Addiction Research and Theory, February 2013; 21(1): 62–69

Copyright ß 2013 Informa UK Ltd.

ISSN: 1606-6359 print/1476-7392 online
DOI: 10.3109/16066359.2012.690053

Internet addiction of adolescents in China: Prevalence, predictors,

and association with well-being

Ligang Wang1, Jing Luo1, Yu Bai1, Jie Kong2, Jing Luo1, Wenbin Gao1, & Xinying Sun3
1
Chinese Academy of Sciences, Institute of Psychology, Beijing, China, 2North College of Beijing University of
Chemical Technology, Langfang, China, and 3Department of Social Medicine and Health Education, School of
Public Health, Peking University, Xuyuan Road 38, Haidian District, Beijing, China
Addict Res Theory Downloaded from informahealthcare.com by University of Alberta on 06/06/13

(Received 9 February 2012; revised 22 April 2012; accepted 27 April 2012)

Internet addiction is a mental health problem that INTRODUCTION

affects a significant number of people worldwide. Since the proliferation of computer networking sys-
Our study attempted to investigate the prevalence of tems, particularly the Internet, there has been a rapid
Internet addiction among Chinese adolescents and emergence of online activities, such as online chatting,
to explore the predictors of Internet addiction and gaming, television, and shopping. Data from the China
its association with well-being. A total of 10,988 Internet Network Information Center showed that as of
adolescents from nine different cities in China were June 2011, 27.3% of the 485 million people who had
For personal use only.

surveyed using the Diagnostic Questionnaire (DQ) used the Internet were teenagers. With the increasing
for Internet addiction, the Center for Epidemiologic number of Internet users, attention has been given in
Studies Depression Scale, the Rosenberg Self-esteem recent years to what some researchers have termed
Scale, and the Adolescent’s Satisfaction with Life ‘‘Internet addiction.’’
Scale. The mean age of the whole sample was 17.2
years (ranging from 13 to 23 years). The prevalence Measurement of Internet addiction
rate of Internet addiction among the surveyed First introduced by Goldberg (1995) and made popular
adolescents was 7.5%, which was influenced by in Young’s (1996) groundbreaking research, the term
gender and grade (2 ¼ 74.027, p < 0.001; 2 ¼ 7.162, Internet Addiction Disorder (IAD) was defined as ‘‘the
p < 0.05). The breadth of extracurricular activities, compulsive overuse of the Internet and the irritable or
the age when people used Internet for the first time, moody behavior when deprived of it’’ (Mitchell, 2000).
and whether people used Internet for the first time Rice (2005) defined Internet addiction as a proclivity
in Internet bar were significant predictors of toward compulsive use of the Internet that interfered
Internet addiction ( ¼ 0.065, p < 0.001; with one’s ability to lead a normal life. One common
¼ 0.101, p < 0.001; ¼ 0.545, p < 0.001). Finally, thread existing in these definitions and conceptualiza-
our study found evidence demonstrating the link tions is that Internet addiction is a form of behavioral
between Internet addiction and well-being. addiction. However, whether Internet addiction is
Increased symptoms of problematic use were asso- classified as impulse control disorder or a behavior
ciated with decreased self-esteem (F ¼ 258.344, addiction has not been confirmed until now.
p < 0.001), satisfaction with life (F ¼ 232.428, Furthermore, the operational definition of Internet
p < 0.001), and increased depression (F ¼ 607.062, addiction has not been conclusively developed and is
p < 0.001). not included in the DSM IV-TR. More empirical data
focusing on Internet addiction are necessary to support
Keywords: Internet addiction, predictor, prevalence, well-being the definition and classification for Internet addiction.
In China, Tao et al. (2008) proposed a criterion for
clinical diagnosis of Internet addiction that was worked
out accordingly with the principles of evidence-based

Correspondence: Wenbin Gao, Chinese Academy of Sciences, Institute of Psychology, Beijing, China. Tel/Fax: þ86-10-64842391.
E-mail: gaowb@psych.ac.cn; Xinying Sun, Xuyuan Road 38, Haidian District, Department of Social Medicine and Health Education,
School of Public Health, Peking University, Beijing 100191, China. Tel/Fax: þ86-10-82801743-804. E-mail: xysun@bjmu.edu.cn
62
 PREVALENCE AND PREDICTORS OF INTERNET ADDICTION 63

medicine, with high consistency on evaluations made results. In Brenner’s (1997) study, men and women
by psychiatric rates, and with operational convenience. did not differ in either the time spent online or the
Although this criterion has long been questioned, the number of related problems experienced. Young (1998)
prevalence of Internet addiction has caught the atten- used an eight-item DQ to assess self-selected samples
tion of researchers and it has been suggested that and reported that female college students suffered from
Internet addiction should be included in the Chinese Internet addiction more than males. Therefore, the
Classification of Mental Disorders Version 3. second aim of this study was to determine the
Previous studies demonstrated that usual comorbid prevalence of Internet addiction in this sample and its
psychiatric symptoms of Internet Addiction were association with demographic variables.
Attention Deficit and Hyperactivity Disorder (ADHD;
Chan & Rabinowitz, 2006; Yoo et al., 2004), Predictors of Internet addiction
Depression (Ha et al., 2006; Kim et al., 2006), Social The third aim of this study was to examine possible
phobia (Shepherd & Edelmann, 2005), and Hostility predictors of Internet addiction. In our study, four
(Gerra et al., 2004). It appears reasonable to suggest variables were examined: the breadth of Internet use,
that effective evaluation of, and treatment for ADHD the breadth of extracurricular activities, the age of first
and depressive disorder are required for adolescents Internet use, and the location of first Internet use. The
Addict Res Theory Downloaded from informahealthcare.com by University of Alberta on 06/06/13

with Internet addiction. selection of these predictors was based on the etiology
Numerous instruments for assessing IAD have been of problematic general Internet use proposed by Grohol
established, including the Internet addiction DQ (1999). In Grohol’s model of pathological Internet use,
(Young, 1999), the Pathological Use Scale (Morahan- Internet addiction involved the phasic behaviors of
Martin & Schumacker, 2000), the Internet-Related individuals. Grohol suggested that most people with
Addictive Behavior Inventory (Brenner, 1997), and the Internet addiction were likely to be newcomers to the
Chinese Internet Addiction Scale (Chen & Chou, Internet. In general, Internet users go through three
1999). Among the instruments mentioned above, the stages. In stage I, new users become ‘‘stuck’’ in online
DQ was developed according to Young’s IAD diag- activities or existing online users indulge in a new
nostic criteria and has been used most often because it online activity. In stage II, the Internet users begin to
For personal use only.

is easy to administer. Because only a few studies report avoid the addicted behavior. Finally, a balance between
the validity and reliability of the DQ, especially among Internet use and other activities emerges in stage III.
for use in Chinese studies, the first aim of this study Some people become caught in stage I and never move
was to explore the internal consistency reliability and beyond it, and these individuals may require help to
the criterion-related validity of the measurement. reach stage III. According to Grohol’s model, we must
In particular, many previous studies have identified consider two questions: (1) what kinds of new users
an association between Internet addiction, online time become ‘‘stuck’’ in online activities most easily or find
(Cao & Su, 2006; Chen & Chou, 1999; Young, 1998), it difficult to avoid the addictive behavior? and (2)
and academic performance (O’Reilly, 1996; Soule, What kinds of users find it easy to obtain a balance
Shell, & Kleen, 2003), suggesting that people who are between Internet use and other activities?
dependent on the Internet spend more time online and A negative correlation has been identified between
exhibit poorer academic performance. Therefore, we Internet addiction and self-control (E.J. Kim,
use online time and academic performance as the Namkoong, Ku, & S.J. Kim, 2008; Oh, 2003). Self-
criterion variables for Internet addiction in our study. control increases with age and is defined as the ability
to resist an impulse, drive, or temptation to perform an
Prevalence of Internet addiction action (Vaughn, Kopp, & Krakow, 1984). It may be
Internet addiction is currently becoming a serious presumed that younger users are more vulnerable to
mental health problem among Chinese adolescents. IAD because of their lack of self-control.
Previous studies have reported that the incidence of Valcke, Bonte, Wever, and Rots (2010) reported that
Internet addiction among Chinese adolescents is parenting styles significantly affected children’s
2.4–10.6% (Cao & Su, 2006; Chou & Hsiao, 2000; Internet usage. The highest level of Internet use by
Wu & Zhu, 2004). Soule et al. (2003) identified several children was evident when parents adopted a permissive
groups of individuals that were vulnerable to IAD, parenting style, and the lowest level was observed when
including singles, young males, college students, gays, parents adopted an authoritarian Internet parenting style.
middle-aged females, and the less educated, which In general, parents have no control when children use the
suggests that demographic features influence the Internet in an Internet bar. It may be presumed that
prevalence of IAD. The effect of gender on Internet adolescents who use the Internet for the first time in an
addiction has been noted internationally. Some Internet bar are more susceptible to IAD.
researchers have reported a greater prevalence of Some previous studies have identified a negative
Internet addiction among male users than among association between Internet addiction and social
female users (Chou & Hsiao, 2000; Griffiths, 2000; involvement (Iacovelli & Valenti, 2009; Kraut et al.,
Morahan-Martin & Schumacker, 2000; Scherer, 1997). 1998). A research study conducted in China revealed
Other research findings have shown inconsistent that physical exercise was an effective method of
 64 L. WANG ET AL.

treating IAD and suggested that physical exercise could H3b: Adolescents who used the Internet for the first
shift adolescents’ attention away from Internet use time in an Internet bar have a higher risk of becoming
(Hua, 2006). Therefore, we presume that adolescents addicted Internet users than those who used the Internet
whose online activities and extracurricular activities for the first time in other places.
are rich and colorful are easy more likely to obtain the H3c: Users with more concentrated use pattern had a
balance identified in stage III. higher likelihood to be addictive.
The types of services users prefer affected the H3d: The more types of extracurricular activities
overall range of their Internet use, with game and adolescents engage in, the lower their tendency
entertainment users being more prone to addiction, so toward Internet addiction.
their Internet use often had a narrow scope governed by H4a: Addicted Internet users have lower levels of
their addictive behavior (Na, Park, & Kim, 2007). self-esteem than non-addicted users.
Park’s (2009) study revealed that the more concen- H4b: Adolescents with IAD score higher for depression
trated a person’s Internet usage, the more likely it is than those without IAD.
that he or she will unexpectedly be exposed to negative H4c: Internet addicts are more dissatisfied with life
content (i.e. violence, sexual content, and slander). than non-addicted Internet users.
Logically, if people use the Internet for a sole purpose
Addict Res Theory Downloaded from informahealthcare.com by University of Alberta on 06/06/13

or their Internet usage are more concentrated, they are

more likely to become ‘‘stuck’’. However, this link has ME TH O D
not been empirically tested.
The Youth Lifestyles Survey in the Internet age was a
The association between Internet addiction and nationally representative, cross-sectional questionnaire
well-being given to adolescents between 13 and 24 years old. Data
From a clinical psychiatry perspective, the profile of were collected in June 2009. The survey was conducted
heavy Internet users may include individuals who in a self-study class and was organized by a researcher
have one or more of the following characteristics: with the help of a head teacher. The survey was
depression, bipolar disorder, sexual compulsion, or approved and supervised by the Institutional Review
For personal use only.

loneliness (Morahan-Martin, 2005). Yang and Tung Board, sponsored by the China Association for Science
(2007) noted that students afflicted with psycholog- and Technology (CAST) and the Ministry of Health of
ical disorders, such as dependence, extreme shyness, the People’s Republic of China.
depression, and low self-esteem, had a high tendency
to become addicted to the Internet. Moreover, both
Young (1998) and Morahan-Martin (2005) have Sample method
argued that Internet abusers use the Internet to In this study, a stratified clustered sample design was
modulate negative moods. A few studies have used, and the primary sampling units, or clusters,
explored the relationship between Internet addiction were classes in schools. We selected 9 of the 34
and users’ social–psychological or personality vari- capital cities in China. Of the nine cities, three lie in
ables, such as sensation seeking (Lavin, Marvin, the east of China (economically developed region),
McLarney, Nola, & Scott, 1999; Lin & Tsai, 2002), three are located in the west of China (undeveloped
pleasure experiencing (C. Chou, J. Chou, & Tyan, region), and the remainder are in the middle of
1999; Chou & Hsiao, 2000), loneliness (Morahan- China (moderately developed countries). Two junior
Martin & Schumacker, 2000), and depression (Young high schools, two senior high schools, and two
& Rogers, 1998). In this study, we conducted a universities in each selected city were randomly
larger-scale investigation that focused on the associ- chosen. We selected two classes from every grade
ation between Internet addiction and depression, self- (junior high schools: junior one (first year) to junior
esteem, and life satisfaction. three (third year); senior high schools: senior one
(first year) to senior three (third year); universities:
Summary of predictions freshman year to senior year). With 40 classes
H1a: Participants who answered "yes" to five or more selected from each city, a total of 360 classes
of the DQ items have poorer academic performance formed our sample group. Within the sample group,
than those below established criteria for IAD. 10,988 paper questionnaires were distributed, and
H1b: Participants who met the criteria of the DQ spent 9532 were entirely completed, resulting in an 86.7%
more time on surfing the Internet than the rest. response rate.
H2a: The prevalence of Internet addiction among male The sample group consisted of 50.5% males
adolescents is higher than females. (N ¼ 4814) and 49.5% females (N ¼ 4718). The mean
H2b: The prevalence of Internet addiction among age was 17.2 years, and the standard deviation was 2.72
college students is higher than other grades. years. Of the 9532 respondents, 29.2% were from
H3a: The younger adolescents are when they use the junior high schools (N ¼ 2783), 41.6% were from
Internet for the first time, the greater their tendency senior high schools (N ¼ 3965), and 29.2% were from
toward Internet addiction. universities (N ¼ 2784).
 PREVALENCE AND PREDICTORS OF INTERNET ADDICTION 65

Measures Variables on Internet use

Internet addiction DQ Criterion variables for Internet addiction.
The DQ comprised eight items. Respondents who Online time: Participants were asked to estimate the
answered ‘‘yes’’ to five or more of the items were average number of hours per day that they used the
classified as addicted Internet users (Dependents), and Internet.
the rest were classified as normal Internet users Academic performance: Participants reported their
(non-dependents) for the purposes of this study. academic achievements as good, average, or poor.
Kuang, Cao, and Dai’s (2011) meta-analysis study
indicated that the criterion-related validity value of DQ
was 0.72 (95% CI: 0.64–0.75), and coefficient alpha Predictors of Internet addiction. Age when you used
was 0.87 (95% CI: 0.84–0.90). the Internet for the first time: Participants were asked
how many years they had been using the Internet.
Location where you accessed the Internet for the
Well-being
first time: Participants were asked at which location
Multiple measures were used to assess well-being in
they accessed the Internet for the first time. Five
this study, including measures of depression,
categories were created: (a) participants who accessed
self-esteem, and satisfaction with life.
Addict Res Theory Downloaded from informahealthcare.com by University of Alberta on 06/06/13

the Internet at home, (b) participants who accessed the

Internet at a school library, (c) participants who
Center for Epidemiologic Studies Depression Scale.The accessed the Internet at an Internet bar, (d) participants
20-item Mandarin Chinese version (Chien & Cheng, who accessed the Internet at another person’s house,
1985) of the Center for Epidemiological Studies’ and (e) participants who accessed the Internet from
Depression (CES-D) Scale was used as a self- another location.
administered evaluation to assess depressive symp- Breadth of Internet use: Respondents reported their
toms, with scaled scores ranging from 0 to 60. Higher breadth of Internet use by choosing from 12 Internet
CES-D scores indicate more severe depression. CES-D activities (e.g., sending email, online chat/discussion,
showed good reliability and structural validity when and Internet games). Higher values on the breadth of
administrated to Chinese adolescents (Chen, Yang, &
For personal use only.

Internet use index were assigned to individuals report-

Li, 2009). The Cronbach’s for the CES-D in this ing the most activities. For example, a value of 1
study was 0.90. represents those users who performed only one activity,
and a value of 12 represents users who performed
Rosenberg Self-esteem Scale. Participants’ self-esteem almost all of online activities.
were assessed using the 10-item Rosenberg Self-esteem Breadth of extracurricular activities: Adolescents
Scale (SES). Each item was answered on a five-point reported their usual extracurricular activities by choos-
Likert-type scale, with higher scores indicating higher ing from 13 extracurricular activities (e.g., watching
self-esteem. Sun’s (2007) study in China reported that TV, doing homework, and playing sports). Higher
the Cronbach’s of SES was 0.736, and validity had values on the breadth of extracurricular activities index
been established through findings of a correlation were assigned to individuals reporting the most
between SES scores and depression (Self-rating activities.
Depression Scale) as well as social anxiety (Social
Anxiety Scale) in expected directions. The Cronbach’s Demographic information
in this study was 0.87. Adolescents reported the following information: age,
gender, grade (i.e., junior high school, senior high
Adolescents’ Satisfaction with Life Scale. This school, and university), whether there was a computer
questionnaire was developed for this study to measure at their home (‘‘yes’’/‘‘no’’), whether they had their
adolescents’ satisfaction with life. The measure con- own computer (‘‘yes’’/‘‘no’’), whether they were the
sists of 11 items across eight domains of the life only child in the family (‘‘yes’’/‘‘no’’), and whether
profile: dietary, sleep, hygiene, study, exercise, enter- they were in residence (‘‘yes’’/‘‘no’’).
tainment, interpersonal relationships, and coping styles.
A four-point scale was used with ‘‘1’’ ¼ very dissat- Statistical analysis
isfied, ‘‘2’’ ¼ dissatisfied, ‘‘3’’ ¼ satisfied, and ‘‘4’’ ¼ We analyzed the associations of Internet addiction with
extremely satisfied. The reliability and validity of the demographic variables, criterion variables, and psy-
scale had been tested previously, and the results chological well-being, by chi-squared or one-way
(presented in our previous paper) indicated that the analysis of variance. The associations between
Adolescents’ Satisfaction with Life Scale (ASLS) Internet addiction and a series of Internet use charac-
could be used as an effective and reliable tool to teristics, including place where youths accessed
measure adolescents’ satisfaction with life (Zhang & Internet for the first time, age when youths used
Gao, 2010). Higher scores indicate more life satisfac- Internet for the first time, breadth of Internet use, and
tion. In this sample, the Cronbach’s of the ASLS breadth of extracurricular activities, were all examined
was 0.83. by hierarchical regression analysis with controlling for
 66 L. WANG ET AL.

Table I. Prevalence of Internet addiction and association with academic performance. Table I shows that adolescents
demographics, criterion variables, and psychological well-being. with Internet addiction received higher scores on online
time than those without Internet addiction, and Internet
Internet addiction addicts were more likely to have poor academic
performance. Therefore, H1a and H1b are empirically
Yes No
accepted, which suggest that the discriminant validity
N(%) or N(%) or of the DQ is good in this study.
Mean(SD) Mean(SD) 2 or F
Prevalence of Internet addiction
Total 712(7.5) 8820(92.5) According to Young’s diagnostic criteria for Internet
Gender
addiction, 7.5% of the participants (N ¼ 712) were
Male 470(9.8) 4344(90.2) 74.027***
Female 242(5.1) 4476(94.9)
addicted Internet users. We found that adolescents with
Grade Internet addiction were more likely to be male (H2a)
Junior high school 192(6.9) 2591(93.1) 7.162* and senior high school students and to have personal
Senior high school 330(8.3) 3635(91.7) computers and computers at home (Table I). These
results suggest that these variables should be controlled
Addict Res Theory Downloaded from informahealthcare.com by University of Alberta on 06/06/13

University 190(6.8) 2594(93.2)

Only child or not when examining the association between predictor
Yes 439(7.9) 5140(92.1) 3.102 variables and Internet addiction.
No 273(6.9) 3680(93.1)
In residence or not
Yes 4586(92.2) 386(7.8) 1.299
Predictors of Internet addiction
No 4234(92.9) 326(7.1) The logistic regression analysis results are presented in
Owning computer Table II. Of the four control variables, grade was
at home or not excluded from the regression equation. After control-
Yes 541(8.2) 6019(91.8) 18.748*** ling for the rest of the demographic variables, adoles-
No 169(5.7) 2783(94.3) cents who accessed the Internet for the first time in
Having personal an Internet bar (H3b) had a higher risk of becoming
For personal use only.

computer or not addicted Internet users ( ¼ 0.545, p < 0.001).

Yes 410(9.1) 4106(90.9) 32.482*** In addition, the age at which adolescents used the
No 300(6.0) 4697(94.0)
Internet for the first time (H3a) and the breadth of
Online time 12.35(14.948) 7.29(10.562) 130.790***
extracurricular activities (H3d) were negatively asso-
Academic
performance ciated with Internet addiction ( ¼ 0.065, p < 0.001;
Good 70(5.7) 1155(94.3) 43.655*** ¼ 0.101, p < 0.001). The abovementioned results
Average 369(6.1) 5702(93.9) were concordant with our hypothesis. However, there
Bad 221(9.9) 2015(90.1) was no significant effect of breadth of online activities
Self-esteem 27.62(5.214) 30.51(4.569) 258.344*** on Internet addiction, and H3c had not been proved in
Depression 22.80(11.048) 13.68(9.370) 607.062*** this study.
Life satisfaction 2.60(0.517) 2.86(0.444) 232.428***

Note: *p < 0.05; and ***p < 0.001. The association of Internet addiction with
well-being
The means and standard deviations of the well-being
variables for the group of addicted Internet users and
demographic variables, including gender, grade, non-addicted users are shown in Table I. Addicted
owning home computer or not, and owning personal Internet users differed significantly from non-addicted
computer or not. In addition, before carrying regression users on the well-being variables of self-esteem,
analysis, some category variables, including locations depression, and life satisfaction, indicating that
of Internet use for the first time and grade, were addicted Internet users had lower levels of well-being
transformed into a series of dummy variables. than non-addicted users. That is, H4a, H4b, and H4c
are empirically accepted.
RESULTS
DISCUSSION
Examination of Internet Addiction DQ
Reliability and validity were explored for the DQ in The DQ utilized in our study to assess the Internet
this study. A reliability analysis of the DQ revealed an addiction was found to have a high internal consis-
internal consistency coefficient of 0.71, indicating that tency. The validity of the measurement was indicated
the items in the measure were highly inter-correlated by showing that participants who had spent more time
for this sample. To provide an indication of validity, we on the Internet or who felt they had poor academic
examined the association between Internet addiction performance showed significantly higher scores on
and criterion variables such as online time and the DQ.
 PREVALENCE AND PREDICTORS OF INTERNET ADDICTION 67

Table II. The association between Internet addiction and Internet usage by forward regression analysis.

Wald OR 95% CI

The place where you accessed Internet for the first time (at home ¼ 0) 24.688***
School library 0.041 1.025 0.808–1.301
Internet bar 17.887*** 1.724 1.340–2.220
Other people’s house 0.517 1.122 0.820–1.534
Other 0.300 1.131 0.728–1.755
The age when you used Internet for the first time 56.489*** 0.904 0.881–0.928
Breadth of Internet use 4.241 1.028 1.003–1.053
Breadth of extracurricular activities 12.248*** 0.937 0.904–0.972
Control variable
Gender 36.385*** 1.717 1.440–2.047
Owning home computer or not 2.909 0.813 0.640–1.031
Having personal computer or not 9.251** 0.753 0.627–0.904

Note: **p < 0.01; and ***p < 0.001.

Addict Res Theory Downloaded from informahealthcare.com by University of Alberta on 06/06/13

This study primarily focused on exploring the Previous studies have found that multi-user dungeon
prevalence and predictors of Internet addiction and its games, Internet relay chat, and chat rooms are the
association with well-being. Our results indicated that Internet activities that most frequently lead to addictive
some adolescents use the Internet excessively, and behavior (Kandell, 1998; Young, 1998). Therefore, we
7.5% of adolescents could be described as suffering presumed that Internet addicts might focus their
from Internet addiction. The prevalence of Internet attention on the abovementioned online activities
addiction varies considerably in previous research (H3c), whereas non-dependent users might conduct
For personal use only.

studies due to different social and cultural contexts, more varied types of online activities. However, the
the methodology used, and different populations of breadth of online activities was not found to predict
respondents. In our study, the male-to-female ratio for addicted Internet use in this study. Internet addicts
Internet-addicted students was 1.9 : 1 (470 males and spent more time using the Internet, which might allow
242 females), indicating a gender difference in Internet addicted users more chances to use online applications
addiction. Most previous studies have reported that other than the applications that contributed to the
college students have an increased risk of developing development of pathological Internet use. Therefore,
Internet addiction because these individuals have more we propose that the distribution of time online is a
free time and of a lack of restrictions from families and more effective predictor than breadth of online activ-
schools after completion of the college entrance ities for future research.
examination (Yu et al., 2009). However, the current The final aim of our study was to explore the
suggested that senior high school students were most relationship between Internet addiction and well-being
likely to be Internet addicts, which did not support our in Chinese adolescents. The results supported our
H2b. A potential reason for these results is the hypothesis that addicted Internet use is related to
decreasing trend of Internet users’ age. With the decreased well-being, including lower self-esteem,
popularity of the Internet, children between the ages less satisfaction with life, and increased depression.
of 5 and 15 have become influential consumers of Because low self-esteem is conceptually and empiri-
online games and social network service (iResearch, cally a function of perceived rejection, abandonment, or
2010). Unfortunately, the Internet content rating indifference by significant others (Leary & MacDonald,
system is currently incomplete in China. 2003; MacDonald & Leary, 2005), the degree of self-
The logistic regression analysis model indicated that esteem is always a predictor of IAD. Previous studies
the breadth of extracurricular activities, the age at first have shown a link between IAD and levels of self-
Internet use, and the location of first Internet use were esteem (Armstrong, Phillips, & Saling, 2000; Ko, Yen,
significant predictors of Internet addiction. This calls C.C. Chen, S.H. Chen, & Yen, 2007; Young & Rogers,
for attention in the Internet policy area. A policy to 1998). In addition, many studies have found that
educate young people about lifestyles is necessary depressed individuals are more likely to engage in
since Internet addiction is related to the breadth of synchronous Internet activities (e.g., chatting, instant
extracurricular activities. If young children learn to messengers, etc.; Caplan, 2003; Kubey, Lavin, &
arrange the spare time in a more balanced way, the risk Barrows, 2001; Young & Rogers, 1998). According to
of indulging in surfing the Internet to children will the association between Internet addiction and well-
decrease. This also has many implications for future being, it could be proposed that some suggestions on
research and for policy regarding when and where to prevention and treatment of IAD. Reeves’ (2000) study
start using the Internet by themselves. revealed that anonymous catharsis via the Internet
 68 L. WANG ET AL.

might be an optional coping strategy for students with Cao, F., & Su, L. (2006). Internet addiction among Chinese
poor interpersonal relationships. In general, adoles- adolescents: Prevalence and psychological features. Child:
cences with bad moods always choose surfing the Care, Health and Development, 33, 275–281.
Internet as a coping strategy, which increased the risk of Caplan, S.E. (2003). Preference for online social interaction: A
theory of problematic Internet use and psychosocial well-
becoming Internet addicts. Therefore, to prevent IAD, it
being. Communication Research, 30, 625–648.
is necessary to pay more attention to depressed students’ Chan, P.A., & Rabinowitz, T. (2006). A cross-sectional analysis
Internet usage. Likewise, in the process of curing IAD, of video games and attention deficit hyperactivity disorder
in addition to changing patients’ Internet use habits, symptoms in adolescents. Annals of General Psychiatry, 5,
raising their self-esteem and elevating their moods are 16.
essential. Chen, S.H., & Chou, C. (1999, August). Development of Chinese
This study has some limitations that should be Internet addiction scale in Taiwan. Poster presented at the
considered when interpreting its findings. The first meeting of the 107th American Psychology Annual
limitation is related to the nature of the self-reported convention. Boston, MA, USA.
data employed in the study to capture individual’s Chen, Z., Yang, X., & Li, X. (2009). Psychometric features of
attitudes and behaviors. Even though such self-report- CES-D in Chinese adolescents. Chinese Journal of Clinical
Psychology, 17, 443–445.
ing is typical of many social science researches,
Addict Res Theory Downloaded from informahealthcare.com by University of Alberta on 06/06/13

Chien, C.P., & Cheng, T.A. (1985). Depression in Taiwan:

findings based on this kind of data cannot thoroughly Epidemiological survey utilizing CES-D. Psychiatria et
account for the operation of the delicate psychological Neurologia Japonica, 87, 335–338.
mechanisms of human minds and behaviors. For Chou, C., Chou, J., & Tyan, N.N. (1999). An exploratory study
example, it is difficult to determine causality in the of Internet addiction, usage and communication pleasure-the
relationship between IAD and well-being. The second Taiwan’s case. Internal Journal of Educational
limitation of the study goes to the measurement of the Telecomunications, 5, 47–64.
breadth of Internet use. In this study, the number of Chou, C., & Hsiao, M.C. (2000). Internet addiction, usage,
online activities adolescents used to take part in was gratification, and pleasure experience: The Taiwan college
regarded as the breadth of Internet use, but maybe a students’ case. Computer and Education, 35, 65–80.
fact was ignored that the longer people stay online, the Gerra, G., Angioni, L., Zaimovic, A., Moi, G., Bussandri, M.,
For personal use only.

Bertacca, S., . . . Nicoli, M.A. (2004). Substance use among

greater was the chance of the person using online
high-school students: Relationships with temperament, per-
applications. Therefore, a sophisticated measurement sonality traits, and parental care perception. Substance Use
of the breadth of Internet use should be used in future and Misuse, 39, 345–367.
study. Zhu and He (2002) introduced the H-statistics Goldberg, I. (1995). Internet addictive disorder (IAD) diagnostic
from entropy in information theory (Shannon & criteria. Retrieved from http://www.psycom.net/
Weaver, 1949) to measure the diversity of online iadcriteria.html
time. In their study, respondents were asked to estimate Griffiths, M.D. (2000). Does Internet and compute ‘‘addiction’’
the minutes per week they spent on six Internet exist? Some cases study evidence. Cyberpsychology and
activities. An H-score was computed with these six Behavior, 3, 211–218.
measures which represents the distribution of six online Grohol, J. (1999). Internet addiction guide. Retrieved from
http://psychcentral.com/netaddiction
activities but was not affected by the amount of online
Ha, J.H., Yoo, H.J., Cho, I.H., Chin, B., Shin, D., & Kim, J.H.
time (Zhu & He, 2002).
(2006). Psychiatric comorbidity assessed in Korean children
and adolescents who screen positive for Internet addiction.
Journal of Clinical Psychiatry, 67, 821–826.
ACKNOWLEDGEMENTS Hua, J. (2006). Research of physical intervention on internet
We express our great appreciation to all of the adolescents addiction disorder of junior middle school students (Published
for participating in this study. Doctoral dissertation, East China Normal University,
Shanghai).
Declaration of interest: This study was funded by the Iacovelli, A., & Valenti, S. (2009). Internet addiction’s effect on
China Association for Science and Technology (CAST). likeability and rapport. Computers in Human Behavior, 25,
The authors report no conflicts of interest. The authors 439–443.
alone are responsible for the content and writing of this iResearch (2010). China children’s netizens and online behavior
article. research report. Retrieved from http://www.iresearch.com.cn/
Report/
Kandell, J.J. (1998). Internet addiction on campus: The
vulnerability of college students. CyberPsychology and
RE FERE N CE S
Behavior, 1, 11–17.
Armstrong, L., Phillips, J.G., & Saling, L.L. (2000). Potential Kim, E.J., Namkoong, K., Ku, T., & Kim, S.J. (2008). The
determinants of heavier internet usage. International Journal relationship between online game addiction and aggression,
of Human Computer Studies, 53, 537–550. self-control and narcissistic personality traits. European
Brenner, V. (1997). Psychology of computer use: XLVII. Psychiatry, 23, 212–218.
Parameters of Internet use, abuse and addiction: The first 90 Kim, K., Ryu, E., Chon, M.-Y., Yeun, E.-J., Choi, S.-Y.,
days of the Internet usage survey. Psychological Reports, 80, Seo, J.-S., & Nam, B.W. (2006). Internet addiction in Korean
879–882. adolescents and its relation to depression and suicidal
 PREVALENCE AND PREDICTORS OF INTERNET ADDICTION 69

ideation: A questionnaire survey. International Journal of Scherer, K. (1997). College life online: Healthy and unhealthy
Nursing Studies, 43, 185–192. Internet use. Journal of College Studies, 38, 655–665.
Ko, C.H., Yen, J.Y., Chen, C.C., Chen, S.H., & Yen, C.F. Shannon, C., & Weaver, W. (1949). The mathematical theory of
(2007). Factors predictive for incidence and remission of communication. Urbana, IL: University of Illinois Press.
Internet addiction in young adolescents: A prospective study. Shepherd, R.M., & Edelmann, R.J. (2005). Reasons for internet
Cyberpsychology and Behavior, 10, 545–551. use and social anxiety. Personality and Individual
Kraut, R., Patterson, M., Lundmark, V., Kiesler, S., Differences, 39, 949–958.
Mukopadhyay, T., & Scherlis, W. (1998). Internet paradox: Soule, L.C., Shell, L.W., & Kleen, B.A. (2003). Exploring
A social technology that reduces social involvement and Internet addiction: Demographic characteristics and stereo-
psychological well-being? American Psychologist, 53, types of heavy internet users. The Journal of Computer
1017–1031. Information Systems, 44, 64–73.
Kuang, X., Cao, Y., & Dai, X. (2011). Meta-analysis of Internet Sun, Q. (2007). Revision of self-esteem scale (Published Master
addiction scales’ reliability and validity. Chinese Journal of thesis, Jinan University, Guangzhou).
Clinical Psychology, 19, 192–194. Tao, R., Huang, X., Wang, J., Liu, C., Zhang, H., Xiao, L., &
Kubey, R.W., Lavin, M.J., & Barrows, J.R. (2001). Internet use Yao, S. (2008). A proposed criterion for clinical diagnosis of
and collegiate academic performance decrements: Early internet addiction. Medical Journal of Chinese People’s
findings. Journal of Communication, 51, 366–382. Liberation Army, 33, 1188–1191.
Addict Res Theory Downloaded from informahealthcare.com by University of Alberta on 06/06/13

Lavin, M., Marvin, K., McLarney, A., Nola, V., & Scott, L. Valcke, M., Bonte, S., Wever, B.D., & Rots, I. (2010). Internet
(1999). Sensation seeking and collegiate vulnerability to parenting styles and the impact on Internet use of primary
Internet dependence. Cyberpsychology and Behavior, 2, school children. Computer and Education, 55, 454–464.
425–430. Vaughn, B.E., Kopp, C.B., & Krakow, J.B. (1984). The
Leary, M.R., & MacDonald, G. (2003). Individual emergence and consolidation of self-control from eighteen
differences in trait self-esteem: A theoretical integration. to thirty months of age: Normative trends and individual
In M.R. Leary & J.P. Tangney (Eds.), Handbook of self and differences. Child Development, 55, 990–1004.
identity (pp. 401–418). New York, NY: Guifford Press. Wu, H.R., & Zhu, K.J. (2004). Path analysis on related factors
Lin, S.S.J., & Tsai, C.C. (2002). Sensation seeking and internet causing internet addiction disorder in college students.
dependence of Taiwanese high school adolescents. Computer Chinese Journal of Public Health, 20, 1363–1364.
in Human Behavior, 18, 411–426. Yang, S.C., & Tung, C.J. (2007). Comparison of Internet addicts
For personal use only.

MacDonald, G., & Leary, M.R. (2005). Why does social

and non-addicts in Taiwanese high school. Computers in
exclusion hurt? The relationship between social and physical
Human Behavior, 23, 79–96.
pain. Psychological Bulletin, 131, 202–223.
Yoo, H.J., Cho, S.C., Ha, J., Yune, S.K., Kim, S.J.,
Mitchell, P. (2000). Internet addiction: Genuine diagnosis or
Hwang, J., . . . Lyoo, I.K. (2004). Attention deficit hyperactiv-
not? The Lancet, 355, 632–635.
ity symptoms and internet addiction. Psychiatry and Clinical
Morahan-Martin, J. (2005). Internet abuse: Addiction? Disorder?
Neurosciences, 58, 487–494.
Symptom? Alternative explanations? Social Science
Young, K. (1996, August 15). Internet addiction: The emer-
Computer Review, 23, 39–48.
gence of a new clinical disorder. Paper presented at the 104th
Morahan-Martin, J.M., & Schumacker, P. (2000). Incidence and
annual meeting of the American Psychological Association,
correlates of pathological Internet use. Computer in Human
Behavior, 16, 13–29. University of Pittsburgh at Bradford, Toronto, Canada.
Na, E., Park, S., & Kim, E. (2007). Media usage and adjustment Young, K.S. (1998). Internet addiction: The emergence of a new
of adolescents according to the types of Internet usage: clinical disorder. Cyberpsychology and Behavior, 1, 237–244.
Focusing on blog-type and game-type. Korean Journal of Young, K.S. (1999). Internet addiction: Symptoms, evaluation
Mass Communication and Journalism, 51, 393–329. and treatment. In L. VandeCreek & T.L. Jackson
Oh, W.O. (2003). Factors influencing internet addiction (Eds.), Innovations in clinical practice: A source book (pp.
tendency among middle school students in Gyeong-buk 19–31). Sarasota, FL: Professional Resource Press.
area. Taehan Kanho Hakhoe Chi, 33, 1135–1144. Young, K.S., & Rogers, R.C. (1998). The relationship between
O’Reilly, M. (1996). Internet addiction: A new disorder enters depression and Internet addiction. Cyberpsychology and
the medical lexicon. Canadian Medical Association Journal, Behavior, 1, 25–28.
154, 1882–1883. Yu, F., Guo, M., Chen, J., Huang, G., Xu, S., Wan, A., . . . Li, M.
Park, S. (2009). Concentration of internet usage and its relation (2009). Epidemiological study and logistic analysis on
to exposure to negative content: Does the gender gap differ internet addiction disorder among 3508 middle and college
among adults and adolescents? Women’s Studies International students. Modern Preventive Medicine, 36, 1675–1677.
Forum, 32, 98–107. Zhang, F., & Gao, W. (2010). Development and validation of
Reeves, P.M. (2000). Coping in cyberspace: The impact of Chinese Adolescent Lifestyles Scale. Chinese Mental Health
internet use on the ability of HIV-positive individuals to deal Journal, 24, 515–519.
with their illness. Journal of Health Communication, 5, Zhu, J.H., & He, Z. (2002). Diffusion, use and impact of the
47–59. Internet in Hong Kong: A chain process model. Journal of
Rice, M. (2005). Online addiction. Beijing Review, 48, 32–33. Computer-Mediated Communication, 7(2), available online.

Not ice of Cor rec tion

The version of this article published online ahead of print on 8 October 2012 contained an error on page 1.
Xinying Sun should also have been listed as a corresponding author. The error has been corrected for this version.