
even less efficacious in this area. Also, teachers with less than one
year of experience were more efficacious in getting students to
work in groups than those with 2e9 years of experience, and in
providing alternative explanations than those with ten years or
more experience.

4. Discussion
Fig. 1. Wright map showing the distribution of item measures relative to participants’
level of efficacy. To summarize the findings, a statistically sound positioning of
 F.T.T. Lai et al. / Teaching and Teacher Education 59 (2016) 338e346 343

Table 3
Extracts of differential item functioning analysis (significant) results.

Grouping variable Item Pairwise comparison Mantel-Haenszel

c2 p

Gender 10 Male * e Female 4.200 0.040

Age range 7 20e29 years old * e 30e39 years old 5.314 0.021
7 20e29 years old * e 40e49 years old 7.824 0.005
15 20e29 years old * e 40e49 years old 4.590 0.032
Qualification 10 Cert/Dip * e Master’s degree 4.840 0.028
17 Cert/Dip e Master’s degree * 4.651 0.031
17 Bachelor’s degree e Master’s degree * 4.425 0.035
Job position 2 SENCO e 10 year-experience * 8.399 0.004
11 10 year-experience e 2e9 year-experience * 7.312 0.007
11 SENCO * e 2e9 year-experience 4.840 0.028
14 2e9 year-experience * e 1 year experience 3.846 0.050
18 10 year-experience * e 1 year experience 6.642 0.010

Note. * denotes the group with higher item measure value for the corresponding item.

inclusive tasks according to the hierarchical location along the
continuum of overall teacher efficacy for inclusive practices was
derived. The task of ‘informing others about relevant educational
government policy and legislation’ was estimated to require the
highest level of efficacy, and ‘making parents feel comfortable
coming to school’ requires the lowest. Influences from the de-
mographic factors were also found in the DIF analyses. Besides, our
previous research hypothesis of high self-efficacy required by
collaboration with parents was confirmed, with the task of
involving parents in activities with disabled students (item 9)
ranked third among all 18 tasks in TEIP. Nevertheless, the hypoth-
esis of tasks related to disruption and aggression being ranked high
was not sufficiently supported by the data. While the task of
dealing with physical aggression (item 17) was ranked second,
calming disruptive students (item 7) was ranked the seventh
lowest.
This study is closely connected with the existing literature in the
sense that it utilized a well-validated scale with inclusive tasks that
have long been an important research focus (Florian & Rouse, 2009;
Jordan & Stanovich, 2001). It complements the literature as it took
one step further in examining the hierarchical ordering of the items
of various tasks using a Rasch analytic model, providing theoretical
and practical insights into the perceptions of the challenges faced
by teachers in inclusive classrooms. As far as the authors are aware,
it is the first of its kind in the field of study with an aim to
prompting future studies on similar topics.
4.1. Implications of results
Together with previous findings from the literature, the item
hierarchy derived from this study provides a dynamic representa-
tion of the continuum of teacher efficacy for inclusive practices. It is
therefore recommended that inclusive education training curricula
designers make reference to this hierarchy to help identify tasks
which are worth investment of resources. Specifically, training in
tasks should be prioritized to reflect both the significance of the
task for effective inclusive education and the required self-efficacy.
On one hand, concentration in training should be upon those tasks
that are essential to teachers’ daily practice. On the other hand,
training should reflect areas where teachers feel a lower level of
self-efficacy which is probably due to the lack of relevant skills and
experience: these areas can be identified at the upper end of the
hierarchy of required self-efficacy. Accordingly, it should follow
from the results that training on controlling and preventing
disruptive behavior (items 7, 8), dealing with physically aggressive
students (item 17), involving parents in school activities (item 9),
assigning accommodating learning tasks (item 10) and gauging
students’ comprehension (item 5) should receive the most atten-
tion in the design of curricula for teacher training on inclusive
education. On the other hand, getting students to follow rules (item
11), getting students to work in groups (item 14), calming
emotional students (item 2) and other tasks towards the lower end
of the efficacy continuum could have less focus in cases where the
duration of and resources for training are limited, because most
teachers would already be efficacious without additional training.
This study’s results on high efficacy demanding tasks are
consistent with previous findings. As discussed in earlier sections,
the behavioral problems of children with developmental disorders
such as ASD and ADHD have been an emerging area of extensive
research effort, and have drawn attention to the difficulties facing
the teachers of those students (Goldin, Matson, Tureck, Cervantes,
& Jang, 2013; Reiersen & Todd, 2008). It is, therefore, natural to
observe related tasks to require higher levels of self-efficacy.
Although there were also behavior management tasks that were
ranked near the lower end of the hierarchy, it actually reflected the
multifaceted nature of teaching where some aspects of managing
disruptive behavior were less demanding than others. It does not
affect the interpretation of the findings that, in fact, some disrup-
tive behavior management tasks require very high self-efficacy and
need more attention paid to them. Also, measuring the learning
achievements of students with SEN differently from the rest of the
students was identified as imposing difficulties for teachers ac-
cording to previous studies (Mayes & Calhoun, 2007). That is why
related tasks were near the upper end of the continuum of teacher
efficacy besides aggressive behavior management. In a primary
school context, getting students, even those with SEN, to follow
rules or to work in groups would be much less challenging since
regulations are relatively simple and easily communicated, and
group work is typically more interesting and motivating than the
usual traditional teaching approaches (Rohrbeck, Ginsburg-Block,
Fantuzzo, & Miller, 2003). It was therefore not surprising to find
that these two tasks were placed toward the lower end of the
continuum of teacher efficacy. It is also worth noting that there was
no obvious pattern in the hierarchical order of tasks by the domains
of the scale. This probably reflects the widely known fact that in-
clusive practices are highly demanding so that there are tasks that
require very high efficacy in each and every aspect of daily practice
(Foote & Collins, 2011; Westwood, 2013).
According to the Wright map, as shown in Fig. 1, there was an
apparent lack of less endorsable items to better discriminate more
efficacious teachers. Such a pattern was possibly due to the par-
ticipants’ greater experience and stronger motivation to teach
students with SEN compared to the rest of the teacher population:
participants in the study were mainly recruited on a voluntary basis
344 F.T.T. Lai et al. / Teaching and Teacher Education 59 (2016) 338e346
through advertisements targeting those who had some experience
with students with SEN, and were eager to acquire relevant useful
skills.
The DIF analysis demonstrates that even given the same overall
efficacy level, some personal factors such as the experience of
teaching and qualifications might exert specific effects on perfor-
mances in individual tasks as suggested by previous studies pre-
sented in earlier sections. The difference between previous and
current results arose from this study’s identification of specific in-
clusive teaching tasks to measure teacher efficacy. For example, it
was found that while teachers with extra teaching experience were
more effective when getting students to follow rules, teachers with
less experience tended to be more efficacious with particular
teaching approaches, such as getting students to work in groups.
This might imply that experienced teachers require additional skills
in various modern teaching approaches, especially in the context of
inclusive classrooms. It adds to the current knowledge by identi-
fying areas in which more experienced teachers are more effica-
cious (Malinen et al., 2013; Sharma, Shaukat, & Furlonger, 2015;
Shaukat et al., 2013). Also, a master’s degree seems to make
teachers more efficacious about designing accommodating tasks,
while it appears more difficult for teachers with such a qualification
to deal with physical aggression. This potentially points to the lack
of real-world practice with children with SEN in their master’s
training, and the lack of adequate preparation for accommodating
those children in learning activities. This finding provides new in-
formation on the specific areas where further training is needed to
improve inclusive teaching (Forlin et al., 2014; Sharma & Sokal,
2015; Sokal et al., 2013). These findings jointly suggest that teach-
ing experience or other personal qualities might only be helpful
with a few respective tasks. Since previous investigations on these
associations are scarce, further inquiries are needed to provide
more useful implications for the target audience of specific mod-
ules of inclusive education training. In particular, for course de-
signers of inclusive teacher education, it would be useful to know,
based on the profile of the audience, what the foci of the course
should be. For instance, according to the results of this study, an
audience with fewer qualifications and a higher proportion of
males might require more additional help in assigning accommo-
dating learning tasks (item 10).
4.2. Relevance to the social context of Hong Kong
Efficacy for different inclusive teaching tasks is highly context-
specific. For instance, as revealed by Savolainen et al. in a
comparative study (2012), Finnish teachers viewed managing
behavior as the weakest aspect of their teaching, while South Af-
rican teachers with comparable backgrounds rated it as their most
efficacious area. Therefore, the results of this study should be un-
derstood within the context of the socio-political factors in Hong
Kong. Although further research is needed to establish evident
causal links, several factors are believed to have played a role in
shaping the identified pattern of efficacy. First, it has been consis-
tently reported that Hong Kong teachers suffer serious stress levels
due to their heavy workload (Cheung & Wong, 2012; Jin et al.,
2008). Ad hoc tasks that fall outside their daily teaching routine
such as dealing with the sudden disruptive behavior of students
distorts their work schedule, further increasing their workload and
therefore requiring considerably more skills to handle. Second,
teachers in Hong Kong faced an average primary school class size of
27.0 in 2014/15 (Hong Kong Education Bureau, 2016), as compared
to 21.2 (SD ¼ 0.18) in the United States (National Center for
Education Statistics, 2014). Tasks that are individualized for a few
students are highly demanding considering the ongoing learning
progress of the rest of the class. Third, government support for
inclusive education has been identified as inadequate, with pro-
fessional support in particular reported to be lacking (Forlin & Sin,
2009). Currently, only a small proportion of teachers within each
school has been given relevant training (Forlin & Sin, 2009).
Therefore, assistance for less experienced teachers within the
school and from external parties is limited. Tasks that are beyond
the scope of teachers’ prior regular school teaching qualification
training might, therefore, require even more efficacy than in the
presence of adequate specialized support.
4.3. Limitations of the study
Firstly, convenience sampling reduces the generalizability of the
results. Nevertheless, there was a resemblance with the overall
study population in terms of demographic characteristics (Hong
Kong Education Bureau, 2016). Also, previous studies adopting
similar sampling methods have served as good examples of careful
investigation and interpretation; important implications can thus
be derived in spite of the adoption of convenience sampling (Forlin
et al., 2013). Secondly, there could be self-report bias. However, as
the concept of self-efficacy is by its very nature dependent upon an
individual understanding of self, there was no alternative method
for soliciting information on efficacy without using a self-report
measurement. Also, previous studies have already shown that the
effects of such bias are only trivial in size (Benítez-Silva, Buchinsky,
Man Chan, Cheidvasser, & Rust, 2004). Thirdly, we used efficacy as
the scale instead of real performance assessment to derive the
sequence which represents the hierarchy of the ‘difficulty’ of the
inclusive tasks. It is possible that although teachers’ efficacy for a
particular task is high, their actual performance may reveal the
opposite. However, it is very costly, if not completely unfeasible, to
perform assessments on ability relating to each inclusive task for a
sufficiently large sample of teachers. Also, it has been found that
self-efficacy consistently correlates with actual performance, even
in field settings and for highly complex tasks (Judge, Jackson, Shaw,
Scott, & Rich, 2007; Stajkovic & Luthans, 1998). Fourthly, a modest
sample size was adopted as compared to other studies in the field.
Although a good reliability of estimates was shown and the model
fit the data well, future studies should increase the sample size for
larger power in the analyses.
4.4. Recommendations for future research
Based on the study results and interpretations as discussed,
there are several future research directions recommended for the
research community. These relate to the facilitation of a more
evidence-based practice in designing inclusive education training
for teachers.
4.4.1. More specific and detailed task descriptions
While this study derived an item hierarchy of the TEIP and
identified a few tasks regarding physical aggression, prevention of
disruptive behavior and involvement of parents as the ones that
require the highest efficacy, it did not inform us about which parts
of the tasks teachers were less efficacious in. For instance, within
the task of dealing with physical aggression, does stopping the
student’s behavior or protecting other students impose more
challenges to teachers? By increasing the specificity of the task
descriptions, more details could be unearthed for better design of
inclusive education training. Thus, future research should cover a
wider yet more specific and detailed range of inclusive tasks in the
analyses. This might involve expansion of the adopted scale (TEIP)
by including sub-tasks under the existing tasks for further analysis.
 F.T.T. Lai et al. / Teaching and Teacher Education 59 (2016) 338e346
4.4.2. Additional diverse stratification variables
As preliminarily suggested by the results of this current study,
the item hierarchy, i.e. ranking of inclusive tasks by the level of
required self-efficacy, was subject to the influence of different
stratification variables such as gender and teaching experience or
qualifications. By including more such variables, the richness of the
data and analyses could be enhanced. For example, if there is data
from two countries with similar inclusive education policies, it
could be adopted as a stratification variable which would enable
comparative research into the differences in the hierarchy of tasks
between the two countries. Other stratification variables that
would potentially yield interesting implications are types of
schools, academic ranking of schools, the religion of teachers, etc.
4.4.3. Longitudinal design
Adding temporality to the analyses would be insightful in the
sense that the change in the hierarchy of tasks by required self-
efficacy over time could be observed. For example, if data were
collected once a year over a ten-year period, the change in how the
inclusive teaching tasks are ranked would imply a change in the
needs of inclusive training or additional assistance in the teachers’
teaching. This could also help us capture which tasks could be
mastered with the accumulation of experience alone and which
could not. For the government authorities, this is a useful practice
to keep track of teacher’s self-efficacy for implementing inclusive
education. This could serve as an evaluative measure of inclusive
education as well as solid evidence based on which further training
on different aspects could be provided.
5. Conclusion
In this study, with the aim of facilitating better evidence-based
inclusive education training for teachers, a sequence of inclusive
tasks according to the level of efficacy required for implementation
was derived with good reliability and fitness to data. It could be
used for identifying the most urgent needs for additional training
for teachers who lack specific skills and knowledge on particular
tasks in inclusive education, such as dealing with physically
aggressive students, involving parents in activities with disabled
children and the prevention of disruptive behavior. By referencing
the results and interpretation of this study, curriculum design could
be more focused on those tasks which require additional assistance.
Also, the DIF analysis reveals the pattern by which the background
of teachers influences their self-efficacy for individual tasks. Cour-
ses on inclusive education could hence be better targeted at those
who need it most urgently.
Acknowledgements
This study was funded by the General Research Fund under the
University Grants Committee (UGC) of Hong Kong (Project number:
843112). The UGC played no role in the study design, data collec-
tion, analysis and interpretation, writing of the report, or the de-
cision to submit the article for publication.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.tate.2016.07.006.
References
Ainscow, M., & Sandill, A. (2010). Developing inclusive education systems: The role
of organisational cultures and leadership. International Journal of Inclusive Ed-
ucation, 14(4), 401e416.
345
Almog, O., & Shechtman, Z. (2007). Teachers’ democratic and efficacy beliefs and
styles of coping with behavioural problems of pupils with special needs. Eu-
ropean Journal of Special Needs Education, 22(2), 115e129.
Andrich, D. (1978). Application of a psychometric rating model to ordered cate-
gories which are scored with successive integers. Applied Psychological Mea-
surement, 2(4), 581e594.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change.
Psychological Review, 84(2), 191e215.
Bandura, A. (2006). Guide for constructing self-efficacy scales. In F. Pajares, &
T. C. Urdan (Eds.), Self-efficacy beliefs of adolescents (pp. 307e337). Charlotte, NC:
Information Age Publishing.
Bearss, K., Johnson, C., Handen, B., Smith, T., & Scahill, L. (2013). A pilot study of
parent training in young children with autism spectrum disorders and
disruptive behavior. Journal of Autism and Developmental Disorders, 43(4),
829e840.
Benítez-Silva, H., Buchinsky, M., Man Chan, H., Cheidvasser, S., & Rust, J. (2004).
How large is the bias in self-reported disability? Journal of Applied Econometrics,
19(6), 649e670.
Bond, T. G., & Fox, C. M. (2007). Applying the Rasch model: Fundamental measurement
in the human sciences. Mahwah, NJ: Lawrence Erlbaum Associates Publishers.
Broderick, A., Mehta-Parekh, H., & Reid, D. K. (2005). Differentiating instruction for
disabled students in inclusive classrooms. Theory into Practice, 44(3), 194e202.
Brophy, J. (2003). Teaching problem students. New York, NY: Guilford Press.
Chan, D. W. (2006). Emotional intelligence and components of burnout among
Chinese secondary school teachers in Hong Kong. Teaching and Teacher Educa-
tion, 22(8), 1042e1054.
Cheung, A. C., & Wong, P. M. (2012). Factors affecting the implementation of cur-
riculum reform in Hong Kong: Key findings from a large-scale survey study.
International Journal of Educational Management, 26(1), 39e54.
Counts, C. A., Nigg, J. T., Stawicki, J. A., Rappley, M. D., & Von Eye, A. (2005). Family
adversity in DSM-IV ADHD combined and inattentive subtypes and associated
disruptive behavior problems. Journal of the American Academy of Child &
Adolescent Psychiatry, 44(7), 690e698.
Crozier, S., & Sileo, N. M. (2005). Encouraging positive behavior with social stories.
Teaching Exceptional Children, 37(6), 26e31.
Crozier, S., & Tincani, M. J. (2005). Using a modified social story to decrease
disruptive behavior of a child with autism. Focus on Autism and Other Devel-
opmental Disabilities, 20(3), 150e157.
Eikeseth, S., Klintwall, L., Jahr, E., & Karlsson, P. (2012). Outcome for children with
autism receiving early and intensive behavioral intervention in mainstream
preschool and kindergarten settings. Research in Autism Spectrum Disorders,
6(2), 829e835.
Engelbrecht, P., Oswald, M., & Forlin, C. (2006). Promoting the implementation of
inclusive education in primary schools in South Africa. British Journal of Special
Education, 33(3), 121e129.
Engelbrecht, P., Swart, E., & Eloff, I. (2006). Stress and coping skills of teachers with a
learner with Down’s syndrome in inclusive classrooms. South African Journal of
Education, 21(4), 256e259.
Falkmer, M., Anderson, K., Joosten, A., & Falkmer, T. (2015). Parents’ perspectives on
inclusive schools for children with autism spectrum conditions. International
Journal of Disability, Development and Education, 62(1), 1e23.
Florian, L., & Rouse, M. (2009). The inclusive practice project in Scotland: Teacher
education for inclusive education. Teaching and Teacher Education, 25(4),
594e601.
Foote, C. J., & Collins, B. (2011). You know, Eunice, the world will never be the same
after this. International Journal of Special Education, 26(3), 285e295.
Forlin, C., Loreman, T., & Sharma, U. (2014). A system-wide professional learning
approach about inclusion for teachers in Hong Kong. Asia-Pacific Journal of
Teacher Education, 42(3), 247e260.
Forlin, C., Sharma, U., & Loreman, T. (2013). Predictors of improved teaching efficacy
following basic training for inclusion in Hong Kong. International Journal of
Inclusive Education, 18(7), 718e730.
Forlin, C., & Sin, K. (2009). Developing support for inclusion: A professional learning
approach for teachers in Hong Kong. International Journal of Whole Schooling,
6(1), 7e26.
Gibbs, S., Billington, T., Greig, A., Mackay, T., Munn, P., & Stringer, P. (2007). Teachers’
perceptions of efficacy: Beliefs that may support inclusion or segregation.
Educational & Child Psychology, 24(3), 47e53.
Goldin, R. L., Matson, J. L., Tureck, K., Cervantes, P. E., & Jang, J. (2013). A comparison
of tantrum behavior profiles in children with ASD, ADHD and comorbid ASD
and ADHD. Research in Developmental Disabilities, 34(9), 2669e2675.
Hong Kong Education Bureau. (2016). Figures and statistics e Primary education.
Retrieved from http://www.edb.gov.hk/mobile/en/about-edb/publications-stat/
figures/pri.html.
Jin, P., Yeung, A. S., Tang, T.-O., & Low, R. (2008). Identifying teachers at risk in Hong
Kong: Psychosomatic symptoms and sources of stress. Journal of Psychosomatic
Research, 65(4), 357e362.
Jordan, A., & Stanovich, P. (2001). Patterns of teacher-student interaction in inclu-
sive elementary classrooms and correlates with student self-concept. Interna-
tional Journal of Disability, Development and Education, 48(1), 33e52.
Judge, T. A., Jackson, C. L., Shaw, J. C., Scott, B. A., & Rich, B. L. (2007). Self-efficacy and
work-related performance: The integral role of individual differences. The
Journal of Applied Psychology, 92(1), 107e127.
Kaat, A. J., Lecavalier, L., & Aman, M. G. (2014). Validity of the aberrant behavior
checklist in children with autism spectrum disorder. Journal of Autism and
346 F.T.T. Lai et al. / Teaching and Teacher Education 59 (2016) 338e346
Developmental Disorders, 44(5), 1103e1116.
Karkhaneh, M., Clark, B., Ospina, M. B., Seida, J. C., Smith, V., & Hartling, L. (2010).
Social Stories™ to improve social skills in children with autism spectrum dis-
order: A systematic review. Autism, 14(6), 641e662.
Kern, L., & Clemens, N. H. (2007). Antecedent strategies to promote appropriate
classroom behavior. Psychology in the Schools, 44(1), 65e75.
Klusmann, U., Kunter, M., Trautwein, U., Lüdtke, O., & Baumert, J. (2008). Teachers’
occupational well-being and quality of instruction: The important role of self-
regulatory patterns. Journal of Educational Psychology, 100(3), 702e715.
Lacey, P., & Scull, J. (2015). Inclusive education for learners with severe, profound
and multiple learning difficulties in England. In C. Forlin (Ed.), Including learners
with low-incidence disabilities (pp. 241e268). Bingley, UK: Emerald Group
Publishing Limited.
Lancaster, J., & Bain, A. (2007). The design of inclusive education courses and the
self-efficacy of preservice teacher education students. International Journal of
Disability, Development and Education, 54(2), 245e256.
Levi, U., Einav, M., Raskind, I., Ziv, O., & Margalit, M. (2013). Helping students with
LD to succeed: The role of teachers’ hope, sense of coherence and specific self-
efficacy. European Journal of Special Needs Education, 28(4), 427e439.
Linacre, J. M. (2009). Winsteps user manual and on-line help. Retrieved from http://
www.winsteps.com/.
Linacre, J. M. (2014). Winsteps® (Version 3.92.0) [Computer software]. Beaverton, OR:
Winsteps.com.
Logan, S. L., Carpenter, L., Leslie, R. S., Garrett-Mayer, E., Hunt, K. J., Charles, J., et al.
(2015). Aberrant behaviors and co-occurring conditions as predictors of psy-
chotropic polypharmacy among children with autism spectrum disorders.
Journal of Child and Adolescent Psychopharmacology, 25(4), 323e336.
Malinen, O. P., Savolainen, H., Engelbrecht, P., Xu, J., Nel, M., Nel, N., et al. (2013).
Exploring teacher self-efficacy for inclusive practices in three diverse countries.
Teaching and Teacher Education, 33, 34e44.
Mayes, S. D., & Calhoun, S. L. (2007). Learning, attention, writing, and processing
speed in typical children and children with ADHD, autism, anxiety, depression,
and oppositional-defiant disorder. Child Neuropsychology, 13(6), 469e493.
National Center for Education Statistics. (2014). Highest degree earned, years of full-
time teaching experience, and average class size for teachers in public elementary
and secondary schools, by state: 2011e12 (table 209.30). Retrieved from http://
nces.ed.gov/programs/digest/d14/tables/dt14_209.30.asp.
Oliver, R. M., Wehby, J. H., & Reschly, D. J. (2010). Teacher classroom management
practices: Effects on disruptive or aggressive student behavior. Oslo, Norway: The
Campbell Collaboration.
O’Toole, C., & Burke, N. (2013). Ready, willing and able? Attitudes and concerns in
relation to inclusion amongst a cohort of Irish pre-service teachers. European
Journal of Special Needs Education, 28(3), 239e253.
Pang, I. W. (2011). Teacher stress in working with challenging students in Hong
Kong. Educational Research for Policy and Practice, 11(2), 119e139.
Park, M., Dimitrov, D. M., Das, A., & Gichuru, M. (2014). The teacher efficacy for
inclusive practices (TEIP) scale: Dimensionality and factor structure. Journal of
Research in Special Educational Needs, 16(1), 2e12.
Pivik, J., McComas, J., & Laflamme, M. (2002). Barriers and facilitators to inclusive
education. Exceptional Children, 69(1), 97e107.
Rasch, G. (1966). An item analysis which takes individual differences into account.
British Journal of Mathematical and Statistical Psychology, 19(1), 49e57.
Reiersen, A. M., & Todd, R. D. (2008). Co-occurrence of ADHD and autism spectrum
disorders: Phenomenology and treatment. Expert Review of Neurotherapeutics,
8(4), 657e669.
Rohrbeck, C. A., Ginsburg-Block, M. D., Fantuzzo, J. W., & Miller, T. R. (2003). Peer-
assisted learning interventions with elementary school students: A meta-
analytic review. Journal of Educational Psychology, 95(2), 240e257.
Savolainen, H., Engelbrecht, P., Nel, M., & Malinen, O. P. (2012). Understanding
teachers’ attitudes and self-efficacy in inclusive education: Implications for pre-
service and in-service teacher education. European Journal of Special Needs
Education, 27(1), 51e68.
Scattone, D., Wilczynski, S. M., Edwards, R. P., & Rabian, B. (2002). Decreasing
disruptive behaviors of children with autism using social stories. Journal of
Autism and Developmental Disorders, 32(6), 535e543.
Shah, R., Das, A., Desai, I., & Tiwari, A. (2016). Teachers’ concerns about inclusive
education in Ahmedabad, India. Journal of Research in Special Educational Needs,
16(1), 34e45.
Sharma, U., Forlin, C., & Loreman, T. (2008). Impact of training on pre-service
teachers’ attitudes and concerns about inclusive education and sentiments
about persons with disabilities. Disability & Society, 23(7), 773e785.
Sharma, U., Loreman, T., & Forlin, C. (2012). Measuring teacher efficacy to imple-
ment inclusive practices. Journal of Research in Special Educational Needs, 12(1),
12e21.
Sharma, U., Shaukat, S., & Furlonger, B. (2015). Attitudes and self-efficacy of pre-
service teachers towards inclusion in Pakistan. Journal of Research in Special
Educational Needs, 15(2), 97e105.
Sharma, U., & Sokal, L. (2015). The impact of a teacher education course on pre-
service teachers’ beliefs about inclusion: An international comparison. Journal
of Research in Special Educational Needs, 15(4), 276e284.
Shaukat, S., Sharma, U., & Furlonger, B. (2013). Pakistani and Australian pre-service
teachers’ attitudes and self-efficacy towards inclusive education. Journal of
Behavioural Sciences, 23(2), 1e16.
Sokal, L., Woloshyn, D., & Funk-Unrau, S. (2013). How important is practicum to pre-
service teacher development for inclusive teaching? Effects on efficacy in
classroom management. Alberta Journal of Educational Research, 59(2),
285e298. Retrieved from http://www.ajer.ca/.
Stajkovic, A. D., & Luthans, F. (1998). Self-efficacy and work-related performance: A
meta-analysis. Psychological Bulletin, 124(2), 240e261.
Tait, K., & Mundia, L. (2013). A comparison of Brunei and Hong Kong-SAR student
teachers’ self-efficacy in implementing inclusive education practices: Implica-
tions for teacher education. Asian Social Science, 10(1), 51e60.
Talmor, R., Reiter, S., & Feigin, N. (2005). Factors relating to regular education
teacher burnout in inclusive education. European Journal of Special Needs Edu-
cation, 20(2), 215e229.
Virues-Ortega, J., Julio, F. M., & Pastor-Barriuso, R. (2013). The TEACCH program for
children and adults with autism: A meta-analysis of intervention studies.
Clinical Psychology Review, 33(8), 940e953.
Westwood, P. (2013). Inclusive and adaptive teaching: Meeting the challenge of di-
versity in the classroom. Oxford, UK: Routledge.
Winstanley, C. A., Eagle, D. M., & Robbins, T. W. (2006). Behavioral models of
impulsivity in relation to ADHD: Translation between clinical and preclinical
studies. Clinical Psychology Review, 26(4), 379e395.
Wong, D. K. P. (2002). Struggling in the mainstream: The case of Hong Kong. In-
ternational Journal of Disability, Development & Education, 49(1), 79e94.