A psychological disorder called ‘Internet addiction’ has newly emerged along with a dramatic increase of worldwide Internet use. However, few studies have used population-level samples nor taken into account contextual factors on Internet addiction.
Methods and Findings
We identified 57,857 middle and high school students (13–18 year olds) from a Korean nationally representative survey, which was surveyed in 2009. To identify associated factors with addictive Internet use, two-level multilevel regression models were fitted with individual-level responses (1st level) nested within schools (2nd level) to estimate associations of individual and school characteristics simultaneously. Gender differences of addictive Internet use were estimated with the regression model stratified by gender. Significant associations were found between addictive Internet use and school grade, parental education, alcohol use, tobacco use, and substance use. Female students in girls' schools were more likely to use Internet addictively than those in coeducational schools. Our results also revealed significant gender differences of addictive Internet use in its associated individual- and school-level factors.
Citation: Heo J, Oh J, Subramanian SV, Kim Y, Kawachi I (2014) Addictive Internet Use among Korean Adolescents: A National Survey. PLoS ONE 9(2): e87819. https://doi.org/10.1371/journal.pone.0087819
Editor: Marianna Mazza, Catholic University of Sacred Heart of Rome, Italy
Received: July 8, 2013; Accepted: January 3, 2014; Published: February 5, 2014
Copyright: © 2014 Heo et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The authors have no support or funding to report.
Competing interests: The authors have declared that no competing interests exist.
Internet use is recognized as an essential part of modern life. Owing to web-based technologies and increases of Internet access in Latin America and Asia, Internet use has increased dramatically across the world reaching the number of global Internet users more than 2.3 billion in 2011 .
On the other side of this popularity, a new psychological disorder has emerged: “Internet addiction”, also inconsistently referred to as “excessive Internet use” , , “problematic Internet use” , , “Internet dependency” , , or “pathological Internet use” , . Such discrepancy is largely attributable to lack of consensus in definitions across studies that focused on different symptoms of Internet addiction. Young  defined Internet addiction as “maladaptive pattern of Internet use leading to clinically significant impairment or distress”. Kandell  later defined it as “a psychological dependence on the Internet, regardless of the type of activity once logged on” . Other studies have even not given it a clear definition. To measure or diagnose these addictive symptoms related with Internet use, some studies have developed their own assessment tools. Most of the Internet addiction studies developed measures based on the Diagnostic and Statistical Manual of Mental Disorders (DSM) criteria . Young  developed the 8-question diagnostic Questionnaire with modification of the criteria for compulsive gambling (DSM-IV). Morahan-Martin and Schumacher  later developed the Pathological Internet Use scale of 13-questions by reconstructing the DSM-IV criteria. More recent studies developed new measures independently with DSM criteria. Using factor analysis methods, Caplan  and Widyanto and Mcmurran  created their own measures. Tao et al.  developed their measure using item-response theory. These variations in the definitions and measures have fueled controversies over inclusion of Internet addiction in the DSM , .
Despite the lack of consensus on its definition and measurement, evidence of Internet addiction has accumulated since the mid-1990s. Case and empirical studies revealed that Internet addiction was characterized by adverse effects on the individual's psychological well-being , , academic failure , , reduced work performance  or job loss , sleep deprivation , social withdrawal , , little or no self-confidence , , poor diet , , family problems , , marital breakdown , and even violence associated with blocked access to online games  or cardiopulmonary-related death from excessive use , .
However, these studies have some limitations. First and most critically, a majority of the research suffered from sampling bias due to convenience sampling and small sample size as they recruited subjects through the Internet , , , –. Inevitably, this sampling of self-selected participants caused mixed or contradicted results between the studies. Second, though the effects of environmental factors on addictive behaviors have been well established , , most of the past papers on Internet addiction have focused primarily on associations with individual personality such as low self-esteem , loneliness , low self-disclosure or anti-social behavior , stronger suicidal intention , and sensation-seeking , , . Specifically, no empirical studies examined the associations with family factors (e.g. family income or parental educational attainment) and school environmental factors though it is well known that parental socioeconomic status (SES) and school characteristics were associated with risks of adolescents' addictive behaviors –. Last, despite past studies having consistently reported higher risks of Internet addiction among boys , , few studies have identified gender differences in Internet addiction.
To fill these gaps in past studies with social epidemiological perspectives, we examine the individual- and contextual-level correlates of Internet addiction with a multilevel statistical method using nationally representative survey data of South Korean adolescents. Due to higher prevalence of Internet addiction in Korean adolescents than adults , we focus on Internet addiction among adolescents. This study also examines gender differences in Internet addiction among the population.
South Korea is one of the most highly digitalized societies in the world. The Internet penetration rate in South Korea exceeded 75 percent in 2011 . More than half of the 50s age-group and almost 100% of teenagers are using the Internet in their daily life . After a series of crimes and death related to Internet addiction, South Korea has regarded Internet addiction as a social and public health problem. The government initially developed the Korean-version of the Internet addiction measurement scale (KS-scale) and has introduced into middle and high schools for screening addictive Internet users . Moreover, to curb excessive online gaming among adolescents, the government implemented compulsive policies called “Internet Shutdown” and “Cooling Off” in 2011 and 2012 respectively to limit adolescents' online gaming at midnight and amounts of time spent for online games . A nation-wide survey specified to Internet addiction in 2010 showed that 8.0% in the whole population were addicted to Internet; 12.4% of adolescents were using Internet addictively . Given that Internet users have been increasing exponentially around the world especially with the popularity of social network services (SNS), this study could provide information to prevent and intervene in adolescent Internet addiction for other countries where it has not emerged yet as a social and public health issue.
We are interested in answering the following questions: 1) Is higher parental SES inversely correlated with adolescents' addictive internet use? 2) Are school contexts associated with adolescents' addictive internet use regardless of individual-level factors? 3) Are these associations of individual- and school-level factors different between genders?
Source of Data
Out of 75,066 samples from the Fifth Korean Youth Risk Behavior Web-based Survey (KYRBWS) conducted in 2009, we identified 57,857 students from 400 middle and 400 high schools after dropping samples missing values for parental education level. The KYRBWS is a nationally representative survey producing annual data to monitor adolescent (13–18 year olds) health behaviors. The KYRBWS was produced by the Korea Centers for Disease Control and Prevention (KCDC) and approved by the ethics committees of the KCDC. The written informed consent was obtained from each student's parents for the survey. To have a nationally representative sample, the survey used the stratified two-stage random cluster sampling method. A total of 800 middle and high schools (primary sampling units) were selected via random sampling from each stratum of 135 strata that were identified using administrative districts and characteristics of schools Then, one class (secondary sampling units) in each school grade was randomly sampled from each selected school. All the students of the sampled classes were requested to complete an anonymous web-based survey during an hour of their regular class time in a computer room of each selected school. Survey objectives and the entire survey process were explained to the students before the survey was conducted. The students were required to log into the KYRBWS website with a randomly assigned number and complete the self-administered questionnaire. The overall response rate of the Fifth KYRBWS study was 97.6%.
Internet addiction was assessed by the simplified Korean Internet Addiction Self-assessment Tool (KS scale) (see Table S1), which was developed by the Korean government and used nation-wide in Korea with a definition of “having trouble in one's daily life due to withdrawal and tolerance in Internet use regardless of devices” . The test for reliability and construct validity of the scale is described in more detail elsewhere . This official measure has been adopted for nationwide Internet addiction screening and annual surveillance among Korean adolescents . The scale consisted of 20 questions inquiring about 6 domains: disturbance of adaptive functions, positive anticipation, withdrawal, virtual interpersonal relationship, deviant behaviors, and tolerance. Responses were scaled with 4 categories from “never” to “always yes”. In this study, rather than adopting the measurement itself that has cut-points of three categories (addiction, latent addiction, and normal), we measured the severity of Internet addiction with a continuous variable by summation of each response [from 1 (never) to 4 (always yes)] with a range from 20 to 80. We treated this score of addictive Internet use as an outcome variable in the study.
As shown in table 1, key individual-level variables used in the analysis included demographic characteristics; self-rated academic achievement; parental socioeconomic status (SES); tobacco, alcohol, and substance use; and physical activities and psychological status. Self-rated academic achievement was a five-level categorized variable from very high to very low. We treated self-rated academic achievement as a continuous variable in the main analysis. Parental SES was measured by parent's educational attainment and the Family Affluence Scale (FAS) . Paternal and maternal educational attainment were categorized in three levels (middle school-or-less, high school, and college-or-higher). The FAS was measured by summation of answers of four items: 1) having one's own bedroom (yes = 1, no = 0); 2) frequency of family trips per year; 3) the number of computers at home; and 4) the number of vehicles owned by family. Tobacco and alcohol use were measured by the average number of cigarettes and average volume of alcohol consumed in the past 30 days. Substance use was categorized into three levels: never, past use, and current use. Categories of physical activity were strenuous exercise, moderate exercise, and weight training, which were estimated by the number of days of exercise over 30 minutes, 20 minutes, and days of weight training, respectively. Of psychological factors, self-rated sleep satisfaction was scaled into five categories from very good to very poor. Depressive symptoms and suicidal ideation were dichotomized as yes or no to questions whether the student has ever had depressed moods or suicidal ideation in the past twelve months. We included two types of school-level variables: the urbanicity of the school's location (metropolitan, urban, and rural) and school type by gender mix (boys', girls', and co-educational).
A two-level, random intercept multilevel regression model was fitted with individuals (level 1) nested within schools (level 2) to estimate the associations of individual determinants and school context simultaneously using MLwiN (development version 2.22). Chow test was applied to detect significant gender differences in terms of slopes and intercepts between the stratified regressions  that were fitted separately to boys and girls. We obtained maximum-likelihood estimates by Iterative Generalised Least Squares (IGLS), and then switched to Markov Chain Monte Carlo (MCMC) function. The MCMC was conducted to burn-in for 500 simulations for starting values of the distribution to discard, and was followed by 5,000 further simulations to get the precise estimate and distribution of interest. Once convergence diagnostics were confirmed, the simulated values and 95% credible intervals (CI) were obtained.
Table 2 shows students' primary and secondary purposes for Internet use aside from academic purposes, according to gender in the middle and high schools. Regardless of the school, boys' primary and secondary purpose of Internet use were online gaming and information searching, respectively. Girls reported blogging and updating a personal homepage, searching for information, and using messengers and chatting as their primary and secondary purposes.
Table 3 presents the result of multilevel regression modeling to predict addictive Internet use among adolescents. Girls were much less likely to be addicted to the Internet than boys. The score of addictive Internet use increased gradually during middle school years, yet they decreased during high school years. Self-rated academic achievement was inversely associated with addictive Internet use. As the parental education level and the FAS increased, the score of addictive Internet use was significantly decreased. Tobacco use was inversely associated with addictive Internet use while alcohol use was not a significant factor. Substance use showed the strongest association with addictive Internet use. All the variables of physical activities showed inverse associations with addictive Internet use. Higher scores of addictive Internet use were associated with higher levels of sleep dissatisfaction. Psychological characteristics such as depressive symptoms and suicidal ideation showed positive associations with addictive Internet use. Regarding school characteristics, girls attending girls' schools were more likely to have addictive Internet use than those attending coeducational schools.
With confirmation of the Chow test [F (17, 57,823) = 163.62, p<0.001], gender stratified analysis revealed different patterns of associations between boys versus girls across all the variables (Table 4). The association of poor self-rated academic achievement with addictive Internet use was stronger in boys than in girls. Parental educational status was inversely associated with addictive Internet use among boys while showing no association among girls. Tobacco and alcohol use showed the opposite associations between boys and girls: 1) a statistically significant association between drinking and addictive Internet use in girls, yet non-significant in boys; 2) a significant association between smoking less and addictive Internet use in boys but not in girls. Boys who reported substance use at the time of survey had much higher risk of addictive Internet use compared to girls. The associations of addictive Internet use with physical activities and psychological characteristics were stronger in boys than girls. With respect to school context variables, girls' schools had a positive association with addictive Internet use; whereas, boys' schools had no association. Urbanicity of school locations showed no correlation with addictive Internet use.
To our knowledge, this is the first study that examined associations of addictive Internet use with individual-level factors and school-level environmental factors using multilevel analysis with a nationally representative sample. Our novel finding is that there were associations between the adolescents' addictive Internet use and school contexts even after controlling for individual-level characteristics: girls in girls' schools were more likely to be addicted to the Internet than those in coeducational schools. Additionally, we found gender differences in addictive Internet use from the gender stratified analysis: 1) lower parental educational attainment was associated only with boys' addictive Internet use, and 2) alcohol use was a risk factor of addictive Internet use for girls only; whereas, smoking is a risk factor for boys only.
First, our hierarchical regression analysis showed that girls in girls' schools were more likely to be addicted to the Internet compared with girls in coeducational schools after controlling for individual-level factors. The contexts of girls' schools may contribute to girls' addictive Internet use with fostering their online networking based on abundant offline same-sex networks within their schools. Korean students in single gender schools seemed to have more same-sex friends than those in coeducational schools because they spend most of their time in school in pursuit of academic excellence, and making opposite gender friends is usually not welcomed by parents concerned about their children's academic achievement . Given that girls have a greater tendency to cherish interpersonal relationships in offline networks and are generally more cautious in creating new relationships online –, they may take advantage of cyberspace to maintain relationships and reinforce their own identities through communicating and sharing information on their common interests via instant messaging, chatting, and visiting friends' personal websites , , . Some girls could also make boyfriends online or offline; however, it might not contribute to Internet addiction as they might want to spend more time face-to-face. Boys in boys' schools also might tend towards Internet addiction based on their relatively abundant offline networks within the schools via online gaming together. However, as shown in the results, school type was not a significant factor for boys' addictive Internet use perhaps because online gaming networks are usually established nationwide or worldwide .
Another novel finding in our study is that parental SES was inversely associated with adolescents' addictive Internet use. Parents of higher education attainment might be able to guide their children toward desirable Internet use and supervise children's Internet use effectively based on their knowledge of the Internet and its devices. Moreover, adolescents whose parents had higher SES might use the Internet less addictively due to their higher self-esteems . Notably, gender stratification showed that a higher parental educational level was only significantly associated a lower score of addictive Internet use in boys (Figure 1-A and 2-A). This could be explained by parents' supervision focused on their boys. Korean parents usually had concerns on their boys' Internet use because they were more accessible and vulnerable to addictive online games and sexual/violent images .
We also found several other variables associated with addictive Internet use among both genders, yet their directions and magnitudes were varied in gender stratification. In the high school grades, the addictive Internet use score was decreased. This is contrasted with past studies that reported no association between age and Internet addiction , . This inconsistency seems to lie in the difference of sampling methods or academic and cultural contexts (Taiwan vs. European countries vs. Korea). Higher pressure for academic achievement of in the Korean society might limit high school students' online networking and/or time spent for online gaming .
Of cigarette smoking and alcohol drinking, our results showed an inverse association of addictive Internet use with smoking and an insignificant association with drinking; however, gender stratification showed complex patterns in the associations of addictive Internet use with drinking and smoking. Drinking and smoking seemed to be complementary for girls' addictive Internet use, whereas smoking might have acted as a substitution for boys. Boys might have fewer opportunities for smoking because they usually played online games at home or Internet café where adolescent smoking is prohibited. In contrast, cyberspace might provide girls more chances to reinforce the drinking and smoking behaviors against a gender-discriminative social atmosphere for women , . Girls might be encouraged to drink and smoke by sharing experiences or information on drinking and smoking with their online peers. Such online interactions may contribute to establishing a favorable norm for smoking and drinking which could lead to offline gatherings in pursuit of drinking or smoking.
Our findings on self-rated academic achievement, physical activities, and psychological status confirm previous studies , , . Self-rated academic achievement was inversely associated with addictive Internet use, yet the association was stronger in boys than girls. The difference might be attributable to unequal pressure for better academic achievement between genders. In a male dominant society, such as in Asian communities with Confucian backgrounds, parental expectations still focus more on boys with the traditional perspective of men as breadwinners, responsible for earning money for their families. As their academic excellence affects later social and economic positions, boys of low academic achievement may be more stressed than their girl counterparts. This societal atmosphere might induce boys to be addicted to the Internet which provides a hideout from reality  or eases their stress with illusory feelings of achievement and self-esteem . The boys addicted to the Internet in this way might waste time for study leading iteratively to poor academic achievement (reverse causality). This study also confirms the past results reporting associations of Internet addiction with depression , suicidal behaviors , lower self-rated sleep satisfaction , and substance use .
Several limitations of this study should be noted. Firstly, this study used cross-sectional data for which causal relations cannot be inferred. Secondly, despite survey administration to guarantee the anonymity of the subject online, adolescents might under-report or over-report in a socially desirable manner. Lastly, respondents were sampled among adolescents who were attending schools. Although it was a nationally representative survey and the rate of entering middle and high school in Korea has been above 99%, selection bias might exist due to excluded adolescents who were out of school, absentees, and exceptional children.
In summary, we found several significant associations of addictive Internet use with individual- and school-level factors and gender differences. Our results suggest that preventing adolescents' addictive Internet use at a population level should take into account gender differences and the association factors of family and school contexts.
Twenty questionnaires of the simplified Korean Internet Addiction Self-assessment Tool (KS scale).
Conceived and designed the experiments: JH JO. Performed the experiments: JH. Analyzed the data: JH JO. Wrote the paper: JH JO YK SS IK.
- 1. International Telecommunication Union (2013) World Telecommunication/ICT Indicators Database 2013 (17th Edition).
- 2. Weinstein A, Lejoyeux M (2010) Internet addiction or excessive internet use. The American journal of drug and alcohol abuse 36: 277–283.
- 3. Young KS (1998) Internet addiction: The emergence of a new clinical disorder. CyberPsychology & Behavior 1: 237–244.
- 4. Thatcher A, Goolam S (2005) Development and psychometric properties of the Problematic Internet Use Questionnaire. South African Journal of Psychology 35: 793.
- 5. Shapira NA, Lessig MC, Goldsmith TD, Szabo ST, Lazoritz M, et al. (2003) Problematic internet use: proposed classification and diagnostic criteria. Depression and Anxiety 17: 207–216.
- 6. Lin SSJ, Tsai CC (2002) Sensation seeking and internet dependence of Taiwanese high school adolescents. Computers in Human Behavior 18: 411–426.
- 7. Lavin M, Marvin K, McLarney A, Nola V, Scott L (1999) Sensation seeking and collegiate vulnerability to Internet dependence. CyberPsychology & Behavior 2: 425–430.
- 8. Morahan-Martin J, Schumacher P (2000) Incidence and correlates of pathological Internet use among college students. Computers in Human Behavior 16: 13–29.
- 9. Durkee T, Kaess M, Carli V, Parzer P, Wasserman C, et al. (2012) Prevalence of pathological internet use among adolescents in Europe: demographic and social factors. Addiction 107: 2210–2222.
- 10. Kandell JJ (1998) Internet addiction on campus: The vulnerability of college students. CyberPsychology & Behavior 1: 11–17.
- 11. American Psychiatric Association (2000) Diagnostic and statistical manual of mental disorders: DSM-IV-TR®: American Psychiatric Pub.
- 12. Caplan SE (2002) Problematic Internet use and psychosocial well-being: development of a theory-based cognitive–behavioral measurement instrument. Computers in human behavior 18: 553–575.
- 13. Widyanto L, Mcmurran M (2004) The psychometric properties of the internet addiction test. CyberPsychology & Behavior 7: 443–450.
- 14. Tao R, Huang X, Wang J, Zhang H, Zhang Y, et al. (2010) Proposed diagnostic criteria for internet addiction. Addiction 105: 556–564.
- 15. Block JJ (2008) Issues for DSM-V: Internet addiction. American Journal of Psychiatry 165: 306.
- 16. Suler J (2004) Computer and cyberspace “addiction”. International Journal of Applied Psychoanalytic Studies 1: 359–362.
- 17. Chou C, Hsiao MC (2000) Internet addiction, usage, gratification, and pleasure experience: the Taiwan college students' case. Computers & Education 35: 65–80.
- 18. Ha JH, Yoo HJ, Cho IH, Chin B, Shin D, et al. (2006) Psychiatric comorbidity assessed in Korean children and adolescents who screen positive for Internet addiction. The Journal of clinical psychiatry 67: 821.
- 19. Kubey RW, Lavin MJ, Barrows JR (2001) Internet use and collegiate academic performance decrements: Early findings. Journal of Communication 51: 366–382.
- 20. Brenner V (1997) Psychology of computer use: XLVII. Parameters of Internet use, abuse and addiction: the first 90 days of the Internet Usage Survey. Psychological reports 80: 879–882.
- 21. Griffiths M (2000) Does Internet and computer “addiction” exist? Some case study evidence. CyberPsychology and Behavior 3: 211–218.
- 22. Flisher C (2010) Getting plugged in: An overview of Internet addiction. Journal of paediatrics and child health 46: 557–559.
- 23. Ko CH, Yen JY, Chen CS, Yeh YC, Yen CF (2009) Predictive values of psychiatric symptoms for Internet addiction in adolescents. Arch Pediatr Adolesc Med 163: 937–943.
- 24. Armstrong L, Phillips JG, Saling LL (2000) Potential determinants of heavier internet usage. International Journal of Human-Computer Studies 53: 537–550.
- 25. Christakis D (2010) Internet addiction: a 21st century epidemic? BMC Medicine 8: 61.
- 26. CNN (2010) Virtually addicted: Weaning Koreans off their wired world. Accessed: 2012.1.20.
- 27. BBC news (2005) S Korean dies after games session. Accessed: 2012.1.20.
- 28. BBC news asia-pacific (2011) Chinese online gamer dies after three-day session. Accessed: 2012.1.20.
- 29. Soule LC, Shell LW, Kleen BA (2003) Exploring Internet addiction: Demographic characteristics and stereotypes of heavy Internet users. Journal of Computer Information Systems 44: 64–73.
- 30. Nalwa K, Anand AP (2003) Internet addiction in students: a cause of concern. CyberPsychology & Behavior 6: 653–656.
- 31. Kaltiala-Heino R, Lintonen T, Rimpela A (2004) Internet addiction? Potentially problematic use of the Internet in a population of 12–18 year-old adolescents. Addiction Research & Theory 12: 89–96.
- 32. Davis RA, Flett GL, Besser A (2002) Validation of a new scale for measuring problematic Internet use: Implications for pre-employment screening. CyberPsychology & Behavior 5: 331–345.
- 33. Scholte EM (1992) Prevention and treatment of juvenile problem behavior: A proposal for a socio-ecological approach. Journal of abnormal child psychology 20: 247–262.
- 34. Sallis JF, Owen N, Fisher EB (2008) Ecological models of health behavior. Health behavior and health education: Theory, research, and practice 4: 465–486.
- 35. Chou C, Condron L, Belland JC (2005) A review of the research on Internet addiction. Educational Psychology Review 17: 363–388.
- 36. Mathy RM, Cooper A (2003) The duration and frequency of Internet use in a nonclinical sample: Suicidality, behavioral problems, and treatment histories. Psychotherapy: Theory, Research, Practice, Training 40: 125.
- 37. Soteriades ES, DiFranza JR (2003) Parent's socioeconomic status, adolescents' disposable income, and adolescents' smoking status in Massachusetts. American Journal of Public Health 93: 1155–1160.
- 38. Fawzy FI, Coombs RH, Simon JM, Bowman-Terrell M (1987) Family composition, socioeconomic status, and adolescent substance use. Addictive behaviors 12: 79–83.
- 39. Garnefski N, Okma S (1996) Addiction-risk and aggressive/criminal behaviour in adolescence: Influence of family, school and peers. Journal of adolescence 19: 503–512.
- 40. Greenfield DN (1999) Psychological characteristics of compulsive Internet use: A preliminary analysis. CyberPsychology & Behavior 2: 403–412.
- 41. Lin MP, Ko HC, Wu JYW (2008) The role of positive/negative outcome expectancy and refusal self-efficacy of Internet use on Internet addiction among college students in Taiwan. CyberPsychology & Behavior 11: 451–457.
- 42. National Information Society Agency (2011) Internet Addiction Survey 2010. In: Agency NIS, editor. Seoul, South Korea.
- 43. Korea Statistical Information Service (2013) Statistics on Internet use.
- 44. Kim D, Jung Y, Lee E, Kim D, Cho Y (2008) Development of Internet Addiction Proneness Scale-Short Form (KS scale). The Korea Journal of Counseling 9: 1703–1722.
- 45. Hawkins M (2012) South Korea introduces yet another law to curb gaming's ills. NBC News.
- 46. Currie C, Gabhainn SN, Godeau E, Roberts C, Smith R, et al.. (2008) Inequalities in young people's health: Health Behaviour in School-aged Children (HBSC) international report from the 2005/2006.
- 47. Chow GC (1960) Tests of equality between sets of coefficients in two linear regressions. Econometrica: Journal of the Econometric Society: 591–605.
- 48. Kim H, Kim E, Min K, Shin J, Lee S, et al.. (2007) International Conference on Socialization in Adolescence III on the relationship of parents-children, teachers-students, and among peers In: National Youth Policy Institute, editor. International Conference on Socialization in Adolescence.
- 49. Jones S (2002) The Internet Goes to College: How Students Are Living in the Future with Today.
- 50. Gross EF (2004) Adolescent Internet use: What we expect, what teens report. Journal of Applied Developmental Psychology 25: 633–649.
- 51. The Korean National Information Society Agency (2012) The survey on Internet Addiction 2011. Seoul, South Korea: The Korean Ministry of Public Administration.
- 52. Ng BD, Wiemer-Hastings P (2005) Addiction to the internet and online gaming. CyberPsychology & Behavior 8: 110–113.
- 53. Rosenberg M (1989) Society and the adolescent self-image (rev: Wesleyan University Press.
- 54. Ko CH, Yen JY, Chen CC, Chen SH, Yen CF (2005) Gender differences and related factors affecting online gaming addiction among Taiwanese adolescents. The Journal of Nervous and Mental Disease 193: 273.
- 55. Kim K, Ryu E, Chon MY, Yeun EJ, Choi SY, et al. (2006) Internet addiction in Korean adolescents and its relation to depression and suicidal ideation: A questionnaire survey. International journal of nursing studies 43: 185–192.
- 56. Ko CH, Yen JY, Chen CC, Chen SH, KUANYI W, et al. (2006) Tridimensional personality of adolescents with internet addiction and substance use experience. Canadian Journal of Psychiatry 51: 887–894.