Internet use and academic achievement: gender differences in early adolescence.

Chen, Su-Yen ; Fu, Yang-Chih

As a new medium of learning in the twenty-first century, the Internet has brought unprecedented opportunities to students. To capitalize upon such opportunities, schools and families eagerly facilitate Internet use, particularly in East Asia, where academic achievement remains the top priority at school. At the same time, however, the Internet has also become a major concern for parents, because some online activities may seriously distract adolescents from their homework. For parents and educators alike, therefore, it is important to determine whether and how Internet use is linked to academic achievement, a key outcome of school learning.

Previous studies have been inconclusive about the relation between Internet use and academic achievement. Among high school students, for example, the amount of time using the Internet has little to do with individuals' academic achievement. Furthermore, students' grade point averages (GPA) are not closely correlated with specific activities, such as searching for information, E-mailing, and playing games (Hunley et al., 2005). Among college students, however, searching information online about course materials helps boost intellectual development and facilitates preparation for future jobs. In contrast, heavily indulging in online recreation has been closely linked to impaired academic performance (Kubey, Lavin, & Barrows, 2001; Kuh & Hu, 2001).

Internet use varies greatly by what students do online and how they do it. Like many other domains in adolescente, the content and other patterns of Internet use also differ widely between boys and girls. Does such a gender gap account for the lack of consistent findings about how these activities are linked with academic achievement? For example, does any one specific online activity help boost academic achievement for boys and girls alike? Or is it possible that boys benefit from one activity while girls gain from another?

Following consistent findings about the gender gap in Internet use, this paper examines whether and how male and female adolescents differ in the ways various aspects of Internet use affect academic achievement. These aspects include the overall frequency of using the Internet, activities students engage in online (such as information seeking, chatting and socializing with friends, and playing games), the location where they use the Internet, and whether parents regulate such use. Data were drawn from the Taiwan Youth Project, a panel survey series that has followed 2,690 youths from grade 7 (age 13) since 2001.

BACKGROUND

Internet Use and Academic Achievement

Some studies have suggested a positive association between college students' Internet use and their learning. In Suhail and Bargee's (2006) survey study with 200 university students from Pakistan, around three quarters of respondents noted positive effects of Internet use on their learning in at least three aspects. First, Internet use improved their grades. Second, the Internet expanded their reading, writing, and information-processing skills. Third, the Internet has proved a helpful tool in their studies. In another study, Kuh and Hu (2001) used data (collected with the College Student Experiences Questionnaire) from 71 four-year colleges and universities in the United States (N = 18,344) and found that surfing the Internet for course material had positive net effects on intellectual development and vocational preparation, in addition to personal development.

Other studies have found a negative link between college students' Internet use and academic performance. For example, non-heavy Internet users had higher academic grades than heavy Internet users as a group (Chen & Peng, 2008). In another study, at a large public university in the United States (N = 572) significantly more students believed that their academic performance had been impaired when they were involved in heavy recreational Internet use, defined as usage of synchronous, computer-mediated communication (CMC), such as multi-user domains (MUDs) and Internet Relay Chat (IRC) (Kubey, Lavin, & Barrows, 2001).

Although Internet use has been equally popular among high school students, relatively fewer studies have explored how Internet use is linked to academic achievement among these adolescents. One rare study (Hunley et al., 2005) recruited 10th-grade students from science and social studies classes at three public high schools in Ohio and asked them to keep a log of their computer use for one full week. Using GPA as the indicator, the study found no significant relation between academic achievement and the amount of time spent on the Internet. Nor did such achievement have any noticeable association with such online activities as searching for information, playing games, or Emailing.

In view of the lack of consistent evidence supporting an association between Internet use and academic achievement, it appears that the Internet may not play a major role in adolescents' learning. As suggested earlier, however, previous studies may have overlooked gender differences in Internet use among adolescents. Taking such a key factor into account, the current study aims to compare and contrast what aspects of Internet use affect the school learning among boys and girls, respectively.

Gender Differences in Internet Use

Studies have suggested that even though the gender gap in computer use is closing among adolescents, boys and girls still differ greatly in what they do online (Clemente, 1998; Imhof, Vollmeyer, & Beierlein, 2007; Odell, Korgen, Schumacher, & Delucchi, 2000). Whereas more female adolescents use the Internet to search for information (Chen & Peng, 2008; Lin & Yu, 2008; Odell et al., 2000) and for E-mail (Chen & Peng, 2008; Lin & Yu, 2008; Odell et al., 2000; Sherman et al., 2000), more male adolescents use the Internet to play games (Chen & Peng, 2008; Griffiths, Davies, & Chappell, 2004; Lin & Yu, 2008; Odell et al., 2000; Sherman et al., 2000).

Such gender differences prevail from elementary school through college in some societies. Ina study of 5th and 6th graders in Taiwan, for example, Lin and Yu (2008) found that boys tended to spend a little more time than girls in terms of weekly use of the Internet. They also differed significantly in their top three online activities: the percentages of time girls spent searching for homework information and using e-mail were higher than those of boys; in" contrast, boys played games more often than did girls.

The patterns remain about the same among college students, at least in the United States and Taiwan. For example, Odell et al. (2000) surveyed American college students from five states and found that more female than male students used the Internet for E-mailing and research, while more male students played online games. Sherman et al. (2000) also investigated the Internet gender gap among American college students by comparing the usage patterns of three student cohorts in 1997, 1998, and 1999. Male college students participated more in WWW surfing, newsgroups, MUDs (multi-user, real-time virtual world online gaming), and chat groups, while female students reported significantly higher E-mail use. Based on a large national survey in Taiwan, Chen and Peng (2008) also found that whereas males spent more time playing online games than did females, females spent more time searching for academic information, as well as making friends and chatting.

Male and female adolescents also differ markedly in terms of where they access the Internet. As revealed in study after study, boys visit Internet Cafes more often than their female counterparts, who use the Internet mostly at home and at school (Hsu & Chuang, 2008; Lin & Yu, 2008; Wu & Cheng, 2007). Internet Cafes may indeed provide a convenient environment and fast access to the Internet so that customers can concentrate on their work without interference from others (Wu & Cheng, 2007). Such a setting, however, has also become a place for adolescents to indulge in online games. While Internet Cafes are seen as a masculine gaming space and are thus considered highly gendered (Wu & Cheng, 2007; Hsu & Chuang, 2008), parents and teachers may become concerned that those who overly indulge in the Internet, especially boys, will tend to lag behind academically. As found in a large survey, high school students in Taiwan who spent more time playing online games had lower academic achievement in later school years (Chen & Lu, in press). Although the association is only marginally significant, it raises an important issue as to the role of gender differences in understanding how Internet use is correlated with academic achievement.

HYPOTHESES

Previous studies have shown some associations between Internet use and academic achievement, while gender differences in online activities and the location of use of the Internet are well documented. By taking into account gender differences in patterns of Internet use, the current study aims to reexamine the association between Internet use and academic achievement. Based on the above review, the following hypotheses were formulated.

Hypothesis 1. Adolescents' academic achievement partly depends on their activities on the Internet earlier in their school years.

1a. Academic achievement will be higher if they use the Internet more often to search for information.

1b. Achievement will be lower if they use the Internet to socialize.

1c. Achievement will be lower if they use the Internet to play games.

Hypothesis 2. Academic achievement will be impaired if adolescents use the Internet Cafe more often for online activities.

Hypothesis 3. The association between Internet use and academic achievement differs between male and female adolescents.

METHOD

Sample and Data Collection

Data were taken from the Taiwan Youth Project (TYP), a panel study based at the Academia Sinica in Taiwan. The project was started in the year 2000 and has conducted 8 waves of interviews as of the end of 2008. Students were sampled from middle schools (ages 13-15) located in the northern part of Taiwan: Taipei City, Taipei County, and Yi-Lan County, using the multi-stage, stratified cluster sampling method. For the first wave of the survey, 40 middle schools were randomly selected. In each school, two classes of 7th graders were chosen at random. All students in these classes, one of each of their parents, and their homeroom teachers were asked to complete self-administered questionnaires.

The initial successful samples included 2,690 students. The respondents were re-interviewed each year afterwards, with 2,683, 2,663, and 2,354 students retained in the second, third, and fourth waves of the surveys, respectively. For the purpose of this study, data were drawn from the second wave of the student survey when the student respondents were in the 8th grade. Only the dependent variables, the standardized test score of the high school entrance examination, was taken from the data collected in the third wave. This test score measures students' academic achievement at the end of 9th grade.

Since many students did not report their scores, the number of valid cases dropped to 1,409. We conducted t-tests and Chi-square analyses to check for the differences in any aspect of Internet use between the respondents that provided their test score and those that did not. The only difference that turned out to be significant was the use of Internet Cafes: more students who did not provide their test scores visited Internet Cafes than those who did provide their scores (p < .05). Although the remaining valid cases are somewhat biased toward the non-Internet Cafe users, we expect the findings to be useful for the understanding of how Internet use is associated with academic achievement among most adolescents.

Measures

Dependent variable. As noted, the dependent variable--the self-reported test score in the high school entrance exam--measures students' academic achievement at the end of the 9th grade as the major outcome. The valid scores ranged from 30 to 289 in the sample, with a mean score of 168.1 (see Table 1).

Independent variables. All independent variables were taken from the students' survey in the 8th grade. For overall frequency of Internet use, respondents reported how often they used the Internet: "at least once a day," "two or three times a week," "once a week," "seldom," or "never." The answers were recorded in reverse order so that a higher number indicated more frequent use. For online activities, respondents were asked what activities they usually engaged in on the Internet. They were allowed to choose more than one activity, from: searching for information, chatting and socializing with friends, playing games, and other activities. Each activity was coded as a dummy variable. The first three activities attracted about the same number of students, ranging from 40% to 47%. The location where one gained access to the Internet most often was divided into three categories: at home, at school, at the Internet Cafes, in addition to "others" (e.g., at cram schools, at someone else's homes). Only the Internet Cafe (about 11%) was used as a dummy variable in the analysis. Finally, respondents were asked whether their parents set rules about certain aspects of their life, including the amount of time spent on Internet use or on computer games. Over 60% of the students had such parental regulation.

Control variables. To take prior academic achievement into account, we controlled for the respondents' class ranking in the 8th grade (i.e., 1 = ranked in the last part of the class, to 2 = the latter part, 3 = ranked between 16-25, 4 = 6-15, or 5 = ranked top 1-5). Since the class size ranged between 30 and 40 in nearly all sampled classes, the ordinal categories should be close to a universal measure for prior achievement. Other controls included gender (male = 1) and father's and mother's education (i.e., 1 = elementary and lower, 2 = middle school, 3 = high school and higher).

Data Analyses

After describing the patterns of Internet use among Taiwanese male and female 8th graders, we first checked gender differences with Chi-square analyses. Then Pearson correlation was used to identify the intercorrelations among the variables. Finally, and most importantly, we performed regression analyses to examine how respondents' academic achievement in the 9th grade varied on the patterns of Internet use in the 8th grade, while controlling for background variables and academic achievement in the 8th grade. To verify if any of such variations differed between boys and girls, the regression analyses were repeated after the sample was split by gender.

RESULTS AND DISCUSSION

Most of the respondents used the Internet two or three times a week. Compared to females, males were online more frequently (see Table 2). In terms of averages, males and females differed markedly in every aspect of Internet use. For example, more than half the female respondents used the Internet to search for information, as well as to chat and socialize. Only about one fourth indicated they used it to play online games. By contrast, playing games was the most popular Internet activity for male respondents, with 66% reporting that they usually used the Internet for games, followed by searching for information (43%), and chatting and socializing (25%). Gender differences were apparent, as indicated by Chi-square analyses (see Table 2): More female students searched for information and chatted and socialized with friends than did their male counterparts; while more males noted playing games as an activity they usually engaged in on the Internet. Likewise, a significantly greater percentage of male respondents reported that they mostly used the Internet in Internet Cafes. Regarding parental regulation, more male students said their parents had established rules about how much time they could spend on the Internet or playing computer games.

Table 3 presents the intercorrelations among variables. All variables related to the patterns of Internet use in the 8th grade were significantly correlated with academic achievement in the 9th grade. While overall frequency of Internet use, Internet use for searching for information, and parental regulation of time spent on Internet use were positively correlated with later academic achievement, chatting and socializing, playing games online, and going to Internet Cafes were negatively correlated.

Some of these associations remained significant after background variables were taken into account. All of these background variables, gender, father's and mother's education, and prior academic achievement (in the 8th grade) helped explain academic achievement in the 9th grade, accounting for 58.5% of the total variance (Model 1, Table 4). As one would expect, academic achievement in the 8th grade had the greatest impact on subsequent academic achievement.

When all the background variables were held constant, the frequency of Internet use alone turned out to be a nonsignificant factor in understanding how well a student performed on the high school entrance exam. What they did on the Internet, however, remained critical to how they performed academically. For example, those who used the Internet to search for information outscored those who did not by 11.28 points (p < .001, Model 2). By contrast, students who used the Internet for chatting and socializing underperformed by an average of 6.32 points. Those who played online games also scored 6.35 lower, on average (p < .01, Model 2). Thus, regardless of gender, parents' education, and how well they were doing academically a year ago, what the adolescents did on the Internet continued to clearly distinguish who scored better on the high school entrance exam. The findings confirm Hypotheses 1a, 1b, and 1c. Spending time on the Internet per se had no definite implication for students' academic achievement, but the types of online activities indeed played a key role.

The same negative effect also lingered from the use of Internet Cafes. Even among students who shared a similar background and engaged in the same Internet activities, those who went to Internet Cafes consistently performed more poorly on the entrance exam, lowering the score by 7.39 points, on average (p < .05, Model 3, Table 4). Such an effect remained significant even after we also controlled for parental regulation (a factor that helped raise the score itself). Therefore, even when these adolescents were identical or similar in background factors, prior academic achievement, the frequency and the activities they did online, and under similar parental regulation, going to Internet Cafes alone weakened their academic performance. Hypotheses 2 is thus confirmed.

Although the combined contribution of Internet use was relatively small, given the rigorous statistical controls we employed and the prospective longitudinal design of the study (which involved a one-year gap between independent variables and the dependent variable), it is impressive that their regression coefficients reached the level of significance at p < .05 or p < .001. Consistent with the zero-order correlation, Internet use for searching for information was positively linked with later academic achievement, whereas chatting and socializing with friends, playing games, and the use of Internet Cafes were negatively linked to that achievement.

Previous studies have suggested that male and female adolescents differ markedly in what they do on the Internet and where they go online. To disentangle any gender differences in the association between Internet use and academic achievement, therefore, we split the full sample by gender and proceeded with further analyses within each group. As expected, the effects from the background variables remained very important and substantial in both groups. How well boys and girls did on the high school entrance exam was definitely affected by both parents' education and the students' class ranking in the 8th grade (Table 5), consistent with findings in Table 4. The effects of Internet use, however, differed somewhat between males and females.

For boys, searching for information online proved to be a positive and important factor in the exam score (p < .001, Models 1 & 2, Table 5). Among the three other factors pertaining to Internet use that resulted in negative impacts on the full sample, however, only playing online games had a significant and negative effect among boys (Model 1). Part of such an effect was determined by where the male adolescents played the games and the extent of parent regulation (Model 2). That is, although that negative effect lost part of its significance after Internet Cafe and parental regulation were added into the model (Model 2), playing online games remained the only factor in using the Internet that hindered academic achievement. Within the male subsample, furthermore, parental regulation turned out to be another factor that helped students score better on the entrance exam, an effect that only emerged after splitting the full sample.

The female subsample revealed a somewhat different pattern as to how Internet use was linked to academic achievement. Like boys, girls who went online to search for information also scored significantly better than those who did not (p < .001, Model 3). Unlike their male counterparts, however, female adolescents did not score worse if they used the Internet to play games. Rather, poor performance on the exam was significantly linked to earlier Internet use for chatting and socializing (p < .05, Model 3). Also unlike boys, parental regulation did not help girls score better (Model 4).

Therefore, not only did male and female adolescents differ in the kinds of activities they mostly engaged in on the Internet and where they went online, but the ways such Internet use linked to later academic achievement also varied. While both boys and girls gained from using the Internet as a main source of information, only boys suffered by going online for playing games, and only girls scored poorly if they used the Internet mainly for social purposes. Thus, the previous findings with the full sample are partly correct, and the current findings with split subsamples partly confirm Hypothesis 3. These findings call for modifications along the gender line: The positive effects of Internet use on academic achievement apply to boys and girls alike, but one needs to stipulate the negative effects more discriminately in terms of gender. Using academic performance as the yardstick, an overuse of the Internet for social purposes makes girls particularly vulnerable whereas indulging in online games is especially harmful to boys.

CONCLUSION

Gender differences in online activities are substantial among Taiwanese adolescents, a finding consistent with a study on Taiwanese 5th and 6th graders (Lin & Yu 2008) and with Western studies on high school and college students (Chen & Peng, 2008; Griffiths, Davies, & Chappell, 2004; Odetl et al., 2000; Sherman et al., 2000). On average, male students use the Internet more frequently than do female students. They use the Internet for recreational purposes (e.g., online games) more often than their female counterparts. In contrast, female students use the Internet more to search for information and to chat and socialize with friends. Boys also visit Internet Cafe more often to gain access to the Internet and have more parental regulation of their Internet use.

Patterns of Internet use are closely linked to academic achievement later in middle school. Although the frequency of using the Internet in the 8th grade is not a key factor in distinguishing who scores better on the high school entrance exam, what students do online clearly distinguishes such academic performance. While using the Internet to search for information is positively linked with later academic performance, Internet use for recreational and social purposes exerts a negative impact on academic achievement. These results are generally consistent with prior studies of American college students (Kubey, Lavin, & Barrows, 2001; Kuh & Hu, 2001). In addition, using Internet Cafes as the location for accessing the Internet exerts a negative effect on later academic achievement.

Most importantly, according to further regression analyses, males and females differ not only in their patterns of Internet use, but in how these patterns affect their academic performance. For female students, the most popular online activity in the 8th grade is chatting and socializing. The more time they spend on this activity, the lower their scores on the high school entrance exam a year later, an effect absent among boys. By contrast, it is gaming, the most popular online activity for males, that significantly lowers boys' scores, but not girls'. Furthermore, Internet Cafes and parental regulation of Internet use partly explain why gaming lowers male students' later test scores, a further modification unique to boys. As a recent study found in the United Kingdom, boys who play computer games often are more likely than girls to quarrel with their parents when facing parental regulation (Livingstone (2007). While parental regulation helps boost performance on entrance exams, playing online games alone hurts scores, even after taking into account such regulation and other aspects of Internet use.

Thus, among students who share similar background characteristics and are at the same academic level in 8th grade, what they do on the Internet, rather than how often they go online, has important implications for how well they will achieve on one of the most important exams of their lives. During such a process, not only does gender make a difference in the patterns of Internet use, but it also plays a key role in differentiating what kinds of online activities help or hinder students' academic achievement in middle school. In studying how the Internet affects learning or how well students perform in early adolescence, then, gender remains a critical factor that deserves further examination. When more cross-national or cross-cultural data become available, it would be even more fruitful to examine the linkage between Internet use and academic achievement. Such a comparative perspective would further identify the extent to which the current findings can be applied in various social and cultural contexts.

REFERENCES

Chen, S. Y., & Lu, L. (in press). After-school time use in Taiwan: Its effects on educational achievement and well-being. Adolescence.

Chen, Y. F., & Peng, S. S. (2008). University students' Internet use and its relationships with academic performance, interpersonal relationships, psychosocial adjustment, and self-evaluation. CyberPsychology & Behavior, 11, 467-469.

Clemente, P. C. (1998). State of the Net: The new frontier. New York: McGraw-Hill.

Griffiths, M. D., Davies, M. N. O., & Chappell, D. (2004). Online computer gaming: A comparison of adolescent and adult gamers. Journal of Adolescence, 27, 87-96.

Hsu, Y. C., & Chuang, O. L. (2008). Are Internet Cafe gendered spaces? CyberPsychology & Behavior, 11(2), 224-226.

Hunley, S. A., Evans, J. H., Delgado-Hachey, M., Krise, J., Rich, T., & Schell, C. (2005). Adolescent computer use and academic achievement. Adolescence, 40(158), 307-318.

Imhof, M., Vollmeyer, R., & Beierlein, C. (2007). Computer use and the gender gap: The issue of access, use, motivation, and performance. Computers in Human Behavior, 23 (2823-2837.

Kubey, R. W., Lavin, M. J., & Barrows, J. R. (2001). Internet use and collegiate academic performance decrements: Early findings. Journal of Communication, 51, 366-382.

Kuh, G. D., & Hu, S. (2001). The relationships between computer and information technology use, selected learning and personal development outcomes, and other college experiences. Journal of College Student Development, 42, 217-232.

Lin, C. H., & Yu, S. F. (2008). Adolescent Internet usage in Taiwan: Exploring gender differences. Adolescence, 43, 317-331.

Livingstone, S. (2007). Strategies of parental regulation in the media-rich home. Computers in Human Behavior, 23, 920-941.

Odell, P. M., Korgen, K. O., Schumacher, P., & Delucchi, M. (2000). Internet use among female and male college students. CyberPsychology & Behavior, 3, 855-862.

Sherman, R. C., End, C., Kraan, E., Cole, A., Campbell, J., Birchmeier, Z., & Kluasner, J. (2000). The Internet gender gap among college students: Forgotten but not gone? Cyber Psychology & Behavior, 3, 885-894.

Suhail, K., & Bargees, Z. (2006). Effects of excessive Internet use on undergraduate students in Pakistan. CyberPsychology & Behavior, 9(3), 297-307.

Wu, C. H., & Chen, F. F. (2007). Internet Cafe addiction of Taiwanese adolescents. CyberPsychology & Behavior, 10(2), 220-225.

This article drew upon data from the Taiwan Youth Project (TYP), a panel study funded by the Academia Sinica as a thematic program from 2004 to 2007 (grant number AS-93-TP-C01, http://www.typ.sinica.edu.tw/). Part of the research framework of this paper was also taken from the first author's research project sponsored by the National Science Council, Taiwan (grant number NSC 98-2410-H-007-004-MY2).

Fu, Yang-Chih, Ph.D., Institute of Sociology, Academia Sinica, Taiwan

Requests for reprints should be sent to Su-Yen Chen, Ph.D. Center for Teacher Education, National Tsing Hua University, 221 Education Building, 101 Sec. 2, Kuang-Fu Rd., Hsinchu, TAIWAN, 30013. E-mail: [email protected]

Table 1 Summary of Variables

Variables Means S.D. Min. Max.

Independent variables

Male 0.506 0.500 0 1
Father's educational level
 (1=elementary or lower,
2=middle school, 3 = high 1.754 0.718 1 3
 school or higher) Mother's
 educational level (1 =
 elementary or lower,
2 = middle school, 3 = high 1.630 0.659 1 3
 school or higher)
Frequency of Internet use 2.166 1.255 0 4
Internet use for searching 0.475 0.500 0 1
 information
Internet use for chatting 0.403 0.491 0 1
 and socializing with friends
Internet use for playing games 0.464 0.499 0 1
Interest use in Interest Cafe 0.112 0.316 0 1
Parental regulation on 0.604 0.489 1 1
 Interest use
8th academic achievement 3.540 1.045 1 5
 (ranking in the class)
Dependent variable
9th grade academic achievement 168.1 53.5 30 289
 (test scores of high-
 school entrance exam)

Table 2 Gender Differences in Patterns of Internet Use

 Male Female
 Mean (SD) Mean (SD)

Frequency of Internet use 2.26 (1.276) 2.07 (1.226)
Internet use for searching
 information 0.43 (0.496) 0.52 (0.500)
Internet use for chatting and
 socializing with friends 0.25 (0.436) 0.56 (0.497)
Internet use for playing games 0.66 (0.475) 0.27 (0.442)
Internet use in Internet Cafe 0.16 (0.364) 0.07 (0.256)
Parental regulation on
 Internet use 0.70 (0.459) 0.51 (0.500)

 [chi square]

Frequency of Internet use 17.138 *
Internet use for searching
 information 8.332 *
Internet use for chatting and
 socializing with friends 115.978 ***
Internet use for playing games 189.333 ***
Internet use in Internet Cafe 25.777 ***
Parental regulation on
 Internet use 53.798 ***

* p <.05, ** p <.01 *** p < .001

Table 3 Inter-correlations Among Variables

 1 2 3 4

1. Gender 1.000
2. Father's
 education 0.033 1.000
3. Mother's
 education 0.010 0.622 *** 1.000
4. Prior
 achievement -0.100 *** 0.106 *** 0.141 *** 1.000
5. Freq. of
 Internet use 0.079 * 0.148 *** 0.152 *** 0.128 ***
6. Search for
 info. -0.082 * 0.079 * 0.138 *** 0.179 ***
7. Chat and
 socialize -0.307 *** -0.037 -0.007 -0.019
8. Play
 games 0.392 *** -0.017 -0.049 -0.115 ***
9. Internet
 Cafe 0.137 *** -0.066 * -0.091 ** -0.062 *
10. Parental
 regulation 0.197 *** 0.080 * 0.071 * -0.007
11. Academic
 achievement -0.031 0.331 *** 0.345 *** 0.719 ***

 5 6 7 8

1. Gender
2. Father's
 education
3. Mother's
 education
4. Prior
 achievement
5. Freq. of
 Internet use 1.000
6. Search for
 info. 0.035 1.000
7. Chat and
 socialize 0.149 *** -0.022 1.000
8. Play
 games 0.118 *** -0.167 *** -0.057 * 1.000
9. Internet
 Cafe 0.027 -0.131 *** 0.041 0.218 ***
10. Parental
 regulation 0.118 *** 0.009 -0.016 0.075 *
11. Academic
 achievement 0.151 *** 0.250 *** -0.086 * -0.134 ***

 9 10

l. Gender
2. Father's
 education
3. Mother's
 education
4. Prior
 achievement
5. Freq. of
 Internet use
6. Search for
 info.
7. Chat and
 socialize
8. Play
 games
9. Internet
 Cafe 1.000
10. Parental
 regulation 0.026 1.000
11. Academic
 achievement -0.108 *** 0.069 *

* p <.05 ** p < .01 *** p < .001

Table 4 Regression Analyses of Academic Achievement
in the 9th Grade

 Model 1 Model 2

Male 3.59 (1.91) 5.03 (2.30) *
Father's education 11.87 (1.69) *** 11.45 (1.75) ***
Mother's education 12.08 (1.84) *** 11.17 (1.91) ***
Prior academic 35.03 (0.94) *** 33.80 (1.02) ***
 achievement
Frequency of 0.04 (0.97)
 Internet use
Use the Internet to:
 search for information 11.28 (2.06) ***
 chat and socialize -6.32 (2.17) *
 play games -6.35 (2.22) *
Go to Internet Cafe
Parental regulation

Constant 2.23 (4.24) 9.04 (5.10)

N 1283 1142
R-square 0.587 0.592
Adjusted R-square 0.585 0.590
F-value 453.66 *** 205.90 ***

 Model 3

Male 4.67 (2.35) *
Father's education 11.30 (1.75) ***
Mother's education 10.78 (1.91) ***
Prior academic 33.80 (1.02) ***
 achievement
Frequency of -0.22 (0.97)
 Internet use
Use the Internet to:
 search for information 10.77 (2.06) ***
 chat and socialize -6.05 (2.18) *
 play games -5.49 (2.24) *
Go to Internet Cafe -7.39 (3.10) *
Parental regulation 4.18 (2.10) *

Constant 8.81 (5.22)

N 1142
R-square 0.596
Adjusted R-square 0.592
F-value 166.82 ***

* p <.05 ** p <.01 *** p <.001

Table 5 Regression Analyses of
Academic Achievement in the 9th Grade by Gender

 Males

 Model 1 Model 2

Father's 10.75 (2.40) *** 10.21 (2.38) ***
education

Mother's 12.16 (2.59) *** 12.25 (2.58) ***
education

Prior academic 34.15 (139) *** 34.10 (137) ***
achievement

Freq. of 0.76 (1.34) 0.58 (1.34)
Internet use

Use the
Internet to:

 search 10.90 (2.87) *** 10.33 (2.85) ***
 for info

 chat and -3.00 (3.24) -2.89 (3.21)
 socialize

 play -10.51 (3.00) *** -8.99 (3.11) *
 games

Go to -6.86 (3.77)
Internet
cafe

Parental 8.68 (3.06) *
regulation

Constant 12.67 (6.87) 8.19 (7.17)

N 576 576

R-square 0.632 0.639

Adjusted 0.628 0.634
R-square

F-value 139.41 *** 111.51 ***

 Females

 Model 1 Model 2

Father's 1234 (2.57) *** 12.31 (2.57) ***
education

Mother's 10.02 (2.83) *** 9.73 (2.86) **
education

Prior academic 32.97 (1.52) *** 32.96 (1.53) ***
achievement

Freq. Of -0.80 (1.41) -0.99 (1.42)
Internet use

Use the
Internet to:

 search 10.88 (2.97) *** 10.53 (2.99) ***
 for info

 chat and -9.31 (2.96) * -8.89 (3.04) *
 socialize

 play -1.72 (3.32) -1.83 (3.32)
 games

Go to -5.13 (5.56)
Internet
cafe

Parental 1.14 (2.94)
regulation

Constant 15.08 (7.56) * 15.85 (7.77) *

N 566 566

R-square 0.554 0.554

Adjusted 0.548 0.547
R-square

F-value 198.82 *** 176.84 ***

* P <.05 ** p <.01 *** p <.001