Prevalence of PIU and Associated Factors in Malaysian Universities

This report, based on a 2015 cross-sectional study of 1023 undergraduate students at a Malaysian public university, investigates the prevalence and determinants of Pathological Internet Use (PIU). The study, utilizing the Young’s Diagnostic Questionnaire, found a PIU prevalence of 28.9%. Significant factors associated with PIU included recreational internet use of three or more hours, pornography use, gambling problems, drug use in the past year, and moderate/severe depression. The report highlights the need for university authorities to develop interventions, such as screening, awareness campaigns, and promoting healthy lifestyles, to address the adverse outcomes of PIU. The study emphasizes the importance of understanding the impact of internet usage patterns and psychosocial factors on student well-being. The research also provides a literature review covering existing research on PIU and its associated factors, including socio-demographic variables, internet use patterns, psychosocial factors, and comorbid symptoms.

Wen Ting Tong, Md. Ashraful Islam, Wah Yun Low, Wan Yuen Choo, and Adina Abdullah
63
Prevalence and Determinants of Pathological Internet Use among
Undergraduate Students in a Public University in Malaysia
Wen Ting Tong1, Md. Ashraful Islam2, Wah Yun Low3, Wan Yuen Choo4,
and Adina Abdullah5
Pathological Internet Use (PIU) affects one’s physical and mental health, and university
students are at risk as they are more likely to develop PIU. This study determines the
prevalence of PIU and its associated factors among students in a public university in
Malaysia. This cross-sectional study was conducted among 1023 undergraduate students in
2015. The questionnaire comprised of items from the Young’s Diagnostic Questionnaire to
assess PIU and items related to socio-demography, psychosocial, lifestyle and co-morbidities.
Anonymous paper-based data collection method was adopted. Mean age of the respondents
was 20.73 ± 1.49 years old. The prevalence of pathological Internet user was 28.9% mostly
Chinese (31%), 22 years old and above (31.0%), in Year 1 (31.5%), and those who perceived
themselves to be from family from higher socio-economic status (32.5%). The factors found
statistically significant (p<0.05) with PIU were Internet use for three or more hours for
recreational purpose (OR: 3.89; 95% CI:1.33 – 11.36), past week of Internet use for
pornography purpose (OR: 2.52; 95% CI:1.07 – 5.93), having gambling problem (OR: 3.65;
95% CI:1.64 – 8.12), involvement in drug use in the past 12 months (OR: 6.81; 95% CI:1.42
– 32.77) and having moderate/severe depression (OR: 4.32; 95% CI:1.83 – 10.22). University
authorities need to be aware of the prevalence so that interventions can be developed to
prevent adverse outcomes. Interventions should focus on screening students for PIU, creating
awareness on the negative effects of PIU and promoting healthy and active lifestyle and
restricting students’ access to harmful websites.
Keywords: internet addiction, prevalence, risk factors, tertiary students, Malaysia
In this digital world, the growing Internet use has led to problematic behavior such as
excessive use and several terms have been coined to describe such behavior such as Internet
addiction (IA), Internet dependence, problematic Internet use, compulsive Internet use,
pathological Internet use (PIU), excessive Internet use (Rial Boubeta et al. 2015). PIU is when
a person has excessive or poorly-controlled preoccupations, urges or behaviors related to
Internet use resulting in impairment and distress to their life (Shaw & Black, 2008).
In the 5th edition of the Diagnostic and Statistical Manual of Mental Disorder (DSM-
5), the American Psychiatric Association (APA, 2012) has included Internet Use Disorder as
their clinical diagnosis. In this paper, PIU is used to define someone with Internet problem
with a potentially pathological behavioral problem and does not refer to a clinical diagnosis,
since the instrument used in this study to assess Internet problem is based on a screening tool.
Also PIU is a preferred term as compared to IA where the latter refers to dependency on
psychoactive substances (Davis, 2001).
1 Research Assistant, Department of Primary Care Medicine, Faculty of Medicine, University of Malaya, Kuala
Lumpur, Malaysia
2 Research Fellow, Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia
3 Professor, Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia. Email: [e-mail]
4 Associate Professor, Department of Social and Preventive Medicine, Faculty of Medicine, University of Malaya,
Kuala Lumpur, Malaysia
5 Lecturer, Department of Primary Care Medicine, Faculty of Medicine, University of Malaya, Kuala Lumpur,
Malaysia
The Journal of Behavioral Science Copyright © Behavioral Science Research Institute
2019, Vol. 14, Issue 1, 63-83 ISSN: 1906-4675 (Print), 2651-2246 (Online)

Pathological Internet Use among Undergraduate Students
64
Currently, most studies on PIU have been focused largely in Europe and US where
PIU has been a prominent issue in the adolescent health literature. Internet addiction has
become more prevalent in Asia than in other parts of the world (Yen, Yen, & Ko, 2010). In a
meta-analysis of 31 nations across seven world regions, the global prevalence of Internet
addiction was 6.0% (Cheng & Li, 2014). In East Asian countries, most studies were from
Taiwan, China, Korea and Singapore, however, literature are scarce in other Asian
counterparts (Kuss, Griffiths, & Binder, 2013; Lam, 2014). In Asia, there is higher variation
in prevalence among young people and adolescents, ranging from 8% to 50.9% (Kim et al.,
2006; Mak et al., 2014). In China, the rates ranged from 6% to 26.5% (Cao et al., 2011, Lai et
al., 2013, Wu et al., 2013, Chi, Lin & Zhang, 2016; Xin et al., 2018). In one study among
adolescents in six Asian countries, namely China, Hong Kong, Japan, South Korea, Malaysia
and the Philippines, there were variations in internet behaviors and addiction across these
countries (Mak et al., 2014). Further, the study found that the prevalence of addictive Internet
use ranges from 1% in South Korea to 5% in the Philippines, and the prevalence of
problematic Internet use ranges from 13% in South Korea to 46% in the Philippines, as
measured by the Internet Addiction Test (IAT). Further, based on the Revised Chen Internet
Addiction Scale (CTAS-R), the prevalence of addictive Internet use in the six countries are as
follows: Philippines (21%), Hong Kong (16%), Malaysia (14%), South Korea (10%), China
(10%) and Japan (6%) (Mak et al., 2014). Elsewhere, a cross-sectional study in five ASEAN
countries (Indonesia, Malaysia, Myanmar, Thailand and Vietnam), the overall prevalence of
pathological Internet use was 35.9% (ranging from 16.1% in Myanmar to 52.4% in Thailand),
maladaptive use 34.8% and adjusted Internet users 29.9% (Turnbull et al., 2018). Among
these five ASEAN countries, the highest prevalence of pathological Internet use is Thailand
(52.4%) followed by Indonesia (38.5%), Vietnam (37.5%), Malaysia (28.9%), and Myanmar
(16.1%) (Turnbull et al., 2018). Other parts of Asia, in Nepal among undergraduate students,
the prevalence rate of Internet addiction was 35.4% (Bhandari et al., 2017). In Japan, among
junior and high school students, the prevalence of Internet addiction was 8.1% (Marioka et al.,
2017), in South India, among 2776 University students, the prevalence was 29.9% for mild
Internet addiction, 16.4% for moderate addictive use and 0.5% for severe Internet addiction
(Anand et al., 2018). In Asia, Internet use is indeed a problematic issue, and a public health
concern.
Malaysia, a multiethnic country located in Southeast Asia, is not without its negative
consequences of technological advancement. As the country becomes more advanced,
developed, and technologically savvy, this comes with a price. Based on the Malaysian
Communication and Multimedia Commission (MCMC), Internet addiction among Malaysians
has reached an alarming rate. According to the MCMC (2017), smartphones are the most
common device to access the Internet (89.4%), with 57.4% of users being male, and 67.2%
being from urban areas. Additionally, 83.2% of children aged 5 – 17 use the internet. In the
Malaysian Internet User survey (MCMC, 2015), university/college students comprised of
62.5% of internet users who were schooling. 80% accessed the web for social media usage
with the average usage period being over four hours a day, and 89% were found to be
addicted to the Internet. Further, 60% of the respondents showed elevated levels of anxiety
and 32% suffered from major depression. These findings are a cause for concern with
negative implications for the individual, family and the community.

Wen Ting Tong, Md. Ashraful Islam, Wah Yun Low, Wan Yuen Choo, and Adina Abdullah
65
Other local Malaysian studies also showed a variation in the prevalence rates of
Internet addiction due to the methodology employed. Cheng and Li’s (2014) meta-analysis of
31 nations across seven regions in the world, among 12-18 years old adolescents, 2.4% of
Malaysian adolescents were reported being addicted to Internet, and 35.1% were found
having problematic Internet use. In one study among secondary school students, 28.6% of the
respondents were addicted to the Internet (Mohd Isa, Hashim, Kaur, & Ng, 2016). Yet, in
another study among undergraduate students in a public university, the prevalence of Internet
addiction was 7.8% and 56.5% were problematic Internet users (Rosliza, Ragubathi,
Mohamad Yusoff, & Shaharuddin, 2018). Zainudin, Md Din, & Othman, (2013) also in their
study among undergraduate students, found 30% prevalence of excessive Internet users.
Among local medical students, a study showed a prevalence of 36.9% Internet addiction
(Ching et al., 2017). As to the impact of Internet addiction on young Malaysian adults, (Alam
et al., 2014) showed those adults using Internet excessively were having problems, such as,
interpersonal, behavioral, physical, psychological and work problems in their daily lives.
Internet addiction in adolescents and young adults has become a public health issue
and has an impact on health education and health promotion. Excessive and inappropriate use
of the Internet can pose serious negative consequences on one’s mental health and quality of
life (Kuss & Griffiths, 2012, Alam et al., 2014). Thus, this paper examined the prevalence of
pathological internal use among university students in Kuala Lumpur and its associated
factors. It is hypothesized that pathological Internet use is associated with socio-demographic
factors, gender, age, life satisfaction, time spent on Internet, Internet use patterns, history of
child abuse and other psychosocial factors. The literature reviewed will further illustrate the
relationship between pathological Internet use and its various associated factors.
Literature Review
There are many factors associated with Internet use, such as sociodemographic
variables, such as gender, time spent online, psychosocial factors, life satisfaction, and history
of child abuse and other comorbid symptoms, such as depression, harmful substance abuse
and sleeping disorder. Socio-demographic factors such as gender, family socio-economic
status, types of residence; duration of Internet use for study or recreational purpose;
psychosocial factors such as low academic achievement, low life satisfaction; and comorbid
symptoms such as alcohol and substance use and depression have been associated with PIU in
adolescents and young people (Kuss, Griffiths, Karila, & Billieux, 2014, Turnbull et al.,
2018). Bozogplan, Demirer, & Sahin, (2013) found that loneliness, self-esteem and life
satisfaction explained 38% of the total variance in Internet addiction.
A number of studies have shown that the male gender is more susceptible to Internet
addiction (Carli et al., 2013; Anand et al., 2018). Anand et al. (2018) in their study among
undergraduate students, aged 18-21 years old in South India found that IA was higher among
male students, i.e. 2.8 times at a higher risk of engaging IA. One study among school
adolescents in China, showed that mild and severe IA was significantly higher in boys than in
girls (Xin et. al., 2018). Among Malaysian medical students, the male students were 1.8 times
more at risk of Internet addiction (Ching, et al., 2017). College and University students are
more susceptible to PIU (Kim, Griffiths, Lau, Fong, & Lam, 2013; Ozcan, & Buzlu, 2007;
Chi, Lin, & Zhang, 2016) due to reasons such as early exposure to the Internet, lack of
parental supervision, the availability and free access to the Internet at the university campus,

Pathological Internet Use among Undergraduate Students
66
the need to use Internet to perform academic activities (Ko, Yen, Chen, Chen, & Yen, 2008)
to cope with anxiety, depression, and stress of university’s life (Hicks & Heastie, 2008) and,
for social networking (van Rooij, Schoenmakers, van de Eijnden, & van de Mheen, 2010).
The quality of the family environment and parent-child relationships were also shown to be
linked to Internet addiction (Chi, Lin, & Zhang, 2016). The excessive use of the Internet has
also impacted on students’ academic performance and social interaction (Yen, Yen, & Ko,
2010; Durkee et al., 2016; Turnbull et al., 2018).
Internet use pattern is also something to reckon with, as the pattern varies from study
to study. Mak et al., (2014) in their six Asian countries epidemiological study of Internet
behaviors among adolescents aged 12-18 years old, found that emails (66%), instant messages
(50%), blogging (25%), and visiting leisure web sites (20%) are relatively more common in
Japan, whereas social networking (65%), newsgroup/discussion groups/forums (19%), non-
purposive web surfing (27%), online shopping (8%), and downloading (28%) are relatively
more common in Hong Kong. In Malaysia, most common are for social networking (38%),
followed by online gaming (19%), downloading (19%), web surfing (14%), visiting leisure
websites (13%), email (12%), listening to online radio (10%), Instant messenger (9%), and
others (Mak et al., 2014). In another ASEAN study of 5 countries among undergraduate
students (Turnbull et al., 2018), it was found that among those with PIU, overall Internet
usage was more than 5 hours/day, followed by Internet use for recreation purposes (more than
3 hours/per), Internet for pornography, Smartphone use, and Internet use for study purposes.
It is obvious that students use the Internet for a variety of purposes. A Malaysian study on
medical students found that the use of Internet was mainly for entertainment purposes,
followed by education and the mixture of both entertainment and education purposes (Ching,
et al., 2017). Based on the recent Internet Users Survey 2017 (MCMC, 2017), text
communication (96.3%) and visiting social network site (89.3%), were the most common
activities for Internet users as well as getting information online (86.9%).
A study among Hong Kong and Macau university students have reported that having
more liberal sexual attitudes, stronger perception that sex is as an instrument for biological
needs, poor attitudes towards contraception and ever had sexual experience were significantly
associated with PIU (Ding et al., 2016). Childhood trauma particularly physical and
emotional abuse significantly increases risks of developing PIU (Zhang et al., 2009;
Dalbudak, Evren, Aldemir, & Evren, 2014; Turnbull et. al., 2018). Furthermore, adolescents
who had experienced sexual abuse showed lower self-esteem, more depressive symptoms, and
greater problematic Internet use compared to adolescents who have not experienced sexual
abuse (Kim, Park, & Park, 2017). Childhood abuse has also been related to post-traumatic
stress disorder (PTSD) (Ginzburg et al., 2009; Hsieh et al., 2016). People who have
experienced traumatic events may use avoidance as a means to cope with their negative
memories and emotions and one way to do that is to use the Internet as distraction and this
may lead to addiction.
PIU has impact on physical and psychological health due to poorer diet, less regular
exercise, sedentary activities, less sleep (Kim & Chun, 2005, Mak et al. 2014) resulting in
obesity (Mak et al., 2014; Tsitsika et al., 2016), lower self-perceived immune function (Reed,
Vile, Osborne, Romano, & Truzoli, 2015) and health status; as well as mental problems such
as depression, social anxiety, attention-deficit hyperactive disorder and psychosocial well-
being (Ko, Yen, Yen, Chen, & Chen, 2012; Lai et al., 2015; Mak et al 2014; Sung, Noh, Park,
& Ahn 2013). Co-morbid symptoms, such as, gambling problem, harmful use of alcohol,

Wen Ting Tong, Md. Ashraful Islam, Wah Yun Low, Wan Yuen Choo, and Adina Abdullah
67
drug use in the past 12 months, mental distress, e.g. severe depression, PTSD symptoms,
sleeping problems and suicidal attempts, are also related to PIU (Turnbull et al. 2018). Carli et
al., (2013) and Yen et al., (2007) suggest that depression is a leading comorbid disorder with
IA. Individuals with negative self-esteem are at risk of engaging in addictive Internet
behaviors which helps to momentarily free themselves of their negative self-esteem, irrational
cognitive assumptions and associated unpleasant emotions (Griffiths, 2000). Thus, one way
of relieving stress among university students is by interacting with their computers, and this
works as a coping mechanisms for them.
PIU is indeed a very complex issue and can manifest itself in a pathological,
behavioral and emotional way and there are also many theories that explain PIU. Among
others, is the cognitive-behavioral model of pathological Internet use proposed by Davis
(2001) to explain the etiology of this phenomenon. This model emphasizes the individual’s
cognitions (or thoughts) as the main source of abnormal behavior, and these cognitive
symptoms (e.g. feeling of self-consciousness, low self-esteem, low self-worth, social anxiety,
etc.) of PIU often precede and cause the affective or behavioral symptoms. Thus, the
etiological factor must be present or must have occurred in order for the symptoms to occur
(in this case, the Internet use). So, the maladaptive cognitions (e.g. distorted thoughts and the
thought processes) are sufficient to cause the symptoms of PIU, such as, the obsessive
thoughts about Internet usage, or having less time to do other things, etc. (Davis, 2001). This
model is important to explain the role of cognitions in PIU.
In view of the literature above, it is thus important to determine the extent of PIU and
its associated factors so that interventions can be developed to prevent the onset of negative
consequences of PIU among university students who are the future policy makers of a nation.
Therefore, this study begs the research questions as to how serious is PIU in Malaysia among
University students, and how the various socio-demographic factors, psychosocial issues,
Internet use patterns, history of child abuse, and co-morbid conditions are affected by PIU. It
is also hypothesized that the socio-demographic variables together with the various psycho-
social variables are related to PIU.
Objectives
This study aimed (1) to determine the prevalence of PIU using the Young’s Diagnostic
Questionnaire (Young, 1998), where the diagnosis of PIU was established when there is a
score of ≥ 5; and (2) to determine the associated factors pertaining to socio-demography (age,
gender, perceived income status, academic performance), post-traumatic stress disorder,
history of child abuse, Internet use patterns and co-morbid symptoms using logistic
regression.
Methods
Study Design and Sampling
This cross-sectional study was conducted between July to September 2015 among
Malaysian undergraduate students in a public university in Kuala Lumpur. The particular
university in this study was purposely selected because it is the premier university in the
country. University students were chosen as the literature review has shown that this is the
age group that are most vulnerable, being Internet savvy and frequently exposed to

Pathological Internet Use among Undergraduate Students
68
communications via social media. The total undergraduate student body at the time of study
was 11,908 from 16 faculties, 2 centers and 2 academies. A stratified cluster sampling was
used to draw the sample. All the faculties, centers and academies formed the clusters and were
included in the sampling frame. Within each cluster, the student populations were stratified by
gender in order to obtain equal representation of both males and females. The number of
students selected from each cluster are proportional to size. Undergraduate students from year
1 to 5 form all the clusters, as those who are currently studying at the university were invited
to participate on a voluntary basis.
Measurements
The questionnaire used for this study was a combination of items from the following:
PIU was assessed using the Young’s Diagnostic Questionnaire (YDQ) (Young, 1998).
The YDQ was developed based on the diagnostic criterion of pathological gambling listed in
the DSM-4 (American Psychiatric Association, 1994). The YDQ comprised of 8 “yes” and
“no” items assessing patterns of Internet usage in terms of preoccupation, tolerance, loss of
control, withdrawal, negative consequences, denial, and escapism (scoring 0-8). One point
was given to each “Yes” answer. Diagnosis of PIU was established when there is a score of ≥
5. The Cronbach’s Alpha value was 0.678.
Socio-demographic variables including age, gender, ethnicity, current year of study,
self-perceived economic status and current residence (6 items). The item on self-perceived
economic status had a response options from 1=wealthy (within the highest 25% in your
country in terms of wealth), 2=Quite well-off (within the 50-75% range for your country),
3=Not very well off (within the 25-50% range for your country) and 4=Quite poor (within the
lowest 25% in your country in terms of wealth).
Internet use variables were open ended items on number of hours spent on the Internet
in a day, number of hours spent on the Internet for study purposes and recreational purposes
in a day, number of hours spent on the Internet for pornography in a week and number of
hours using smartphone in a day (5 items).
Psychosocial variables included items from the World Health Organization adverse
childhood experience scale (CDC, 2016; WHO, 2016) to measure child abuse experiences in
terms of emotional (5 items; Cronbach’s Alpha (0.78)), physical (2 items; Cronbach’s Alpha
(0.74)) and sexual abuse (4 items; Cronbach’s Alpha (0.81)).
Self-perceived life satisfaction was measured using one-item: “All things considered,
how satisfied are you with your life as a whole?” adapted from Lucas & Donnellan (2012).
The response options ranged from 1=very satisfied to 5=very dissatisfied.
Self-perceived academic performance was measured using one-item “How would you
rate your academic performance” with response options from 1=excellent to 5=poor.
Co-morbid symptoms measured were: gambling, measured using the item “Have you
felt that you might have a problem with gambling?” with response options from 0=never,
1=sometimes, 2=most of the time, 3=almost always; tobacco use measured using the item
“Do you currently use one or more of the following tobacco products (cigarettes, snuff,
chewing tobacco, cigars, etc.) with response options “yes” and “no” (World Health
Organization, 1998); and drug use measured using the item “How often have you taken drugs
in the past 12 months; other than prescribed by healthcare providers?” with response options

Wen Ting Tong, Md. Ashraful Islam, Wah Yun Low, Wan Yuen Choo, and Adina Abdullah
69
from 1= 0 times, 2=1-2 times, 3=3-9 times and 4 = ≥ 10 times. The Alcohol Use Disorder
Identification Test (AUDIT-C, 3 items). was used to assess harmful alcohol use, with
response options ranging from 0=never, 1=less than monthly, 2=monthly, 3=weekly and
4=daily or almost daily. A score greater than or equal to three was considered harmful for
women and a score greater than or equal to four for was considered harmful for men (Bush,
Kivlahan, McDonell, Fihn, & Bradley, 1998).
Depression was screened using the Centre for Epidemiological Studies Depression
Scale (CES-D, 10 items) with a Likert-scale from 1=Rarely (<1day), 2=Some/little (1-2 days),
3= Much (3-4 days) and 4=Most (5-7 days). A score of more than 10 indicates moderate
depression, ≥ 15 indicates severe depression (Andresen, Malmgren, Carter, & Patrick, 1994).
The Cronbach’s Alpha value was 0.71.
Post-traumatic stress disorder (PTSD) in the past month was screened using Breslau’s
7-item screening scale. A score of ≥ four indicates positive for PTSD (Kimerling et al., 2006).
The Cronbach’s Alpha value was 0.81.
Data Collection Process
This study received ethics approval from the University of Malaya Medical Ethics
Committee (ref: MECID.NO: 201412-905).
On data collection days, the enumerators were stationed at common student areas
within the faculties, centers and academies where students were most likely to be found.
Trained enumerators approached students and explained about the study purpose, anonymity,
voluntariness and the consent of participation study. Written informed consent were obtained
from them before paper questionnaire were given for self-administration. Data collection
ceased when the number of samples required for both gender within each cluster were
achieved.
Data Analysis
Descriptive analyses were performed to provide socio-demographic, psychosocial,
Internet use patterns and co-morbid symptoms information for non-PIU and PIU and,
addiction symptoms that were common among the respondents. Bivariate logistic regression
was performed to determine the associations between the socio-demographic, Internet use,
psychosocial and co-morbid symptoms variables with respondents with PIU. Factors, which
were found to be significant, were then included in the model for multiple binary logistic
regression using the ‘Enter’ method, and the factors associated with PIU were identified. The
crude and adjusted odds ratios with its 95% confidence interval were reported, where
applicable. Data analysis was conducted using Statistical Package for the Social Sciences
Version 20 (IBM Corporation, 2015). Significance were determined by using p<0.05.
Results
A total of 1132 students were approached during data collection. However, only 1023
completed the questionnaire were included in the data analysis (response rate: 90.4%). The
mean age of the respondents was 20.73 ± 1.49 years old (ranging from 18 to 28 years old).
There was almost equal proportion of female (50.90%) and male (49.1%) respondents. More
than half of the respondents were Malay (52.3%) followed by Chinese (40.0%), Indian (3.9%)

Pathological Internet Use among Undergraduate Students
70
and Others (3.8%). Most respondents were in Year 1 (37.8%) and perceived themselves to be
from family from low socio-economic status (58.8%).
Regarding Internet use, 28.9% were pathological users and among these, majority
were Chinese (31.5%), 22 years old and above (31.0%), in Year 1 (31.5%) and perceived
themselves to be from family from higher socio-economic status (32.5%).
Overall, the respondents’ average hours of Internet use per day was 7.18 hours; an
average of 3.09 ± 2.23 hours were used for study purpose (per day), 4.09 ± 3.30 hours for
recreational purpose (per day) and 0.55 ± 2.39 hours for pornography purpose (per week).
Among those who use Internet for pornography purpose, highest proportion was found among
the pathological users (36.3%). Almost seventy percent (68.9%) of the respondents use
Internet at least five hours per day whereas 72.2% student use smart phone at least four hours
per day.
Among the pathological users, most of them had low self-perceived life satisfaction
(33.2%) and self-perceived poor academic performance (34.5%). Gambling problem (45.4%),
tobacco use (41.9%), drug use (44.9%), depression (46.2%) and PTSD symptoms (50.0%)
were most commonly reported among pathological users compared to non-PIU (Table 1).
Table 1
Distribution of respondents by socio-demography, psychosocial, Internet use patterns and co-
morbid symptoms and types of Internet users
Characteristics Sample
n (%)
Non-pathological
Internet users n (%)
Pathological users
n (%)
Socio-demographic
All 1023 727 (71.1) 296 (28.9)
Gender
Male 501 (49.1) 338 (67.4) 163 (32.6)
Female 519 (50.9) 387 (74.6) 132 (25.4)
Age, mean ± S.D. (years) 20.73 ± 1.49
18-19 254 (24.8) 189 (74.4) 65 (25.6)
20-21 443 (43.3) 313 (70.6) 130 (29.4)
22 and above 326 (31.9) 225 (69.0) 101 (31.0)
Ethnicity
Malay 535 (52.3) 386 (72.1) 149 (27.9)
Chinese 409 (40.0) 280 (68.5) 129 (31.5)
Indian 40 (3.9) 32 (80.0) 8 (20.0)
Others 39 (3.8) 29 (74.4) 10 (25.6)
Current year of study
Year 1 387 (37.8) 265 (68.5) 122 (31.5)
Year 2 284 (27.8) 204 (71.9) 80 (28.1)
Year 3 154 (15.1) 113 (73.4) 41 (26.6)
Year 4 187 (18.3) 137 (73.3) 50 (26.7)
Year 5 11 ( 1.1) 8 (72.8) 3 (27.2)
Self-perceived family
socio-economic status
High 421 (41.2) 284 (67.5) 137 (32.5)
Low 602 (58.8) 443 (73.6) 159 (26.4)

Wen Ting Tong, Md. Ashraful Islam, Wah Yun Low, Wan Yuen Choo, and Adina Abdullah
71
Table 1 (Continued)
Characteristics Sample
n (%)
Non-pathological
Internet users
n (%)
Pathological users
n (%)
Duration of Internet use
variables
Mean ± SD
Duration of Internet use
(overall) (hours)
(≥5 hours per day)
7.18 ± 4.37
705 (68.9) 482 (68.4) 223 (31.6)
Duration of Internet use for
study purpose (hours)
(≥3 hours per day)
3.09 ± 2.23
493 (48.2) 347 (70.4) 146 (29.6)
Duration of Internet use for
recreational purpose (hours)
(≥3 hours per day)
4.09 ± 3.30
610 (59.6) 417 (68.4) 193 (31.6)
Duration of Internet use for
pornography purpose
(hours) (past week)
0.55 ± 2.39
201 (19.6) 128 (63.7) 73 (36.3)
Duration of smart phone use
(hours) (≥4 hours per day)
7.72 ± 5.64
739 (72.2)
503 (68.0) 236 (31.9)
Psychosocial factors
Childhood emotional abuse 480 (46.9) 319 (66.5) 161 (33.5)
Childhood physical abuse 410 (40.1) 271 (66.1) 139 (33.9)
Childhood sexual abuse 127 (12.4) 83 (65.4) 44 (34.6)
Self-perceived life satisfaction
Low 190 (18.6) 127 (66.8) 63 (33.2)
Medium 624 (61.0) 440 (70.5) 184 (29.5)
High 209 (20.4) 160 (76.6) 49 (23.4)
Self-perceived academic
performance
Poor 316 (30.9) 207 (65.5) 109 (34.5)
Satisfactory 549 (53.7) 405 (73.8) 144 (26.2)
Excellent 158 (15.4) 115 (72.8) 43 (27.2)
Skipping breakfast
Never 516 (50.4) 383 (74.2) 133 (25.8)
Sometimes 402 (39.1) 271 (67.4) 131 (32.6)
Everyday 105 (10.3) 73 (69.5) 32 (30.5)
Comorbid symptoms
Having gambling problem 55 (25.7) 30 (54.6) 25 (45.4)
Current tobacco use 31 (3.0) 18 (58.1) 13 (41.9)
Harmful alcohol use 20 (10.1) 12 (60.0) 8 (40.0)
Drug use
(Past 12 months)
49 ( 4.8) 27 (55.1) 22 (44.9)
Depression (moderate/severe) 242 (23.7) 130 (53.8) 112 (46.2)
Sleeping problem
(moderate/severe)
680 (66.5) 469 (69.0) 211 (31.0)
PSTD symptoms
(4 or more)
112 (10.9) 56 (50.0) 56 (50.0)

Pathological Internet Use among Undergraduate Students
72
In terms of various addiction symptoms, over half of the respondents were
preoccupied with the Internet and/or smart phone (59.9%) followed by staying on-line longer
than originally intended (57.1%), felt the need to use the Internet and/or smart phone with
increasing amounts of time in order to achieve satisfaction (53.8%) and used the Internet
and/or smart phone as a way of escaping from problems or of relieving a dysphoric mood
(52.0%). Higher proportion of non-PIU stayed on-line longer than originally intended, felt the
need to use the Internet and/or smart phone with increasing amounts of time in order to
achieve satisfaction compared to PIU, repeatedly made unsuccessful efforts to control, cut
back, or stop Internet and/or smart phone use, jeopardized or risked the loss of significant
relationship, job, educational or career opportunity because of the Internet and/or smart phone
and lied to family members, therapist, or others to conceal the extent of involvement with the
Internet and/or smart phone as compared to PIU (Table 2).
Table 2
Distribution of respondents according to various Internet addiction symptoms
Items All
N (%)
Non-pathological
Internet use
N (%)
Pathological
Internet use
N (%)
Preoccupied with the Internet and/or smart
phone (think about previous on-line activity or
anticipate next on-line session)
613 (59.9) 433 (59.6) 180 (60.8)
Stay on-line longer than originally intended 585 (57.1) 420 (57.8) 165 (55.7)
Feel the need to use the Internet and/or smart
phone with increasing amounts of time in order
to achieve satisfaction
551 (53.8) 401 (55.2) 150 (50.7)
Use the Internet and/or smart phone as a way of
escaping from problems or of relieving a
dysphoric mood (e.g., feelings of helplessness,
guilt, anxiety, depression)
532 (52.0) 374 (51.4) 158 (53.4)
Repeatedly made unsuccessful efforts to control,
cut back, or stop Internet and/or smart phone use
449 (43.9) 330 (45.4) 119 (40.2)
Feel restless, moody, depressed, or irritable
when attempting to cut down or stop Internet
and/or smart phone use
325 (31.8) 229 (31.5) 96 (32.4)
Jeopardized or risked the loss of significant
relationship, job, educational or career
opportunity because of the Internet and/or smart
phone
186 (18.1) 139 (19.1) 47 (15.9)
Lied to family members, therapist, or others to
conceal the extent of involvement with the
Internet and/or smart phone
146 (14.2) 104 (14.3) 42 (14.2)
Determinants of PIU
Simple logistic regression shows that gender, self-perceived family socio-economic
status, overall Internet use (≥5 hours per day), Internet use for recreational purpose (≥3 hours
per day), Internet use for pornography purpose (past week), smart phone use (≥4 hours per
day), childhood emotional abuse, childhood physical abuse, self-perceived life satisfaction,
self-perceived academic performance, skipping breakfast, having gambling problem,
involvement in drug use (past 12 months), having depression (moderate/severe), sleeping

Wen Ting Tong, Md. Ashraful Islam, Wah Yun Low, Wan Yuen Choo, and Adina Abdullah
73
problem (moderate/severe) and PSTD symptoms (4 or more) were significantly associated
with PIU.
After adjusting for the confounding factors, the factors significantly associated with
PIU were Internet use for recreational purpose (≥3 hours per day) (AOR: 3.89; 95% CI: 1.33
– 11.36 and p<0.05), Internet use for pornography purpose (past week) (AOR: 2.52; 95% CI:
1.07 – 5.93 and p<0.05)), having gambling problem (AOR: 3.65; 95% CI: 1.64 – 8.12 and
p<0.01), involvement in drug use (past 12 months), (AOR: 6.81; 95% CI: 1.42 – 32.77 and
p<0.05) and having depression (moderate/severe) (AOR: 4.32; 95% CI: 1.83 – 10.22 and
p<0.01) (Table 3).
Table 3
Logistic regressions on the determinants of PIU
Characteristics Unadjusted Odds Ratio
(95% confidence interval)
Adjusted Odds Ratio
(95% confidence interval)
Socio-demographic factors
Gender
Female 1 (Reference) 1 (Reference)
Male 1.41 (1.07 – 1.85)* 0.39 (0.15 – 1.01)
Age in Year
18-19 1 (Reference) ---
20-21 1.21 (0.88 – 1.71) ---
22 and above 1.29 (0.89 – 1.87) ---
Ethnicity
Malay 1.11 (0.53 - 2.35) ---
Chinese 1.33 (0.63 - 2.82) ---
Indian 0.72 (0.25 - 2.08) ---
Others 1 (Reference) ---
Current year of study
Year 1 1.22 (0.32 - 4.70) ---
Year 2 1.04 (0.27 - 4.04) ---
Year 3 0.96 (0.24 – 3.82) ---
Year 4 0.93 (0.24 – 3.81) ---
Year 5 1 (Reference) ---
Self-perceived family socio-economic status
Low 1 (Reference) 1 (Reference)
High 0.75 (0.57 – 0.98)* 0.61 (0.27 – 1.33)
Internet use factors
Overall Internet Use (≥5 hours per
day)
1.54 (1.13 – 2.09)* 0.44 (0.15 – 1.26)
Study purpose (≥3 hours per day) 1.05 (0.80 – 1.38) ---
Recreational purpose (≥3 hours per
day)
1.38 (1.04 – 1.83)* 3.89 (1.33 – 11.36)*
Pornography purpose (past week) 1.54 (1.11 – 2.14)* 2.52 (1.07 – 5.92)*
Smart phone use (≥4 hours per
day)
1.74 (1.25 – 2.40)* 1.030 (0.448 – 2.367)
Psychosocial factors
Childhood emotional abuse 1.53 (1.17 – 2.01)* 0.85 (0.29 – 2.48)
Childhood physical abuse 1.50 (1.14 – 1.97)* 2.26 (0.79 – 6.43)
Childhood sexual abuse 1.37 (0.92 – 2.03)

Pathological Internet Use among Undergraduate Students
74
Table 3 (Continued)
Characteristics Unadjusted Odds Ratio
(95% confidence interval)
Adjusted Odds Ratio
(95% confidence interval)
Self-perceived life satisfaction
Low 1 (Reference) 1 (Reference)
Medium 0.83 (0.59 – 1.18) 1.55 (0.59 – 4.12)
High 0.60 (0.36 – 0.93)* 2.21 (0.58 – 8.31)
Self-perceived academic performance
Poor 1 (Reference) 1 (Reference)
Satisfactory 0.67 (0.50 – 0.91)* 0.63 (0.27 – 1.48)
Excellent 0.69 (0.54 – 1.05) 2.17 (0.80 – 5.85)
Skipping breakfast 1.36 (1.03 – 1.79)* 1.77 (0.82 – 3.82)
Comorbid symptom factors
Having gambling problem 3.30 (1.71 – 6.38)** 3.64 (1.63 – 8.12)*
Current tobacco use 1.81 (0.87 – 3.74) ---
Harmful alcohol use 1.96 (0.75 – 5.08) ---
Drug use (Past 12 months) 2.08 (1.16 – 3.72)* 6.81 (1.41 – 32.77)*
Depression (moderate/severe) 2.78 (2.05 – 3.77)* 4.31 (1.82 – 10.21)*
Sleeping problem (moderate/severe) 1.36 (1.01 – 1.83)* 0.99 (0.42 – 2.36)
PTSD symptoms (4 or more) 2.91 (1.94 – 4.35)** 0.54 (0.15 – 1.90)
Note: PSTD= Posttraumatic stress disorder; **p<0.001; *p<0.05
Multiple logistic regression, ‘Enter’ method was applied; Multicollinearity were checked and not found;
Hosmer-Lemeshow test, (p=0.725); Pearson chi-square and Significant for Model (p< 0.001) and Classification
table (overall correctly classified percentage=79) were applied to check the model fitness
Discussion
Prevalence of PIU
The prevalence of PIU found in this study was 28.9%, using the 8-items YDQ. In
another Malaysian study, which also used the YDQ, a higher prevalence was reported (43%).
However, the sample size of the study was only 162 (Ng, Isa, Hashim, Pillai, & Harbajan
Singh, 2015). Compared to studies in other Asian nations, the prevalence rate of PIU among
1262 undergraduates from Hong Kong was lower at 15.7%; determined using the Chen’s
Internet Addiction Scale (CIAS) (Kim, Griffiths, Lau, Fong, & Lam, 2013). Similarly, in a
nationally representative sample of 3616 college students in Taiwan, the prevalence of IA was
found to be 15.3% using the revised version of the CIAS (Lin, Ko, & Wu, 2011). A lower
prevalence of IA at 9.7% was also found among 1123 Turkish college students using the
Turkish version of the 36-items Internet Addition Scale (IAS) (Canan, Ataoglu, Ozcetin, &
Icmeli, 2012). However, Jordanian undergraduates students reported a higher prevalence of
40% (n=235/587); determined using the 20-items from Young’s Internet Addiction Test
(IAT) (Alzayyat, Al-Gamal, & Ahmad, 2015). The variation in the prevalence of IA was
attributed to the different assessment tools, classification criteria used and sample sizes.
Currently, only Internet gaming disorder was listed in the Appendix of the Diagnostic and
Statistical Manual for Mental Disorders (DSM) version 5, which calls for further empirical
and clinical research for the condition. There is no gold standard for assessment of Internet