Pertanika J. Soc. Sci. & Hum.

29 (S3): 309 - 334 (2021)

SOCIAL SCIENCES & HUMANITIES

Journal homepage: [Link]

English Language Speaking Anxiety, Self-Confidence and

Perceived Ability among Science and Technology Undergraduate
Students: A Rasch Analysis
Kamal J I Badrasawi1*, Noor Lide Abu Kassim1, Ainol Madziah Zubairi1,
Elia Md Johar2 and Siti Sakinah Sidik1
Kulliyyah of Education, International Islamic University Malaysia (IIUM), 50728 Kuala Lumpur, Malaysia
1

Akademi Pengajian Bahasa, Universiti Teknologi MARA (UiTM), 40450 Shah Alam, Selangor, Malaysia
2

ABSTRACT
The purpose of this paper is to analyse English language speaking anxiety, self-confidence,
and perceived ability in English oral communication among Science and Technology
undergraduate students. It also aims to identify any significant differences in these constructs
based on selected students’ demographic variables. The study employed the survey method
with a 41-item questionnaire administered to a voluntary response sample of three hundred
3rd and 4th-year science and technology undergraduates from three Malaysian public
universities. The Polytomous Rasch model was used to analyse the data. The analysis
showed that the participants experienced English speaking anxiety, low confidence,
and high perceived ability in English oral communication. There were significant mean
differences in English speaking anxiety across the type of university as well as in confidence
and perceived ability based on academic program. The participants were more confident
and could perform better in familiar situations and communicate on familiar topics to
familiar audiences. The findings suggest
that the participants need more training on
ARTICLE INFO
English oral communication. More authentic
Article history: situations are also needed for them to practise
Received: 16 July 2021
Accepted: 04 October 2021 and improve their proficiency levels. Other
Published: 30 November 2021
suggestions include providing lecturers with
DOI: [Link]
training modules, re-assessing the current
E-mail addresses:
kamalbadrasawi@[Link] (Kamal J I Badrasawi) language policies, and implementing certain
noorlide@[Link] (Noor Lide Abu Kassim)
ainol@[Link] (Ainol Madziah Zubairi)
programmes at the tertiary education level.
elia@[Link] (Elia Md Johar) Language programmes could be directed
g1716646@[Link] (Siti Sakinah Sidik)
* Corresponding author towards more social situations to enable

ISSN: 0128-7702
e-ISSN: 2231-8534 © Universiti Putra Malaysia Press
 Kamal J I Badrasawi, Noor Lide Abu Kassim, Ainol Madziah Zubairi, Elia Md Johar and Siti Sakinah Sidik

undergraduates to make English a social language in the world (Hudson & Hudson,
practice, lower English speaking anxiety, 2003). Companies and institutions hire
and boost confidence. employees who can communicate in the
English language efficiently within a wide
Keywords: Confidence, English oral communication
skill, perceived ability, Rasch Model Analysis, science range of workplace communicative events
and technology undergraduates, speaking anxiety due to the substantial roles the language
plays in the current worldwide transactions
INTRODUCTION (Pandarangga, 2015; Sheth, 2016). The same
The English language is considered a idea is mooted in Roshid and Chowdhury’s
lingua franca due to its dominance in many (2013) notion that employers are looking for
social, economic, scientific and political graduates with high English communication
activities (Crystal, 2004; Nishanti, 2018; skills, mainly those who can explain ideas,
Pandarangga, 2015; Pennycook, 2014; identify issues, and solve problems related
Reddy, 2016). Reports have shown that to their work constructively. For instance,
about 750 million people use English as a English communication skill is an imperative
second language (ESL), and it is prioritised employability requirement in India to
in around 70 countries in the world (Reddy, get a better job (Clement & Murugavel,
2016). Furthermore, English language 2015). Clement and Murugavel further
learners have reached 1 billion, while emphasize that engineering graduates can
roughly 2 billion people have mastered only internationally communicate if they are
the language, with an estimation that half proficient in English communication skills,
of the world population might be English mainly those related to their profession.
proficient in the next few decades (English Wijewardene et al. (2014) assert that
Cultural Council as in Xue & Zuo, 2013). competency in English—especially the
The English language has also become spoken and written—is among the crucial
the medium of instruction in many higher factors determining graduates’ employment
education institutions, and it has also been in the private and public sectors in Sri Lanka.
used as a criterion with which students The same trend is seen almost worldwide.
secure admissions into tertiary education Of the four language skills, speaking
programs (Pandarangga, 2015). As such, skill is considered the most important.
English communication competency is Ur (1996) argues that those who know
considered among the highly significant the language speak it, implying that it is
requirements of university graduates’ important to use it effectively rather than
quality in Asia (UNESCO, 2012). knowing it (Scrivener, 2005). We live in
Globally, many countries are constantly a time where the need to speak English
working to improve English language fluently is dire, especially for those who
proficiency among their people and learners, want to advance in certain fields of human
as English has become the most important endeavours (Al-Sibai, 2004). The literature

310 Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021)
 A Rasch Analysis on Speaking Anxiety, Self-Confidence and Ability

on English oral proficiency has shown 2002). If individuals fail to minimise this
several factors affecting its improvement. negative feeling, it becomes permanent in
These factors include speaking anxiety their temperament (Riasati, 2011). On the
(Ahmed et al., 2017; Bux et al., 2015; other hand, situational anxiety is anxiety
Dordinejad & Ahmadabad, 2014; Kumar, expressed by a second language learner in
2018; Ramamuruthy, 2019; Salem & Al specific situations (MacIntyre, 1999). This
Dyiar, 2014; Zhang & Zhong, 2012); low kind of anxiety is usually felt because of the
self-confidence (Gürler, 2015; Kalanzadeh lack of familiarity of language learners with
et al., 2013; Mandokhail et al., 2018; the particular situation in which they are to
Tridinanti, 2018) and perceived ability in use the language, and once they become
English oral communication (Alawiyah, familiar with it, their anxiety diminishes or
2018; Linnenbrink & Pintrich, 2003; even vanishes (Riasati, 2011).
Pajares, 1996; Sunyi, 2017; Zahiri et al., Interestingly, speaking skills in a second
2017). This study primarily aims to analyze or foreign language is the most anxiety-
English language speaking anxiety, self- inducing skill among individuals (Zhang
confidence, and perceived ability in English & Zhong, 2012). Furthermore, research
oral communication among science and has revealed a connection between English
technology undergraduate students using speaking anxiety and students’ speaking
the Rasch Model in Malaysia. fluency; meaning that speaking anxiety
has an unfavourable effect on learners’
English Language Anxiety and its English-speaking fluency (Salem & Al
Effects on Spoken/Oral Interaction Dyiar, 2014) and, in particular, students’ oral
Horwitz et al. (1986) describe language presentation as well as their conversation
anxiety as a multiplex phenomenon of with English native speakers (Bux et al.,
“self-perceptions, beliefs, feelings, and 2015). Therefore, a number of research
behaviours related to classroom language (quantitative and qualitative) has been
learning arising from the uniqueness of conducted to identify the causes of speaking
the language learning process” (p.128). anxiety, its destructive effects on speaking as
Spielberger (1983) defines language well as how the problem can be addressed
anxiety as “a subjective feeling of tension, (Ahmad et al., 2017; Bux et al., 2015;
apprehension, nervousness, and worry Kumar, 2018; Ramamuruthy, 2019; Salem
associated with an arousal of the autonomic & Al Dyiar, 2014). For instance, Ahmad
nervous system” (p.1). It could be classified et al. (2017) found the inter-language
into two categories, namely ‘trait anxiety’ meaning system as the cause of postgraduate
and ‘situational anxiety’ (i.e., state anxiety) students’ English language speaking anxiety,
(Akkakoson, 2016; Spielberger, 1983). Trait while Kamaruddin et al. (2019) found low
anxiety is the anxiety that language learners self-esteem and social anxiety as the factors
experience in every situation (Pappamihiel, contributing to the listening and speaking

Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021) 311
 Kamal J I Badrasawi, Noor Lide Abu Kassim, Ainol Madziah Zubairi, Elia Md Johar and Siti Sakinah Sidik

anxieties among Malaysian university Perceived Ability and its Effects on

students majoring in non-English programs. Spoken/Oral Interaction
Moreover, Mulyono et al. (2019) identified Perceived ability refers to individuals’ self-
negative attitudes, language barriers and perception regarding their ability to perform
intercultural communication apprehension effectively in a specific situation based on
as the main factors provoking English- their skills and capabilities. A good example
speaking anxiety among non-English is the Common European Framework of
native speakers studying in the Indonesian References (CEFR), where perceived ability
universities. is directly linked to learners’ ability to
Moreover, Ramamuruthy (2019) found achieve the stipulated ‘Can do’ statements
fear of being negatively evaluated as the (Alderson, 2017). The CEFR describes what
main factor for English speaking anxiety learners can do across five language skills:
among diploma students studying at an spoken production, spoken interaction,
international college in Malaysia. Amiri reading, listening and writing (Alderson,
and Puteh (2018) found that several factors 2017). For all five skills at each level, there
are sets of detailed ‘Can Do’ statements. For
such as insufficient linguistic competency,
example, the spoken interaction focuses on
inadequate knowledge of the presentation’s
the learners’ production and participation
content, students’ negative perception
in conversations and discussions. Perceived
towards the examiners, and examiners’
ability is important for students to participate
linguistic deficiency in understanding
in maintaining, starting, taking turns and
presentations lead to speaking anxiety
ending conversations. Without adequate
among international students studying
perceived ability, learners will stumble
doctoral programs in different Malaysian
thus fail to achieve the target performance
universities. Sadighi and Dastpak (2017)
stipulated in the ‘can do’ statements.
found that fear of making mistakes, being
Literature has shown that perceived
negatively evaluated, and inadequate ability is among the factors that affect
vocabulary knowledge were the major students’ English oral communication
causes of speaking anxiety among ESL skills. This is not because of their deficiency
Iranian students. It is important to maintain in mastering the linguistic aspect of
that speaking anxiety level might differ due communication, but it is as a result of
to demographic variables (Badrasawi et al., the negative feeling they usually have
2020). In conclusion, it could be inferred about themselves and the audience such
that language anxiety affects the speaking as communication apprehension, fear of
performance of ESL/EFL learners, and the making mistakes and fear of negative
higher the level of language anxiety, the evaluation as affirmed by Horwitz et al.
worse their performance in speaking will (1986). Speaking anxiety tends to be higher
be, and vice versa. when the ESL/EFL speakers perceive their

312 Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021)
 A Rasch Analysis on Speaking Anxiety, Self-Confidence and Ability

speaking ability to be low, and vice versa. Confidence and its Effects on Spoken/
Cognitive component, as emphasised by Oral Communication
Shrauger and Schohn (1995), is an integral Koriat et al. (1980) describe confidence
part of individuals’ perceived ability. It as the belief in oneself to perform tasks
refers to the self-evaluation of performance, successfully. It could also refer to one’s
meeting of own expectations and continuous realistic sense of capacity and possessing
excellence compared to others. Several sufficient knowledge. Brown (2004) stresses
studies have contended that English that successful activities require a high
language learners who perceive their ability level of self-confidence, including second
as high are confident to successfully engage language acquisition (Kalanzadeh et al.,
themselves in English conversations with 2013). Hart (1989) purports that confident
others expressing their ideas overtly, and learners would most likely get the task done.
Confidence in spoken interaction is usually
the reverse is also true (Alawiyah, 2018).
associated with the speakers’ certainty
When people are optimistic in
about using the language. Thus, confidence
performing excellently in a task, they
plays a crucial role in motivating learners
become strongly motivated and driven by
to communicate (Tanveer, 2007). The
their enthusiasm and interest to achieve
higher the confidence, the most likely the
the desired goals (Linnenbrink & Pintrich,
learners will be involved in communication
2003; Pajares, 1996; Sunyi, 2017). Zahiri
activities. Tsou (2005) reported that high
et al. (2017) studied the effects of speaking
self-confidence was positively correlated
anxiety and perceived ability on monologue with oral performance and concluded
speaking skills of students at a state senior that self-confidence is crucial in learners’
Islamic high school in Medan, Indonesia. inclination to communicate. It is supported
They found that both speaking anxiety by Stenstrom (2014), who asserts that
and perceived ability affected students’ confidence is important in spoken interaction
monologue speaking skills. While the former as it is a two-way process. Other researchers
adversely affected students’ speaking skills, found a positive, strong relationship between
the latter had a positive effect on it. The self-confidence or self–esteem of FL or
more anxious students are in speaking, the SL learners and their oral proficiency of
worse the speaking skills will be. Similarly, speaking skills (Gürler, 2015; Mandokhail
the higher their perception of their ability, et al., 2018; Tridinanti, 2018).
the better they will be in speaking skills.
Desmaliza and Septiani (2017) showed a Study Setting
significant relationship between students’ In Malaysia, English has its status as an
perceived ability and their speaking skills, important second language, and it is used
positively influencing the performance of as the medium of instruction for science
students in oral communication activities. and technology in higher institutions. In

Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021) 313
 Kamal J I Badrasawi, Noor Lide Abu Kassim, Ainol Madziah Zubairi, Elia Md Johar and Siti Sakinah Sidik

addition, it is extensively used in various measurement framework, educational

settings, for instance, social, commercial institutions and employers can easily
and national and international transactions. compare qualifications to other exams in
The Malaysian educational system values their countries. The minimum expectation
English language acquisition among for university graduates is B2 which ensures
students at all school levels, and the they can understand more complex texts,
Malaysia Education Blueprint (2013- handle more abstract topics and technical
2025) has stressed improving the students’ discussions, and communicate and interact
English language proficiency at all stages. comfortably with native speakers. C2 is the
Considerable efforts have been devoted highest level CEFR scale of achievement,
to improving graduates’ English language required only for those entering certain
proficiency to work in a globalised economy professions, such as English language
whereby the English language is the teachers.
international language of communication, However, undergraduate Science and
as clearly mooted in the Blueprint 2013- Technology students must be concerned
2025 (Ministry of Education, 2012). More because a recent report has indicated
importantly, this focus encourages the an urgent need for more qualified and
graduates to participate in the workforce and skilled graduates in science, technology,
contribute to the country’s development in engineering and mathematics (Chin, 2016).
the future. Graduates proficient in English Overall, extensive research has shown that
and who have leadership and technical skills undergraduate students have low proficiency
get more opportunities to find a job in their (i.e. not up to the desired levels) in the
respective fields (Ismail, 2011). English language (Ismail, 2011; Musa et al.,
In order to improve the level of English 2012; Nair et al., 2012, Rusli et al., 2018).
language proficiency from preschool to The language competence of Malaysian
tertiary education, Malaysians have adopted undergraduates is still a long way from
CEFR with the establishment of the English satisfactory level though they have learned
Language Standards and Quality Council English for 11 to 13 years in schools (David
(ELSQC) in 2013. Malaysian Science and et al., 2015). As a result, communication
Technology undergraduates went through skills amongst Malaysian graduates have
lessons for ESL and CEFR-aligned tests deteriorated (Shakir, 2009). The Malaysia
such as Malaysian University English Education Blueprint states that “poor
Test (MUET), Cambridge Placement Test English proficiency among graduates has
(CPT) and an English assessment test been consistently ranked as one of the top
administered by the British Council (Aptis) five issues facing Malaysian employers
before graduating. The purpose of taking since 2006” (Ministry of Education, 2012,
at least one of these tests is primarily to p.12). In addition, recent reports have shown
measure their proficiency in the language. that the number of unemployed Malaysian
Since CEFR provides a globally accepted graduates are increasing due to their lack

314 Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021)
 A Rasch Analysis on Speaking Anxiety, Self-Confidence and Ability

of the required levels of English speaking Reference (CEFR) Can-Do Statement

skill to get or secure a job (Free Malaysia rubrics was applied to measure the perceived
Today, 2017; The Sun Daily, 2018). Besides, ability. Altogether, there were 41 items on
the Salary Surveys 2016 by the Malaysian a five-point Likert scale, categorised into
Employers Federation (MEF) found that three sections: English speaking anxiety
over 90% of respondents were required to (ANX) (12 items); Confidence in oral
improve their English capability to get a job communication/interaction (CON) (19
(Malaysian Employers Federation, 2016). items); Can-Do statements (perceived ability
Evidently, in such pressing situations in oral communication) (CAN) (10 items).
where further investigation is needed, this These items were piloted, and the Cronbach’s
study aims to analyse English language Alpha values for the three constructs were
s p e a k i n g a n x i e t y, s e l f - c o n f i d e n c e , 0.87, 0.74 and 0.87, respectively. Two
and perceived ability in English oral Malaysian research universities and a
communication among undergraduate comprehensive university offering Science,
ICT and Engineering programmes were
science and technology students using
identified as the study population. Third-
the Rasch Model. Also, it aims to find the
and fourth-year undergraduate students
significant differences in mean scores of
from those academic programmes were
the factors about selected demographic
invited to participate in the study, from
variables (i.e. gender, academic year,
which three hundred students volunteered.
university type/category and faculty).
The questionnaires were administered in
person to those who agreed to participate—
METHODOLOGY
However, some who were unable to join
This study used the survey method to answered the questionnaire via Google
determine English language speaking Forms. The breakdown of the sample
anxiety, self-confidence and perceived in terms of the programme of study and
ability in English oral communication among institution is presented in Table 1.
science and technology undergraduate The collected data were analysed
students. Based on the related literature, based on the Polytomous Rasch model
items measuring Foreign Language Anxiety using Winsteps version 4.1.0 (Linacre,
Scale (FLAS) were pooled from previous 2018). Unlike other kinds of analysis,
studies (Ali, 2017, McCroskey, 1970; interval data are always used in Rasch
Pappamihiel, 2002; Yim & Yu, 2011) while analysis. It uses logit units; therefore, it
items measuring Confidence and Task is possible to get the difficulty to measure
Difficulty were based on a questionnaire for each item and for category. All persons
that was developed to measure confidence and items are placed on the same interval
and task difficulty in oral proficiency testing, scale to see their distributions. The most
a study by Kassim and Zubairi (2003). difficult items to endorse are positioned
The Common European Framework of toward the upper part of the scale and

Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021) 315
 Kamal J I Badrasawi, Noor Lide Abu Kassim, Ainol Madziah Zubairi, Elia Md Johar and Siti Sakinah Sidik

Table 1
Study sample’s demographic characteristics

Variable Level Frequency (n) Percentage %

University Comprehensive University 110 36.7
Research University A 90 30.0
Research University B 100 33.3
Faculty Science 100 33.3
ICT 100 33.3
Engineering 100 33.3
Year Third 3 rd
140 46.7
Fourth 4th 160 53.3
Gender Male 116 38.7
Female 184 61.3

vice versa. Another important point is that the two misfit items (i.e. items 6 and
that Rasch analysis ensures if the items 10) with infit value > 1.5 were not included
contribute meaningfully to the construct in the final analyses as recommended in the
by investigating the item Fit statistics. The literature (Bond & Fox, 2015). Though the
inferential analyses (One-way analysis two items were recoded, they still showed
of variance (ANOVA) and Independent misfit values. Both items were misfits as
samples t-test) were also conducted to they shared the same characteristics in
compare mean scores of English-speaking that they were worded positively. The item
anxiety, Confidence and Perceived ability in reliability was high (0.95), with separation
English oral communication across selected index (3.04) > 2; and the reliability of a
demographic variables (i.e. university person’s ability is also high (0.90) with
type/category, faculty, academic year and person separation index (4.29) > 2. For the
gender). The results are displayed in Tables point-measure correlation coefficients, ten
and Figures. items had positive values, ranging from
0.74 to 0.83. It means that all items were
RESULTS working in the same direction to define
Psychometric Properties of the English the English language speaking anxiety
Speaking Anxiety Scale construct. The ten items’ infit and outfit
A 12-item scale was used to measure Mean-square statistics were within the
English language speaking anxiety of the recommended range (0.5–1.5), indicating
300 Science and Technology undergraduate that they contributed meaningfully to the
students from Malaysia’s three selected measured construct (i.e., English speaking
public universities. Table 2 shows the anxiety). Thus, the scale’s unidimensionality
psychometric properties for all items using was met, with the variance explained by
the Rasch Model. It is important to note the measures being 64%. The largest factor

316 Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021)
 A Rasch Analysis on Speaking Anxiety, Self-Confidence and Ability

Table 2
Reliability, separation, item fit statistics and point-Measure correlation coefficients

Infit Outfit PT-

No Item Measure
MNSQ ZSTD MNSQ ZSTD CORR
1 I feel very self-conscious about speaking English in 1.07 0.8 1.08 0.9 .77
front of others.
2 I am afraid that native speakers will laugh at me .96 -0.5 .94 -0.7 .80
when I speak English.
3 I lack self-confidence when I speak in English to .80 -2.6 .80 -2.6 .83
others.
4 I always feel that others speak English better than 1.14 1.6 1.09 1 .78
I do.
5 I doubt my ability to speak English properly. .83 -2.2 .84 -2 .82
6 I enjoy speaking English with native speakers. Misfit item (infit > 1.5) DELETED
7 My hands tremble when I am giving a speech. 1.33 3.7 1.38 4.1 .74
8 My heart beats very fast while waiting for my turn 1.09 1 1.11 1.2 .78
to start a speech.
9 While preparing for giving a speech, I forget facts I .89 -1.4 .86 -1.7 .81
really know due to tension and nervousness.
10 I feel relaxed while giving a speech. Misfit item (infit > 1.5) DELETED
11 When I make a mistake while giving a speech, I .92 -1 .94 -0.6 .79
find it hard to concentrate on the parts that follow.
12 While giving a speech, I experience a feeling of .87 -1.6 1.03 0.4 .77
helplessness building up inside me.
Means .99 -0.2 1.01 0.0
[Link] .16 1.9 .16 1.9
Reliability of item difficulty measures 0.95
Item separation 3.03
Reliability of person ability .90
Person separation 4.26
Variance explained by measures 60.4%
Unexplained variance in 1st contrast 2.3 (< 2 items) Linacre 2019.

extracted from the residuals was equivalent and distribution of items and persons on
to 2.3 units, which had a strength of about the same interval scale. Overall, students
two items (Linacre, 2019). easily endorsed the scale items as the
person ability mean (0.27) was higher than
English Speaking Anxiety and its the item difficulty mean (0.00), indicating
Factors that the participants had experienced
Table 3 shows the order of the item difficulty English language speaking anxiety. The
measures arranged from the highest to the least endorsed items were placed towards
lowest. Figure 1 displays the hierarchy the upper part of the scale, and highly

Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021) 317
 Kamal J I Badrasawi, Noor Lide Abu Kassim, Ainol Madziah Zubairi, Elia Md Johar and Siti Sakinah Sidik

Table 3
Item difficulty measures (English speaking anxiety and factors)

No Item Difficulty measures S.E.

12 While giving a speech, I experience a feeling of helplessness building 0.57 0.08
up inside me.
2 I am afraid that native speakers will laugh at me when I speak 0.29 0.08
English.
7 My hands tremble when I am giving a speech. 0.19 0.08
3 I lack self-confidence when I speak in English to others. 0.17 0.08
11 When I make a mistake while giving a speech, I find it hard to 0.17 0.08
concentrate on the parts that follow.
5 I doubt my ability to speak English properly. 0.14 0.08
9 While preparing for giving a speech, I forget facts I really know due -0.09 0.08
to tension and nervousness.
1 I feel very self-conscious about speaking English in front of others. -0.28 0.08
4 I always feel that others speak English better than I do. -0.56 0.08
8 My heart beats very fast while waiting for my turn to start a speech. -0.60 0.08
Means 0.00 0.08
[Link] 0.36 0.00

MEASURE PERSON - MAP - ITEM

endorsed items were placed towards the
<more>|<rare>
5 ## +

.# |
| lower part. It indicated that the participants
felt worried and most anxious before they
| Difficult to Endorse
|
|
4 +
# |
T|
|
spoke. They were thinking about others
whom the participants believed were better
### |
|
3 .##### +
|
.###### |
.### |
|
than themselves.
#### |
2 .### S+
## | Person Mean 0.27 logit
|
.### |
##### | Psychometric Properties of Self-
Confidence in Oral English
.##### |
1 ###### +
.###### |

Communication Scale
.####### |T
####### M| 12_feeling helpless while giving a speech
##### |S 2_Afraid native speakers laugh at me
####### | 7_ Hands tremble 11_ can't focus 3_Lack self-confidence 5_doubt my ability
0 .####### +M
###### | 9_ forget facts due to tension
.#### |S 1_Self-conscious about speaking in front of others
##### | 4_feel others speak better than me
A 19-item scale was used to measure the
confidence of 300 Science and Technology
## |T 8_ Heart beats waiting form my turn to speak
###### |
-1 ######### +
.# S|
.## |
.## |
|
undergraduates from three Malaysian public
universities when they communicated
# |
-2 .## +
# |
.## |
|
. T|
|
in an Individual Interview (IV1-IV7), in
Paired Discussion (PD8-PD12) and Group
-3 ## +
|
.# |
|

-4
|
|
+
Easy to Endorse Discussion (GD13-GD19). The results in
Table 4 showed that the item reliability
|
|
. |
|

-5 .###### +
|

<less>|<freq>
was very high (0.98), with separation
index (6.28) > 2; and the person ability
EACH "#" IS 2: EACH "." IS 1

Figure 1. Person-Item map (English Speaking reliability was also high (0.92) with person
Anxiety)

318 Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021)
 A Rasch Analysis on Speaking Anxiety, Self-Confidence and Ability

Table 4
Item fit statistics, point-measure correlation coefficients, reliability and separation (individual interview IV,
paired discussion PD and group discussion GD)
Infit Outfit PT-
No Item Measure
MNSQ ZSTD MNSQ ZSTD CORR
Individual Interview
IV1 Tasks that require me to respond (answer) 1.12 1.5 1.12 1.4 0.69
immediately make me nervous.
IV22 I find tasks that require my response (answer) 1.19 2.2 1.26 2.8 0.68
without preparation frightening.
IV3 I worry when I do not know what the interviewer 0.85 -1.8 0.84 -1.9 0.75
is going to ask me.
IV4 I perform well on tasks that give me time to 1.04 0.5 1.13 1.5 0.71
prepare.
IV5 Talking about familiar topics make me feel more 1.05 0.6 1.06 0.7 0.74
confident.
IV6 Interacting with an interviewer I am familiar with 0.99 -0.1 1.03 0.3 0.73
makes me feel comfortable.
IV7 My performance on a task depends on the 0.99 -0.1 1.03 0.4 0.70
interviewer that I get.
Paired Discussion (in Peers)
PD8 I find tasks that require my immediate response 1.10 1.3 1.12 1.3 0.68
(answer) frightening.
PD9 I worry when I do not know what my peer is 1.27 3.2 1.25 2.6 0.64
going to ask me.
PD10 I perform well on tasks that give me time to 0.88 -1.5 0.92 -0.9 0.74
prepare.
PD11 I perform well on tasks that require me to interact 0.88 -1.5 0.91 -1 0.73
with a peer.
PD12 My performance on a task depends on the peer 1.07 0.9 1.09 1 0.72
that I get.
Group Discussion
GD13 Tasks that give me time to prepare my response 0.72 -3.7 0.77 -2.9 0.78
(answer) make me feel comfortable and relaxed.
GD14 I like tasks where I am given time to prepare my 0.89 -1.4 0.9 -1.1 0.75
response (answer) while the other group members
take turns speaking.
GD15 I worry when I do not know what my group 1.09 1.1 1.25 2.6 0.66
members are going to ask me.
GD16 It does not matter to me whether I am given time Misfit item (infit > 1.5) DELETED
to prepare my response (answer).
GD17 I perform well when I interact with group 0.83 -2.1 0.85 -1.7 0.77
members with whom I am familiar.
GD18 Talking about familiar topics make me feel more 0.8 -2.5 0.8 -2.4 0.78
confident.
GD19 Interacting with group members I am familiar 0.91 -1.1 0.93 -0.8 0.75
with makes me feel more confident.

Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021) 319
 Kamal J I Badrasawi, Noor Lide Abu Kassim, Ainol Madziah Zubairi, Elia Md Johar and Siti Sakinah Sidik

Table 4 (continue)

Infit Outfit PT-

No Item Measure
MNSQ ZSTD MNSQ ZSTD CORR
Means 0.98 -0.3 1.01 0.1
[Link] 0.14 1.8 0.15 1.7
Reliability of Item difficulty measures 0.98
Item separation 6.28
Reliability of person ability 0.92
Person Separation 3.37
Variance explained by measures 52.7%
Unexplained variance in 1st contrast 4.07 ( No issue & Disattenuated correlations
are 1 or closer to 1)

separation index (3.37) > 2. As for the MEASURE PERSON - MAP - ITEM
<more>|<rare>
point-measure correlation coefficients, all 5 # +
|
|
items were found to have positive values, . |
|

ranging from 0.64 to 0.78. It means that all

|
4 . +
|

items were working in the same direction |

T|
.# |
to define the construct of confidence in 3
#
#
|
+
.# |
English oral communication. However, one ##
##
|
|
## |
item (i.e. GD 16) was deleted because it 2
.# S|
.## +

was a misfit (infit > 1.5). Other items’ infit

## |
.#### |
.###### |

and outfit mean-square values were within .########

.####
|
|
1 #### +T PDS2
the accepted range (0.5 to 1.5), indicating .######### M|
######## |
GDS3
PDS1
.### |S
that they contributed meaningfully to the #####
.####
|
|
IIS1
IIS3
IIS2 IIS7

0 .#### +M IIS4 PDS5

measured construct. In Table 4, the variance ######
##
|
|
GDS2
GDS5
IIS6
PDS3
PDS4

explained by the measures was 52%, and the . S|S GDS1

. | GDS6
GDS7

. |
largest factor extracted from the residuals -1 . +T IIS5
|
|
was equivalent to 4.07, which had a strength .#
.
|
|
. T|
of about four items. Therefore, it would not -2
.#
+
|

affect the scale as all other indicators have

|
. |
. |

been met. It is supported by the values of -3 .

|
+
. |
Disattenuated correlations, which were one |
|
|
or very close to 1 (Linacre, 2019). -4
|
+
. |
|
|

Self-Confidence in English Oral .

|
|

Communication
-5 .## +
<less>|<freq>
EACH "#" IS 3: EACH "." IS 1 TO 2

Overall, Figure 2 shows that the participants Figure 2. Person-Item Map (interview IV, paired
lack confidence in English oral interaction discussion PD and group discussion GD)

320 Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021)
 A Rasch Analysis on Speaking Anxiety, Self-Confidence and Ability

or communication in individual interviews, the items interacted on familiar topics or

paired or group discussions. They were familiar people, and the participants had
easy to endorse the items on self-confidence enough time to prepare in all contexts. On
(0.76 logits), which was negative. Table 5 the contrary, the items they stated were
indicated that students did not feel confident not very worried, not frightened, and not
when asked to speak about unfamiliar nervous or not concentrated when they
topics, speak to unfamiliar people, or not were asked to interact with other people,
have enough time to prepare regardless of talk on unseen topics and interact without
the context, either an individual interview, preparation received low endorsement.
paired or group discussions. It is indicated by Though the participants showed a lack
their high endorsement of the items placed of confidence in oral communication on
at the bottom part of the interval scale. All unfamiliar topics, interacting with unfamiliar

Table 5
Item difficulty measures (confidence in English oral communication: individual interview IV, paired discussion
PD and group discussion GD)
No Item Difficulty Measures S.E
PD 9 I worry when I do not know what my peer is going to ask me. 1.05 0.07
GD 15 I worry when I do not know what my group members are going to 0.83 0.07
ask me.
PD 8 I find tasks that require my immediate response (answer) 0.61 0.07
frightening.
IV 7 My performance on a task depends on the interviewer that I get. 0.40 0.08
IV2 I find tasks that require my response (answer) without preparation 0.37 0.08
frightening.
IV1 Tasks that require me to respond (answer) immediately make me 0.32 0.08
nervous.
IV3 I worry when I do not know what the interviewer is going to ask 0.16 0.08
me.
PD12 My performance on a task depends on the peer that I get. 0.07 0.08
IV4 I perform well on tasks that give me time to prepare. (Individual) 0.00 0.08
PD11 I perform well on tasks that require me to interact with a peer. -0.11 0.08
GD14 I like tasks where I am given time to prepare my response (answer) -0.19 0.08
while the other group members take turns speaking.
IV 6 Interacting with an interviewer I am familiar with makes me feel -0.20 0.08
comfortable.
PD10 I perform well on tasks that give me time to prepare. (Paired) -0.34 0.08
GD17 I perform well when I interact with group members with whom I -0.40 0.08
am familiar.
GD13 Tasks that give me time to prepare my response (answer) make me -0.46 0.08
feel comfortable and relaxed.
GD19 Interacting with group members I am familiar with makes me feel -0.50 0.08
more confident.
GD18 Talking about familiar topics make me feel more confident. -0.62 0.08
IV 5 Talking about familiar topics make me feel more confident. -0.98 0.09

Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021) 321
 Kamal J I Badrasawi, Noor Lide Abu Kassim, Ainol Madziah Zubairi, Elia Md Johar and Siti Sakinah Sidik

people or interacting without enough time Psychometric Properties of Perceived

or preparation, it is imperative to highlight Ability (Can-Do) Scale in English Oral
Communication
that their confidence differed based on the
context (i.e. individual task, paired or group A 10-item scale was used to measure the
discussions). Figure 3 shows the means and participants’ perceived ability (Can-Do
the hierarchy order of the items under each Statements) on their oral communication in
context. They mostly lacked confidence in the English language. The results in Table
interacting in pairs (Mean = 0.26), followed 5 indicated that the item reliability was high
by individual interviews (0.01), and in (0.91), with separation index (3.15) > 2; and
groups (-.022). Having investigated the the person ability reliability was also high
items under each category, it is noticeable (0.94) with person separation index (2.89)
that the participants were more confident in > 2. Moreover, all items had positive point-
interacting on familiar topics, with familiar measure correlation coefficients, ranging
people and having enough time to prepare from 0.77 to 0.88. It means that all items
(Figure 3). were working in the same direction to define

Figure 3. Person-Item Map (interview IV, paired discussion PD and group discussion GD)

322 Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021)
 A Rasch Analysis on Speaking Anxiety, Self-Confidence and Ability

the construct (i.e., perceived ability in oral explained by the measures was 66.8%,
communication in English). All items also and the largest factor extracted from the
had infit and outfit mean-square values residuals was equivalent to 2.00, which has
within the recommended range (0.5 - 1.5), a strength of about two items. Thus, it shows
indicating meaningfully to the measured that the scale’s unidimensionality was not
construct. Table 6 also shows the variance violated.

Table 6
Reliability, separation item fit statistics and point-measure correlation coefficients (Can-Do statements)

Infit Outfit PT-

No Item Measure
MNSQ ZSTD MNSQ ZSTD CORR
DO1 I can take part effortlessly in any conversation or
1.28 3.1 1.33 3.6 0.80
discussion.
DO2 I can have a good familiarity with idiomatic
1.40 4.4 1.44 4.7 0.77
expressions and colloquialisms.
DO3 I can express myself fluently and spontaneously
1.06 0.7 1.05 0.6 0.84
without much obvious searching for expressions.
DO4 If I do have a problem, I can restructure the
0.8 -2.5 0.82 -2.3 0.87
conversation without stopping any interaction.
DO5 I can use language flexibly and effectively for
0.88 -1.5 0.86 -1.7 0.87
social and professional purposes.
DO6 I can interact with a degree of fluency and
spontaneity that makes regular interaction with 0.75 -3.3 0.74 -3.4 0.87
native speakers possible.
DO7 I can enter unprepared into conversation on topics
that are familiar, of personal interest or pertinent 1.12 1.4 1.09 1.1 0.84
to everyday life.
DO8 I can present clear, detailed descriptions on a wide
1.06 0.7 1.04 0.5 0.85
range of subjects related to my field of interest.
DO9 I can explain a viewpoint on a topical issue giving
the advantages and disadvantages of various 0.81 -2.4 0.8 -2.5 0.88
options.
DO10 I can present a clear, smoothly-flowing description
or argument in style appropriate to the context 0.73 -3.6 0.73 -3.5 0.88
with an effective logical structure.
Means 0.99 -0.3 0.99 -0.3
[Link] 0.22 2.6 0.23 2.7
Item difficulty measure Reliability 0.91
Item separation 3.15
Person ability reliability 0.94
Person separation 2.89
Raw variance explained by measures 66.8%
Unexplained variance in 1st contrast 2.00

Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021) 323
 Kamal J I Badrasawi, Noor Lide Abu Kassim, Ainol Madziah Zubairi, Elia Md Johar and Siti Sakinah Sidik

Perceived Ability in English Oral MEASURE PERSON - MAP - ITEM

<more>|<rare>
Communication 7 .## +
|
## |
|

Overall, the participants were positive 6 +

|
|
. T|
(not high) toward their oral interaction in 5
##
+
|
|

English. The person ability means (0.31 4

####
##
##
|
+
|
## |
logits) could be deduced, higher than the 3
|
.# S+
.###### |

item difficulty mean (0.00 logits). However, 2

#####
.###
#######
|
|
+
.########### |
the results in Table 7 and Figure 4 show #########
#####
|
|
1 +

that they could not interact fluently or .#########

########
|T CDS2
###### M|
|S CDS3 CDS4 CDS6
0 ############# +M CDS1
spontaneously because they might have not .###
##
|S CDS10
| CDS7
CDS5 CDS8 CDS9

.#### |T

enough vocabulary and expressions that -1 .#

.#
+
|
|
## S|
helped them interact mainly in unfamiliar -2 .##
###
+
|
# |

situations. Therefore, it was difficult for -3

######

##
|
+
|
.## |
them to endorse the items at the upper part -4
###
.##
|
+
T|

of the scale. On the other hand, they felt they -5

## |
|
+
# |
could interact in familiar situations and on #
|
|
-6 ###### +

topics related to their social and professional <less>|<freq>

EACH "#" IS 2: EACH "." IS 1

settings. Figure 4. Person-Item Map (Can Do Statements) in

English oral communication skills
Table 7
Item difficulty measures (Can Do statements)

Difficulty
No Item S.E
Measures
DO 2 I can have a good familiarity with idiomatic expressions and colloquialisms. 0.70 0.1
DO4 If I do have a problem, I can restructure the conversation without stopping 0.32 0.1
any interaction.
DO 3 I can express myself fluently and spontaneously without much obvious 0.29 0.1
searching for expressions.
DO6 I can interact with a degree of fluency and spontaneity that makes regular 0.19 0.1
interaction with native speakers possible.
DO1 I can take part effortlessly in any conversation or discussion. -0.03 0.1
DO10 I can present a clear, smoothly-flowing description or argument in style -0.13 0.1
appropriate to the context with an effective logical structure.
DO5 I can use language flexibly and effectively for social and professional -0.15 0.1
purposes.
DO8 I can present clear, detailed descriptions on a wide range of subjects related -0.3 0.1
to my field of interest.
DO9 I can explain a viewpoint on a topical issue giving the advantages and -0.37 0.1
disadvantages of various options.
DO7 I can enter unprepared into conversation on topics that are familiar, of personal -0.52 0.1
interest or pertinent to everyday life.

324 Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021)
 A Rasch Analysis on Speaking Anxiety, Self-Confidence and Ability

RESULTS OF INFERENTIAL university (A) had the highest level of

ANALYSIS speaking anxiety (M= .8638 logits), which
The inferential analyses (One-way analysis significantly differed from the other two
of variance (ANOVA) and independent universities, comprehensive M= .0441 and
samples t-test were conducted to identify research university (B) M= -.0116 logits,
the significant differences in mean scores of p<0.5. Furthermore, the ANOVA results
English-speaking anxiety, confidence, and show a significant difference in students’
perceived ability in English language oral confidence in speaking English mean scores
communication with selected demographic across faculty (Science, Information and
variables (i.e., gender, academic year, Technology, and Engineering), p<.05. Post-
university type/category and faculty. The hoc analysis using Tukey test indicated that
results in Table 8 show no statistically the students in Engineering and Information
significant differences in the mean scores and Technology Faculties had higher
of speaking anxiety, confidence, and significant difference from those in Science
perceived ability in English language oral Faculty (1.04 logits, 1.03 logits and .208
communication for both students’ academic logits respectively), p< .05. Finally, the
year and gender, p>.05. In contrast, it same results were found for the perceived
shows a significant difference in speaking ability in English oral communication. The
anxiety mean scores for university type/ students in Engineering and Information
category (comprehensive and other two and Technology Faculties had a higher
research universities), (p<.05). Post-hoc significant difference from those in Science
analysis using the Tukey test indicated Faculty (1.013 logits, .688 logits and -.760
that the students who came from research logits), p < .05.

Table 8
Results of One-way analysis of variance (ANOVA) and Independent samples t-test

Construct Variable N Mean±SD (Logit) p value

Gender
Anxiety Male 116 -.0381±2.383 .055
Female 184 .467±1.899
Academic Year
Third 140 .451±1.560 .156
Fourth 160 .114±2.489
University Category
Comprehensive 110 .044±2.033 .006*
Research A 90 .864±1.787
Research B 100 -.012±2.362
Faculty
Science 100 .411±1.224
Information & Technology 100 .503±1.504 .093
Engineering 100 -.100±3.080

Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021) 325
 Kamal J I Badrasawi, Noor Lide Abu Kassim, Ainol Madziah Zubairi, Elia Md Johar and Siti Sakinah Sidik

Table 8 (continue)

Construct Variable N Mean±SD (Logit) p value

Gender
Confidence Male 116 .537±1.743 .081
Female 184 .896±1.768
Academic Year
Third 140 .675±1.618 .460
Fourth 160 .826±1.885
University Category
Comprehensive 110 .489±2.006 .128
Research A 90 .962±1.584
Research B 100 .862±1.607
Faculty
Science 100 .208±2.753
Information & Technology 100 1.03±.857 .001*
Engineering 100 1.04±.788
Gender
Can-Do Male 116 .313±2.652 1.000
Perceived ability Female 184 .314±3.186
Academic Year
Third 140 .144±2.859 0359
Fourth 160 .462±3.094
University Category
Comprehensive 110 .527±3.037 .232
Research A 90 .516±3.162
Research B 100 -.104±2.743
Faculty
Science 100 -.760±4.269
Information & Technology 100 .688±1.828 .000*
Engineering 100 1.013±1.887
* The mean difference is significant at the 0.05 level

DISCUSSION native speakers (Bux et al., 2015). As a

This study suggests that, in general, the result, the participants in this study mainly
participants had experienced English felt worried and anxious before they were
language speaking anxiety that could asked to speak. They often lost their control
affect their English oral communication before they delivered the speech in front of
that concurs with Salem and Al Dyiar others. They prefer to talk about familiar
(2014), who found a negative relationship topics and interact with familiar persons,
between English speaking anxiety and and they need enough time to prepare
speaking fluency. In addition, speaking for the communication. These findings
anxiety adversely affects students’ oral show that the participants demonstrate
presentation and conversation with English situational anxiety since it appears in

326 Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021)
 A Rasch Analysis on Speaking Anxiety, Self-Confidence and Ability

specific situations (MacIntyre, 1999), for settings. It is because they had enough
example, in unfamiliar contexts where the vocabulary and expressions with which to
learners are required to use the language. interact. Sadighi and Dastpak (2017) found
This anxiety ends or reduces when learners inadequate vocabulary knowledge as one of
become more familiar with the new context the main sources of speaking anxiety among
(Riasati, 2011). The participants also tend to students. Ahmed et al. (2017) found that
think about others whom they believed to be the participants with insufficient linguistic
better than them. It might be because most competence led to speaking anxiety and
participants showed a lack of confidence affected their oral communication since
in English oral communication during they could not express themselves in a
individual interviews, paired or group wide range of communicative situations.
discussions (i.e. regardless of the context). Zahiri et al. (2017) found that speaking
Kamaruddin et al. (2019) found that low anxiety and perceived ability affect students’
self-esteem and social anxiety contributed monologue speaking skills. The former was
significantly to the level of both listening found to have a negative effect on speaking
and speaking anxieties among Malaysian performance, while a higher perception of
university students majoring in non-English their ability has made them better in speaking
programs. On the same note, Desmaliza performance. Besides, lack of vocabulary
and Septiani (2017) found a significant and expressions result in the participants
correlation between students’ perceived to have negative or low perception on
ability and speaking skills, positively their self-confidence. Stenstrom (2014)
influencing students’ performance in oral maintained that confidence is an important
communication activities. In this study, the factor in spoken interaction as a two-way
participants perceived that they could not process. This idea was coined in Tsou (2005)
interact fluently or spontaneously because who found that high self-confidence is
they might not have enough vocabulary and positively correlated with oral performance
expressions to help them interact mainly in as it determines the learners’ willingness to
unfamiliar situations. communicate. Other researchers found the
The finding agrees with Amiri and Puteh exact relationships between self-confidence
(2018)’s study, which found that among the and speaking competency (Gürler, 2015;
factors that caused anxiety to international Mandokhail, 2018; Tridinanti, 2018).
postgraduate students in different Malaysian Though the surveyed participants showed
universities are inadequate linguistic a lack of confidence in oral communication
competency and inadequate knowledge of on unfamiliar topics, interacting with
the presentation’s content. On the other unfamiliar people or interacting without
hand, they perceived a higher ability to enough time for prior preparation, they
interact about familiar situations and topics lacked confidence in interacting in pair,
related to their social and professional individual and group interviews. They

Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021) 327
 Kamal J I Badrasawi, Noor Lide Abu Kassim, Ainol Madziah Zubairi, Elia Md Johar and Siti Sakinah Sidik

might feel more comfortable in groups as Pakistan. Other studies reported that females
they took additional time to prepare, and the scored higher levels of EFL speaking anxiety
same applied when they were in individual (Ahmed & Alansari, 2004), whereas Elaldi
interviewing. McDonough (2004) reported (2016) reported that male EFL University
that learners improved their speaking skills students had higher speaking anxiety
when put in group and pair tasks. These levels than female students. It seems these
findings are alarming as university students variations depend on research contexts.
are expected to be able to communicate For confidence and perceived ability,
at the C1 level of the CEFR, where the one-way ANOVA analysis indicated
communication expectations are about the that engineering students were more
ability to smoothly engage in synchronous confident and perceived higher ability in
discourse involving a wide range of social, oral communication than their counterparts
academic and professional topics without in science or information and technology
much searching and assistance (Council of faculties despite the type of university. This
Europe, 2018). result indicated that undergraduate engineers
The independent samples t-test shows might have realised the importance of
no significant differences in the mean scores English speaking skills in their future
of speaking anxiety, confidence in speaking careers. Past studies found that many
English and perceived ability for gender and engineering graduates could not secure a
student academic year. However, the female job due to their inability to command good
students reported a higher mean score in English (Kakepoto, 2013; Sheth, 2016;
speaking anxiety, as indicated by the mean Ting et al., 2017). Over the years, research
differences. The female students might be has focused on the significance of English
more concerned about their appearance in for engineers at the workplace (Božić &
front of others in the conversation, which Pintarić, 2018; Dewi et al., 2015; Hossain,
might affect their confidence and ability in 2013; Rajprasit & Hemchua, 2015; Spence
oral communication. Literature has reported & Liu, 2013). Sheth (2016) proclaims
different findings in terms of gender and that engineering employers give priority
ESL or EFL speaking anxiety. For instance, to graduate engineers with competence in
Batiha et al. (2016) found no significant English speaking over their counterparts who
differences in the mean scores of speaking tend to be highly tech-savvy but with a low
anxiety in research conducted on the factors level of English language speaking skills.
of speaking anxiety among EFL university As for Science undergraduate students, there
learners due to gender. The same findings is a need to improve awareness about the
were reported by Ahmed et al. (2017), who importance attached to the English language
conducted research to identify the factors in their future profession. It would motivate
responsible for ESL oral communication them towards enhancing their competence
anxiety among postgraduate students in in spoken English.

328 Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021)
 A Rasch Analysis on Speaking Anxiety, Self-Confidence and Ability

CONCLUSIONS AND varied in familiarity and content to prepare

IMPLICATIONS them for future careers. In other words,
This study mainly aimed to analyse speaking they should be provided with authentic
anxiety, confidence and perceived ability situations to practice and improve their
in English oral communication among levels in English oral communication.
Science and Technology undergraduates in Besides, students should be encouraged to
comprehensive and research universities do oral presentations individually, in pairs
in Malaysia. Furthermore, it aimed to and groups, reflecting the real discourse they
find the significant differences in the are expected to engage in as they enter the
three sub-constructs based on the selected working world. Furthermore, the university
demographic variables. Overall, the should provide students with training
participants experienced speaking anxiety, modules on enhancing their confidence and
low confidence, and yet high-perceived perceived ability as these factors play a
ability in English oral communication, with substantial role in students’ level of English
significant differences in mean scores of oral communication. Contextually, the
English speaking anxiety across university programs could also be geared towards more
category and confidence and perceived social-like situations and the classroom
ability due to students’ specialisation setting so that undergraduates can make
(i.e., science, technology and information English a social practice. As a result,
and engineering). However, engineering
students will be more qualified for future
students were more confident and could
employability.
perform better in oral communication than
their science or information and technology ACKNOWLEDGEMENTS
faculties counterparts. In addition, the
This research paper is one of the outputs
findings show that the participants felt
of an FRGS Grant project (FRGS-16-051-
more confident and more able in familiar
0550.) sponsored by the Malaysian Ministry
situations, communicating on familiar
of Education. Therefore, we would like
topics to familiar audiences, contrary to the
expectations for the C1 level in the CEFR. to extend our appreciation and gratitude
The students enrolled in science, to the Malaysian Ministry of Education
engineering and technology programmes for the financial support and the Research
need more training on English oral Management Centre, International Islamic
communication. Lecturers need to encourage University, for their cooperation and support
students to practice oral communication in to complete the project. Special thanks
English in front of the class to improve are also extended to all participants in the
their self-confidence and mitigate their project from the selected public universities
anxiety. Furthermore, the topics should be in Malaysia.

Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021) 329
 Kamal J I Badrasawi, Noor Lide Abu Kassim, Ainol Madziah Zubairi, Elia Md Johar and Siti Sakinah Sidik

REFERENCES Badrasawi, K. J., Solihu, A., & Ahmad, T. B. (2020).

Second language speaking anxiety among
Ahmed, M., & Alansari, B. (2004). Gender differences
Malaysian postgraduate students at a Faculty of
in anxiety among undergraduates from ten
Education. International Journal of Education
Arab countries. International Journal of Social
and Literacy Studies, 8(2), 54-61. [Link]
Behavior and Personality, 32(7), 649-655.
org/10.7575/[Link].v.8n.2p.54
[Link]
Batiha, J., Noor, N., & Mustaffa, R. (2016).
Ahmed, N., Pathan, Z. H., & Khan, F. S. (2017).
Speaking anxiety among English as a foreign
Exploring the causes of English language
language learner in Jordan: Quantitative
speaking anxiety among postgraduate students of
research. International Journal of Education
University of Baluchistan, Pakistan. International
and Research, 4(10), 63-82.
Journal of English Linguistics, 7(2), 99-105.
[Link] Bond, T., & Fox, C. M. (2015). Applying the Rasch
Model: Fundamental measurement in the human
Akkakoson, S. (2016). Speaking anxiety in
sciences. Routledge.
English conversation classrooms among Thai
students. Malaysian Journal of Learning and Božić, D. L., & Pintarić, L. (2018). Comparison
Instruction, 13, 63-82. [Link] of employers’ and students’ perceptions
mjli2016.13.1.4 regarding communication skills. The Journal
of Teaching English for Specific and Academic
Alawiyah, T. (2018). Speaking self-efficacy and EFL
Purposes, 6(1), 63-82. [Link]
student teachers’ speaking achievement. Jurnal
jtesap1801063b
Pendidikan dan Pengajaran, 5(1), 87-96. https://
[Link]/10.19109/ejpp.v5i1.2052 Brown, H. D. (2004). Teaching by principles: An
interactive approach to language pedagogy.
Alderson, J. C. (2017). The Common European
Harper and Row Publisher.
Framework of Reference for Languages (CEFR)
for English language assessment in China Bux, I. G., Ali, F. B., Umrani, T., & Ali, S. L.
(Foreword to special issue). Language Testing (2015). English speaking anxiety: A study
in Asia, 7, Article 20. [Link] of undergraduates at Lasbela University,
s40468-017-0049-9 Baluchistan. International Researchers, 4(1),
54-61.
Ali, M. A. K. (2017). English language anxiety:
Development and validation of a brief measure. Chin, C. (2016, February 28). Nation builders wanted.
International Journal of Psychology and The Star Online. [Link]
Educational Studies, 4(2), 42-53. [Link] news/nation/2016/02/28/nation-builders-wanted-
org/10.17220/ijpes.2017.02.005 malaysias-dream-of-becoming-a-highincome-
nation-by-2020-could-crumble-if-we-d/
Al-Sibai, D. (2004). Promoting oral fluency of second
language learners: Educational linguistics. King Clement, A., & Murugavel, T. (2015). English for
Saud University. employability: A case study of the English
language training need analysis for engineering
Amiri, F., & Puteh, M. (2018). Oral communication
students in India. English language teaching,
apprehension among international doctoral
8(2), 116-125. [Link]
students. English Language Teaching, 11(2),
v8n2p116
164-171. [Link]

330 Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021)
 A Rasch Analysis on Speaking Anxiety, Self-Confidence and Ability

Council of Europe (2018). Common European Gürler, İ. (2015). Correlation between self-confidence
Framework of Reference for languages: and speaking skill of English language
Learning, teaching, assessment - Companion teaching and English language and literature
volume. Council of Europe Publishing. preparatory students. Current Research in Social
Sciences, 1(2), 14-19.
Crystal, D. (2004). The Cambridge encyclopedia
of the English language. Ernst Klett Sprachen Hart, L. E. (1989). Classroom processes, sex of student,
Verlag. and confidence in learning mathematics. Journal
for Research in Mathematics Education, 20(3),
David, A., Thang, S., & Azman, H. (2015).
242-260. [Link]
Accommodating low proficiency ESL students’
language learning needs through an online Horwitz, E. K., Horwitz, M. B., & Cope, J. A. (1986).
writing support system. Journal of Social Foreign language classroom anxiety. The Modern
Sciences and Humanities, Special Issue1(2015), Language Journal, 70, 125-132. [Link]
118-271. org/10.1111/j.1540-4781.1986.tb05256.x

Desmaliza, D., & Septiani, T. (2017). Students’ Hossain, J. (2013). ESP needs analysis for engineering
self- efficacy and their speaking skill at lower students: A learner-centered approach. Journal of
secondary school. In International Conference PU, Part B, 2(2), 16-26.
on Education in Muslim Society (ICEMS
Hudson, P., & Hudson, S. (2003). An evaluation of
2017) (pp. 122-127). Atlantis Press. [Link]
an English language immersion programme for
org/10.2991/icems-17.2018.24
ESL pre-service teachers. GEMA Online Journal
Dewi, Z. Z., Azlin, M. Z., Bakar, S. B. A., & of Language Studies, 3(1), 42-65.
Sumaiyyah, O. (2015). English language
Ismail, N. A. (2011). Graduates’ characteristics and
communication needs at workplace as perceived
unemployment: A study among Malaysian
by students. Journal of Science & Engineering
graduates. International Journal of Business and
Technology, 3(2), 27-31.
Social Science, 2(16), 94-102.
Dordinejad, F. G., & Ahmadabad, R. M. (2014).
Kakepoto, I. (2013). Oral presentation and
Examination of the relationship between
communication skills in workplace environments
foreign language classroom anxiety and English
among engineers and engineering students in
achievement among male and female Iranian
Pakistan [Doctoral thesis, Universiti Teknologi
high school students. International Journal of
Malaysia]. Universiti Teknologi Malaysia
Language Learning and Applied Linguistics
Institutional Repository. [Link]
World, 6(4), 446-460.
id/eprint/44038/
Elaldi, S. (2016). Foreign language anxiety of students
Kalanzadeh, G., Mahnegar, F., Hassannejad, E.,
studying English Language and Literature: A
& Bakhtiarvand, M. (2013). The influence of
Sample from Turkey. Journal of Educational
EFL students’ self-esteem on their speaking
Research and Reviews, 11(6), 219-228. https://
skills. The International Journal of Language
[Link]/10.5897/err2015.2507
Learning and Applied Linguistics World, 2(2),
Free Malaysia Today. (2017, November 13). 77-84. [Link]
Economist: Poor English, lack of experience
Kamaruddin, S. F., Daneil, I. L., Rijeng, J. S., Bohari,
costing grads jobs. Free Malaysia Today. https://
M. M., & Eng, T. H. (2019). The impact of
[Link]/2XFBWXk
listening and speaking anxieties on the Fourth

Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021) 331
 Kamal J I Badrasawi, Noor Lide Abu Kassim, Ainol Madziah Zubairi, Elia Md Johar and Siti Sakinah Sidik

Industrial Revolution: What can educators their oral proficiency. International Journal of
do? Language & Communication, 6(1), 409-423. English Linguistics, 8(3), 210-222. [Link]
org/10.5539/ijel.v8n3p210
Kassim, N. L. A., & Zubairi, A. M. (2003, July 22-23).
Interaction between test-taker characteristics, M c C r o s k e y, J . ( 1 9 7 0 ) . M e a s u r e s o f
task facets and L2 oral proficiency test communication‐bound anxiety. Speech
performance [Paper presentation]. Language Monographs, 37(4), 269-277. [Link]
Testing Research Colloquium, Reading, United org/10.1080/03637757009375677
Kingdom.
McDonough, K. (2004). Learner-learner interaction
Koriat, A., Lichtenstein, S., & Fischhoff, B. (1980). during pair and small group activities in a Thai
Reasons for confidence. Journal of Experimental EFL context. System, 32(2), 207-224. [Link]
Psychology: Human Learning and Memory, org/10.1016/[Link].2004.01.003
6, 107-118. [Link]
Ministry of Education. (2012). Malaysia education
7393.6.2.107
blueprint 2013-2025. Kementerian Pendidikan
Kumar, U. H. (2018). English language anxiety Malaysia.
and academic achievement. North Asian
Mulyono, H., Sari, R. N., & Ningsih, S. K. (2019).
International Research Journal of Social Science
Factors contributing to EFL students’ speaking
& Humanities, 4(3), 138-147.
anxiety. Register Journal, 12(1), 13-27. https://
Linacre, J. M. (2018). A user’s guide to WINSTEPS [Link]/10.18326/rgt.v12i1.13-27
MINISTEP Rasch-model computer programs
Musa, N. C., Koo, Y. L., & Azman, H. (2012).
(Program Manual No. 4.1.0). [Link].
Exploring English language learning and
com
teaching in Malaysia. GEMA: Online Journal
Linacre, J.M. (2019). A user’s guide to Winsteps & of Language Studies, 12(1), 35-51.
Ministeps (Rasch-Model Computer Programs).
Nair, G. K. S., Rahim, R. A., Setia, R., Husin, N.,
(Program Manual 4.4.5). [Link]
Sabapathy, E., Jalil, N. A. A., & Yusoff, N. I. M.
Linnenbrink, E. A., & Pintrich, P. R. (2003). The role (2012). Malaysian graduates English adequacy
of self-efficacy beliefs in student engagement in the job sector. Asian Social Science, 8(4),
and learning in the classroom. Reading 143-147. [Link]
&Writing Quarterly, 19(2), 119-137. https://
Nishanti, R. (2018). The importance of learning English
[Link]/10.1080/10573560308223
in today world. International Journal of Trend
MacIntyre, P. D. (1999). Language anxiety: A review in Scientific Research and Development, 3(1),
of literature for language teachers. In D. J. Young 871-874. [Link]
(Ed.), Effect in Foreign Language and Second
Pajares, F. (1996). Self-efficacy beliefs in
language learning (pp.24-43). McGraw Hill
academic settings. Review of Educational
Companies.
Research, 66(4), 543-578. [Link]
Malaysian Employers Federation. (2016). English org/10.3102/00346543066004543
proficiency critical to enhance employability
Pandarangga, S. (2015). The transformation of
of graduates. [Link]
English as a global language in the world.
Attachments/[Link]
LiNGUA, 10(2), 90-96. [Link]
Mandokhail, S., Khan, F. R., & Malghani, M. ling.v10i2.3132
(2018). Impact of ESL learners’ self-esteem on

332 Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021)
 A Rasch Analysis on Speaking Anxiety, Self-Confidence and Ability

Pappamihiel, N. E. (2002). English as a second Salem, M. S., & Al Dyiar, M. A. (2014). The
language students and English language anxiety: relationship between speaking anxiety and oral
Issues in the mainstream classroom. Research in fluency of special education Arab learners of
the Teaching of English, 36(3), 327-355. English. Asian Social Science, 10(12), 170-176.
[Link]
Pennycook, A. (2014). The cultural politics of English
as an international language. Routledge. https:// Scrivener, J. (2005). Learning teaching: A guidebook
[Link]/10.4324/9781315843605 for English teachers. Macmillan Education.

Rajprasit, K., & Hemchua, S. (2015). The English Shakir, R. (2009). Soft skills at the Malaysian
language & communication in the international institutes of higher learning. Asia Pacific
workplace: An examination of Thai computer Education Review, 10(3), 309-315. [Link]
engineering professionals. The Southeast Asian org/10.1007/s12564-009-9038-8
Journal of English Language Studies, 21(3),
Sheth, T. D. (2016). Communication skill: A
109-124.
prerequisite for engineers. International Journal
Ramamuruthy, V. (2019). The effects of task-based on Studies in English Language and Literature
approach on speaking anxiety among ESL low (IJSELL), 3(7), 51-54.
proficiency diploma students. Universal Journal
S h r a u g e r, J . S . , & S c h o h n , M . ( 1 9 9 5 ) .
of Educational Research, 7(6), 1363-1375.
Self-confidence in college students:
[Link]
Conceptualization, measurement, and behavioral
Reddy, R. (2016). Development of communication implications. Assessment, 2(3), 255-278. https://
skills in English for engineering [Link]/10.1177/1073191195002003006
graduates. IOSR Journal of Humanities and
Spence, P., & Liu, G. (2013). Engineering English and
Social Science (IOSR-JHSS), 21(5), 102-104.
the high-tech industry: A case study of an English
Riasati, M. J. (2011). Language learning anxiety from needs analysis of process integration engineers
EFL learners’ perspective. Middle-East Journal at a semiconductor manufacturing company in
of Scientific Research, 7(6), 907-914. Taiwan. English for Specific Purposes, 32, 97-
109. [Link]
Roshid, M. M., & Chowdhury, R. (2013). English
language proficiency and employment: A case Spielberger, C. D. (1983). Manual for the state-trait
study of Bangladeshi graduates in Australian anxiety (From Y). Consulting Psychologists
employment market. Online Submission, 3(1), Press.
68-81. [Link]
Stenstrom, A. B. (2014). An introduction to
Rusli, R., Yunus, M. M., & Hashim, H. (2018, April spoken interaction. Routledge. [Link]
23-24). Low speaking proficiency among the org/10.4324/9781315845449
Malaysian undergraduates: Why and how? In
Sunyi, D. P. (2017). Investigating students’ self-
e-Persidangan Antarabangsa Sains Sosial dan
efficacy of public speaking. International
Kemanusiaan (pp. 678-689). Kolej Universiti
Journal of Education and Research, 5(10),
Islam Antarabangsa Selangor.
97-108.
Sadighi, F., & Dastpak, M. (2017). The sources of
Tanveer, M. (2007). Investigation of the factors that
foreign language speaking anxiety of Iranian
cause language anxiety for ESL/EFL learners
English language learners. International Journal
in learning speaking skills and the influence it
of Education and Literacy Studies, 5(4), 111-115.
casts on communication in the target language
[Link]

Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021) 333
 Kamal J I Badrasawi, Noor Lide Abu Kassim, Ainol Madziah Zubairi, Elia Md Johar and Siti Sakinah Sidik

[Master’s thesis, University of Glasgow]. Ur, P. (1996). A course in language teaching: Practice
ResearchGate. [Link] and theory. Cambridge University Press.
publication/293263722
Wijewardene, L., Yong, D., & Chinna, K. (2014).
Ting, S., Marzuki E., Chuah K., Misieng, J., & Jerome, English for Employability-the need of the hour
C. (2017). Employers’ views on the importance for Sri Lankan graduates. British Journal or Arts
of English proficiency and communication and Social Sciences, 17(1), 137-145.
skill for employability in Malaysia. Indonesian
Xue, J., & Zuo, W. (2013). English dominance and
Journal of Applied Linguistics, 7(2), 315-327.
its influence on international communication.
[Link]
Theory and Practice in Language Studies,
Tridinanti, G. (2018). The correlation between 3(12), 2262-2266. [Link]
speaking anxiety, self-confidence, and speaking tpls.3.12.2262-2266
achievement of undergraduate EFL students of
Yim, S. Y., & Yu, Y. L. (2011). Validating the English
private university in Palembang. International
learning anxiety scale for primary school students
Journal of Education and Literacy Studies, 6(4),
in Korea. English Teaching, 66(2), 101-123.
35-39. [Link]
[Link]
ijels.v.6n.4p.35
Zahiri, A., Sibarani, B., & Sumarsih (2017). To what
The Sun Daily. (2018, January 22). Language main
extent do anxiety and self-efficacy effect the
barrier to grads getting jobs (Updated). The Sun
EFL students’ English monologue speaking
Daily. [Link]
skill? International Journal of Education and
language-main-barrier-grads-getting-jobs-
Research, 5(9), 151-160.
updated-BUARCH520890
Zhang, R., & Zhong, J. (2012). The hindrance of
Tsou, W. (2005). The effects of cultural instruction
doubt: Causes of language anxiety. International
on foreign language learning. RELC
Journal of English Linguistics, 2(3), 27-33.
Journal, 36(1), 39-57. [Link]
[Link]
org/10.1177/0033688205053481

UNESCO. (2012). Graduate employability in Asia.

UNESCO Bangkok.

334 Pertanika J. Soc. Sci. & Hum. 29 (S3): 309 - 334 (2021)