The Relation of Internet Use to Prosocial and Antisocial Behavior in Chinese Adolescents

Abstract

Prosocial and antisocial Internet use is investigated in this study by constructing an objective measure called the Adolescent Internet Use Questionnaire (AIUQ). The sample consists of 509 secondary school students in Hong Kong. Based on the previous review of the literature, the present study tests the following two hypotheses: (a) sex difference hypothesis: boys are more delinquent in Internet use than are girls; (b) positive association hypothesis: there is a positive association between Internet behavior and daily social behavior. In other words, positive Internet behavior is positively associated with positive daily social behavior, and negative Internet behavior is positively associated with negative daily social behavior. In general, the two hypotheses are supported by the data in this study. The internal consistency reliability and predictive validity of the major indices of the AIUQ are also substantiated by the present data.

Introduction

Problematic Internet use or pathological Internet use (PIU) or Internet addiction is quite prevalent among secondary school and university students.^1,2 Research indicates that 4% to 8.1% of university Internet users show excessive or pathological use.^3,4 If we assume that 6.05% (an estimated average of the above two studies) of undergraduate students at a medium-sized university are problematic Internet users, then a student population of 10,000 would have 605 problematic Internet users. This figure is quite shocking, since the problem is very imminent and serious, bearing in mind that every university student is an asset to society. The situation is also serious in secondary schools. For example, Lin and Tsai reported that 11.8% of senior secondary school students in their study were regarded as Internet dependent.² These Internet dependents consisted of more boys than girls and also more 12th graders, who, on average, spent an average of 17.57 hours per week online. The situation in Hong Kong is even worse, according to a survey taken from 500 10- to 29-year-old people conducted by the Hong Kong Youth Association, where the average Internet usage among this group is 18.4 hours per week. What is even more shocking is that about 10% of the sample spends 42 hours or more online per week, that is, 6 hours each day.⁵

While the mass media often describes Internet addiction as an emerging psychological disorder, there is still disagreement among researchers over whether it is a real condition.⁶ Some researchers regard Internet addiction as a new type of psychiatric disorder similar to impulse control disorder, or a kind of behavioral addiction.^7,8 Beard and Wolf attempt to propose diagnostic criteria for Internet addiction based on DSM-IV criteria for pathological gambling.⁷ For example, the first two criteria for Internet addiction are as follows: a typical user “1. is preoccupied with the Internet (think about previous online activity or anticipate next online session)” and “2. needs to use the Internet with the increased amount of time in order to achieve satisfaction” (p. 381).⁹ Since little research has been done in this area, the clinical aspect of Internet addiction requires further study to find out the actual situation.¹⁰

Further, Davis and Caplan argue that there are two types of PIU:^11,12 (a) specific PIU, where people are dependent on a specific function of the Internet (e.g., overuse of online sexual materials, online auction services, online gambling, etc.), and (b) generalized PIU, indicating a general, multidimensional overuse of the Internet. Caplan argued that the main symptoms of PIU are “obsessive thoughts about the Internet, diminished impulsive control, inability to cease Internet usage, and importantly, feeling that the Internet is an individual's only friend” (p. 193).¹² Additionally, research has also discovered that: (a) pathological users are more likely to be male;^3,4,13 (b) PIU is positively related to depression, loneliness, shyness, and social isolation;^12,14 (c) excessive use of the Internet is associated with lower self-esteem;¹⁵ (d) for pathological users, the influence of the Internet on daily routines, school performance, and parental relation is negative;^2,16 (e) PIU is negatively related to emotional competence and social skills;¹⁷ and (f ) excessive use of the Internet is also associated with decreased perceived social support.¹⁸

Prosocial and antisocial behavior

Current research on child and adolescent behavior tends to emphasize either prosocial or antisocial behavior, but seldom are both kinds of behavior investigated in a single study.^19–25 To fill this research gap, the present study investigates both the prosocial and antisocial daily social behavior of Chinese adolescents in one single setting,^26,27 as well as the prosocial and antisocial or delinquent Internet behavior in one single study.

Prosocial behavior in children and adolescents usually involves “sharing, cooperating, helping, feeling empathy and caring for others” (Radke-Yarrow et al., p. 528).²⁰ In addition, altruistic behavior with some personal sacrifice and normative behavior (e.g., good boy and nice girl) can also be regarded as prosocial behavior. In this article, prosocial behavior refers to altruistic, socially acceptable, or normative behavior (e.g., giving gifts to parents, helping people with disabilities to cross the road, apologizing to others after committing a misdeed, and serving as a volunteer worker).

According to Hindelang et al. (pp. 54–5), the major antisocial and delinquent behavior of adolescents includes: (a) general deviance, such as theft, alcohol use, cheating on exams, and coming to school late; (b) drug use; (c) defying parents, such as shouting at one's father or mother or going against your parents' wishes; and (d) aggressiveness, such as group fist fighting.²⁸ In the present study, these antisocial and delinquent behaviors, except for drug use, were measured. More emphasis was placed in this study on the investigation of the antisocial and aggressive acts in school and home settings.

People usually think that girls are more prosocial and less antisocial than boys. However, research findings on the sex differences in prosocial behavior are not as clear and consistent as commonly believed.¹⁹ On the other hand, research findings support that boys are generally more antisocial than girls.²²

Prosocial and antisocial Internet behavior

It is argued that the behavior of Internet use is similar to daily life social behavior. In other words, the behavior of Internet use is a kind of social behavior. Examples of prosocial Internet behavior include the use of the Internet: (a) as a means for teaching and learning, and thus enhancing one's knowledge and intelligence; (b) as a means of effective communication; (c) in dealing with daily matters such as banking, as well as buying food and daily necessities; (d) in organizing altruistic and voluntary activities to help people in need; and (e) for entertainment and relaxation such as watching video clips or listening to music. On the other hand, antisocial Internet behaviors include: (a) carrying out illegal activities such as selling faked products or offensive pornographic materials; (b) bullying others (i.e., cyberbullying), for example, distributing libelous statements against a certain person; (c) cheating others; and (d) gambling.

In short, the effects of excessive or problematic Internet use on academic achievement, mental health, and social relationships appear to be quite negative.^16,29–33 On the other hand, positive, prosocial, or healthy Internet use has been less investigated. In fact, proper use of the Internet may enrich one's knowledge about the world, improve one's communication with others, help solve some of the problems encountered in daily life, and provide interesting entertainment.^1,4,34,35 The present study attempts to fill this research gap, and investigates both prosocial and antisocial Internet use in one setting. The study of both prosocial and antisocial Internet behavior will help the investigators understand the full scope of social behavior in the Internet, which may help educators develop a suitable curriculum to prevent antisocial Internet behavior, as well as promote prosocial Internet behavior.

Hypotheses

Based on the previous review of the literature, the present study tests the following two major hypotheses:

(a) Sex differences hypothesis: Boys are more delinquent in Internet use than are girls. Past research indicates that boys are more delinquent or antisocial in daily social behavior than are girls, but the sex difference in prosocial behavior is usually insignificant.^26,36

(b) Positive association hypothesis: There is a positive association between the Internet behavior and daily social behavior. In other words, positive Internet behavior is positively associated with positive daily social behavior, and negative Internet behavior is positively associated with negative daily social behavior. The test of this hypothesis may shed some light on the similarities and differences between the social behavior in the cyber world and the real-life world.

Method

Participants

The participants were 65 Grade 7 (37 male, M, 28 female, F), 190 Grade 8 (108M, 82F), 148 Grade 10 (77M, 71F), and 106 Grade 12 (60M, 46F) students in Hong Kong, giving a total of N = 509 (282M, 227F). All participants were Chinese. The mean age (standard deviation in parentheses) of the Grades 7, 8, 10, and 12 participants was 13.03 (0.90), 13.94 (0.95), 15.60 (0.76), and 18.02 (0.88) respectively. The participants came from five government-funded schools that admitted students of heterogeneous social backgrounds and mixed academic standards, and were located in different areas of Hong Kong.

Instruments

Adolescent Internet Use Questionnaire

The construction of the Adolescent Internet Use Questionnaire (AIUQ) was based on previous studies by Morahan-Martin and Schumacher, and Caplan,^3,12 assessing the participants' Internet experience and pathological use. The AIUQ also measures the cognition, behaviors, and the negative outcomes associated with Davies's concept of generalized PIU.¹¹ Participants are asked to rate the items on a 5-point Likert scale ranging from “strongly disagree” to “strongly agree.” Three major scales of the AIUQ were constructed: (a) NIU: normal or prosocial Internet use, for example, “I would use the net when I'm down” (Q39) and “I seek others online when I feel isolated” (Q38); (b) DIU: delinquent or antisocial Internet use, for example, “I have gambled on the Internet” (Q32), “I have been late for appointments because I was online” (Q15), “I have failed an examination because of my excessive online activity” (Q26), and “I have missed family and school meetings because of online activities” (Q19); and (c) HIU: holistic Internet use, which is defined as HIU = NIU – DIU. A high (low) rating of HIU refers to high (low) rating of NIU and low (high) rating of DIU. In other words, a high HIU value means more prosocial and less antisocial use of the Internet, and a low HIU value refers to less prosocial and more antisocial use of the Internet. The AIUQ is a comprehensive questionnaire containing a substantial number of newly constructed items, of which 25 were NIU items and 18 DIU items. The total number of items in the questionnaire, which was constructed in Chinese, was 43. A section on Internet experience was also included, for example, (a) YEARIU in years: how long have you used the Internet? (b) WEEKIU in hours: on average, how many hours do you spend surfing the Internet every week?

Adolescent Behavior Questionnaire³⁷

The Adolescent Behavior Questionnaire (ABQ) measures the prosocial and antisocial/delinquent behavior of adolescents. Participants were asked to report the frequency of 65 acts performed in the past year, on a 7-point scale ranging from “none” to “more than 10 times.” There are two general scores on the ABQ. The antisocial/delinquent Behavior (DB) score measures the frequency of deviant or socially disapproved acts in a classroom or school setting, socially undesirable sexual activities, antisocial acts against one's teachers or school authority, antisocial acts against parents, and aggressive or hostile acts. Generally, a high DB score indicates a high frequency of delinquent acts performed in the past year. The prosocial behavior score (PB) measures the frequency of altruistic and socially acceptable or normative acts. A high PB score indicates a high frequency of prosocial acts performed in the past year. The DB or PB score was computed by averaging the ratings of relevant items in the score. Thus the rating of the DB or PB score ranges from 1.0 to 7.0. An overall score, adolescent behavior (AB) score is generated using the following formula: AB = PB − DB. The reliability and validity of the ABQ scores were demonstrated to be good in studies by Ma et al.^26,38 The reliability of the ABQ scores is usually in the 0.80s to 0.90s.

Prosocial Orientation Questionnaire³⁹

The Prosocial Orientation Questionnaire POQ measures various aspects of prosocial orientation and behavior of adolescents. It was constructed based on Ma and Leung's Child Altruism Inventory (CAI).³⁸ Some items in the CAI were used in the POQ. Participants responded to the POQ by indicating their degree of agreement to 40 statements on a 4-point scale: “strongly agree,” “agree,” “disagree,” and “strongly disagree.” There are four subscales in the POQ: (a) helping behavior (HB)—the tendency to help others in various situations (e.g., “I would spend time and money to help those in need”); (b) cooperation and sharing (CS)—the tendency to cooperate with others and to share things with others (e.g., “I welcome other classmates to join in while I am playing”); (c) affective relationship (AR)—the tendency to maintain an affective, friendly, and sympathetic relationship with others (e.g., “I feel very sad when my family member is sick”); and (d) normative behavior (NB)—the tendency to behave in compliance with the social norm (e.g., “I am very attentive during class lessons”). The average of all the items in the POQ is a measure of the general tendency to perform prosocial acts, and is therefore called general prosocial orientation (GPO). The internal consistency reliability of the four subscales of the POQ ranges from 0.55 to 0.69, and that of the GPO score is 0.87 (Cheung et al., p. 36).⁴⁰ The reliability of the POQ is thus regarded as acceptable.⁴¹

Procedures

Since the present study involved human subjects, details of our empirical study had to be submitted to the University' Committee on the Use of Human and Animal Subjects in Teaching and Research for endorsement. The Committee employs international ethical standards in scrutinizing all applications. After endorsement by the University's Committee, the investigators wrote to the principals of the participating schools to seek their consent for data collection.

Subjects were reminded that all their answers and personal information were for the use of research only and would be kept confidential. Subjects were free to withdraw from the study at any stage. The questionnaires were group administered to participants during their normal class periods. The experimenter went through the test instructions with the participants before they answered the questions.

Results

Reliability analysis

The data were analyzed using SPSS statistical software (V16.0; SPSS, Inc., Chicago, IL). The internal consistency reliability of the AIUQ was studied with the sample of 442 Hong Kong participants who completed all 43 items of the AIUQ. The reliability of the NIU and DIU was 0.86 and 0.84 respectively. The internal consistency reliability of the whole AIUQ was 0.88. The reliability of the AIUQ is therefore regarded as acceptable.

Correlation analysis

The means and standard deviations of the test scores by sex are presented in Table 1. The correlation of Grade with NIU, DIU, and HIU was −0.03, 0.04, −0.06 respectively. All these correlations are not significant at the 5% level. The variable Grade was therefore not included in the following analysis.

Table 1.

Means and Standard Deviations of Test Scores

	Total sample		Boys		Girls
Score	M	SD	M	SD	M	SD	T
Adolescent Internet Use Questionnaire (Total N = 509; 282 boys and 227 girls)^#
NIU	3.25	0.50	3.26	0.52	3.25	0.48	0.19
DIU	2.11	0.55	2.23	0.59	1.97	0.45	5.55^***
HIU	1.14	0.59	1.03	0.61	1.28	0.53	−4.81^***
YEARIU (Years)	5.53	2.27	5.46	2.32	5.63	0.2.20	−0.81
WEEKIU (Hours)	21.84	22.33	25.66	23.87	17.05	19.22	4.45^***
Adolescent Behavior Questionnaire (Total N = 509; 282 boys and 227 girls)
PB	2.52	0.84	2.47	0.86	2.58	0.81	−1.50
DB	1.70	0.65	1.83	0.72	1.54	0.52	5.19^***
AB	.82	0.91	0.64	0.94	1.04	0.83	−5.00^***
Prosocial Orientation Questionnaire (Total N = 113; 57 boys and 56 girls)
HB	2.92	0.37	2.88	0.37	2.95	0.36	−1.05
CS	3.05	0.36	3.01	0.35	3.08	0.36	−0.93
AR	3.08	0.40	3.08	0.38	3.08	0.43	−0.08
NB	3.08	0.35	3.05	0.35	3.11	0.34	−0.97
GPO	3.03	0.32	3.00	0.32	3.05	0.32	−0.82

Note. NIU, normal/prosocial Internet use; DIU, delinquent/antisocial Internet use; HIU, holistic Internet use; YEARIU, no. of years using Internet; WEEKIU, no. of hours using Internet each week; PB, prosocial behavior; DB, delinquent behavior; AB, adolescent behavior; HB, helping behavior; CS, cooperation and sharing; AR, affective relationship; NB, normative behavior; GPO, general prosocial orientation.

The number of participants for YEARIU and WEEKIU in the Hong Kong study is N = 492, 273 boys and 219 girls, and N = 498, 277 boys and 221 girls, because of missing data in these two variables.

***

p < 0.001.

There were significant sex differences in DIU, HIU, DB, and AB. Boys scored significantly higher in DIU and DB and significantly lower in HIU and AB than did girls, indicating that boys were, in general, more delinquent/antisocial and less prosocial in Internet use and daily social behavior than were girls. In addition, the WEEKIU was significantly higher in boys than in girls.

The correlations of the AIU scores with other test scores are presented in Table 2. Generally, the correlations NIU and PB, DIU and DB, and HIU and AB were significant and positive. In other words, normal/prosocial Internet use was associated positively with prosocial behavior; however, delinquent/antisocial Internet use was associated positively with delinquent behavior. Overall or holistic Internet use was associated positively with overall adolescent social behavior. The correlation pattern of AIUQ and POQ scores indicates that NIU and HIU scores were significantly and positively associated with the POQ scores. On the other hand, the DIU score was significantly and negatively associated with the POQ scores. The strong and consistent relationship of AIUQ scores with the ABQ and POQ scores provides some support for the predictive validity of the AIUQ.

Table 2.

Correlation of AIU Scores with Other Test Scores

Score	NIU	DIU	HIU	YEARIU	WEEKIU
Adolescent Internet Use Scores
YEARIU^#	0.06	0.06	−0.01	1.00	0.15^**
WEEKIU^#	0.17^***	0.33^***	−0.16^***	0.15^**	1.00
Adolescent Behavior Scores (N = 509)
PB	0.22^***	−0.18^***	0.35^***	0.07	−0.02
DB	0.16^***	0.39^***	−0.23^***	0.16^**	0.26^**
AB	0.08	−0.44^***	0.48^***	−0.05	−0.21^***
Prosocial Orientation Scores (N = 113)
HB	0.30^**	−0.26^**	0.47^***	−0.05	−0.31^**
CS	0.32^**	−0.08	0.36^***	0.04	−0.15
AR	0.20^*	−0.24^*	0.37^***	−0.05	−0.23^*
NB	0.32^**	−0.19^*	0.43^***	−0.09	−0.16
GPO	0.32^**	−0.23^*	0.47^***	−0.05	−0.25^**

The number of participants for YEARIU and WEEKIU in the Hong Kong study is N = 492, 273 boys and 219 girls, and N = 498, 277 boys and 221 girls, because of missing data in these two variables.

p < 0.05; **p < 0.01; ***p < 0.001.

The correlations of WEEKIU and DIU and WEEKIU and DB were significant and positive, while WEEKIU and HIU and WEEKIU and AB were significant and negative. That is, the more a participant spent time using the Internet each week, the more antisocial and less prosocial was the Internet use and daily social behavior. The negative and significant correlations between WEEKIU and GPO (as well as HB and AR) also indicate a similar result.

Multiple regression analysis

Two multiple regression analyses were conducted. The first used AB as the dependent variable, and the following scores were entered as independent variables: (a) NIU, (b) DIU, and (c) WEEKIU. The results are presented in Table 3. The multiple R, R², and adjusted R² were 0.527, 0.277, and 0.273 respectively. Since there were 11 cases of missing data on the WEEKIU variable, the total number of participants in this analysis was 498. This regression analysis produced the following overall result: F(3, 494) = 63.197, p < 0.001, indicating that the regressions of AB on the three independent variables (NIU, DIU, and WEEKIU) fit nicely. Hierarchical multiple regression was used to examine further the contribution of individual independent variables to the overall variance of AB. The results given in Table 4 indicate that the change in R² due to NIU, DIU, and WEEKIU was 0.008, F(1, 496) = 3.8, p = 0.053; 0.264, F(1, 495) = 179.1, p < 0.0005; and 0.006, F(1, 494) = 4.2, p = 0.041 respectively, implying that the percentage variance contributed by NIU and WEEKIU was rather weak.

Table 3.

Multiple Regression Analyses

Independent variable	B	SE	β	t
Dependent variable: AB (N = 498)
NIU	0.542	0.075	0.296	7.184^***
DIU	−0.881	0.072	−0.527	−12.246^***
WEEKIU	−0.003	0.002	−0.083	−2.049^*
Dependent variable: HIU (N = 498)
PB	0.310	0.029	0.440	10.794^***
DB	−0.292	0.038	−0.324	−7.667^***
WEEKIU	−0.002	0.001	−0.070	−1.714

Note. NIU, normal/prosocial Internet use; DIU, delinquent/antisocial Internet use; HIU, holistic Internet use; WEEKIU, no. of hours using Internet each week; PB, prosocial behavior; DB, delinquent behavior; AB, adolescent behavior.

p < 0.05; ***p < 0.001.

Table 4.

Hierarchical Multiple Regression (Using AB as the Dependent Variable)

					Change statistics
Model	R	R²	Adj. R²	Std. error of the estimate	R² change	F change	df1	df2	Sig. F change
1	0.087^a	0.008	0.006	0.91482	0.008	3.777	1	496	0.053
2	0.521^b	0.271	0.268	0.78474	0.264	179.067	1	495	0.000
3	0.527^c	0.277	0.273	0.78221	0.006	4.200	1	494	0.041

Predictors: (constant), normal/prosocial Internet use.

Predictors: (constant), normal/prosocial Internet use; PIU: delinquent Internet use: DIU.

Predictors: (constant), normal/prosocial Internet use: NIU; delinquent Internet use: DIU; WEEKIU.

The second regression analysis used HIU as the dependent variable, and the following scores were entered as independent variables: (a) PB, (b) DB, and (c) WEEKIU. The results are presented in Table 3. The multiple R, R², and adjusted R² were 0.492, 0.242, and 0.237 respectively. This regression analysis produced the following overall result: F(3, 494) = 52.537, p < 0.001, indicating that the regressions of HIU on the three independent variables (PB, DB, and WEEKIU) also fit nicely. In a similar vein, hierarchical multiple regression was used to examine further the contribution of individual independent variables to the overall variance of HIU. The results given in Table 5 indicate that the change in R² due to PB, DB, and WEEKIU was 0.127, F(1, 496) = 72.1, p < 0.0005; 0.110, F(1, 495) = 71.7, p < 0.0005; and 0.005, F(1, 494) = 2.94, p = 0.087 respectively, implying that the percentage variance contributed by PB and DB was large and comparable with each other and that the variance contributed by WEEKIU was insignificant.

Table 5.

Hierarchical Multiple Regression (Using HIU as the Dependent Variable)

					Change statistics
Model	R	R²	Adj. R²	Std. error of the estimate	R² change	F change	df1	df2	Sig. F change
1	0.356^a	0.127	0.125	0.55230	0.127	72.113	1	496	0.000
2	0.487^b	0.237	0.234	0.51671	0.110	71.680	1	495	0.000
3	0.492^c	0.242	0.237	0.51570	0.005	2.939	1	494	0.087

Predictors: (constant), prosocial behavior; PB.

Predictors: (constant), prosocial behavior: PB; delinquent behavior: DB.

Predictors: (constant), prosocial behavior: PB; delinquent behavior: DB; WEEKIU.

Classification of social behavior groups

The classification of participants into four social behavior groups based on the means of their prosocial and antisocial behavior scores was found to be a valid and useful method for investigating the social behavior of adolescents by Ma.⁴² This kind of classification method is also used in the study of gender roles. The mean or median of the masculinity and femininity scores is used to divide subjects into four gender-role groups (for details, see Ma's study).³⁶

The participants were classified into four daily social behavior groups using the mean DB (1.7017) and mean PB (2.5200) scores as follows: (a) prosocial daily behavior group: PB was greater than or equal to its mean, and DB was less than its mean; (b) active group: PB was greater than or equal to its mean, and DB was greater than or equal to its mean; (c) passive group: PB was less than its mean, and DB was also less than its mean; (d) delinquent (or antisocial) daily behavior group: PB was less than its mean, and DB was greater than or equal to its mean. Results of the one way analysis of variance (ANOVA) of the AIUQ scores for these four social behavior groups indicated significant differences in all the mean AIU scores across the four groups (see Table 6). Post-hoc comparison using the Scheffé test showed that: (a) the general HIU score was more differentiating than specific scores (NIU and DIU); (b) the DIU score was more differentiating than the NIU score; (c) the WEEKIU was more differentiating than the YEARIU; and (d) five out of six pairs of comparisons were significant for the HIU score. In general, the HIU score was significant between the prosocial daily behavior group and the other three groups.

Table 6.

Mean AIU Scores by Social Behavior Group in Hong Kong Study

	Social behavior group
Adolescent Internet Use scores	Prosocial (n = 99)	Active (n = 114)	Passive (n = 218)	Delinquent (n = 78)	F
NIU	3.32	3.38	3.16	3.25	5.79^**
DIU	1.77	2.24	2.08	2.43	28.30^***
HIU	1.55	1.14	1.07	0.82	28.64^***
YEARIU^#	5.26	6.24	5.27	5.55	5.19^**
WEEKIU^##	13.44	29.75	18.35	30.82	16.60^***
Scheffe contrast
NIU	A-PA^**
DIU	PR-A^***	PR-PA^***	PR-D^***	D-PA^***
HIU	PR-A^***	PR-PA^***	PR-D^***	A-D^**	PA-D^**
YEARIU^#	PR-A^*	A-PA^*
WEEKIU^##	PR-A^***	PR-D^***	A-PA^***	PA-D^***

Note. NIU, normal/prosocial Internet use; DIU, delinquent/antisocial Internet use; HIU, holistic Internet use; PAT, online activities; PLT, real-life behavior; PR, prosocial group; A, active group; PA, passive group; D, delinquent group.

The number of participants in the prosocial, active, passive, and delinquent groups is 97, 113, 207, and 76 respectively.

The number of participants in the prosocial, active, passive, and delinquent groups is 99, 114, 211, and 75 respectively.

p < 0.05; **p < 0.01; ***p < 0.001.

Discussion

The internal consistency reliability of the major indices of the AIUQ is in the range of 0.84 to 0.88 in this study. The AIUQ is therefore regarded as a reliable instrument.

Significant sex differences were found in DIU and HIU. Boys were more delinquent in Internet use than were girls. Hypothesis 1 is therefore supported. In addition, boys were more delinquent in daily behavior than were girls.^26,36

Boys spent more time using the Internet each week than did girls. Perhaps this finding is related to the fact that Chinese people tend to have higher expectations in the educational achievements of boys than of girls, and boys are often given more opportunities and resources than are girls.⁴³

The findings in Table 2 indicate that NIU is associated positively with PB, and DIU is also associated positively with DB. In addition, the prosocial orientation scores (HB, CS, AR, NB, GPO) are associated positively with NIU and HIU, and negatively with DIU. All correlations are significant except CS and DIU. The results provide clear support for Hypothesis 2.

Although the multiple regression analysis with the AB score as the dependent variable and the NIU, DIU, and WEEKIU scores as independent variables indicate that the AB score is significantly related to the three independent variables, the percentage variance of the AB score explained by the model was mainly contributed by NIU (26.8% out of the overall 27.3%). In addition, the multiple regression analysis with the HIU score as the dependent variable and the PB, DB, and WEEKIU scores as independent variables indicate that the HIU score is significantly related to the three independent variables. In particular, the three independent variables account for 23.7% of the variance of the HIU score. However, results from the hierarchical multiple regression analysis indicate that the contribution of WEEKIU to the variance of HIU is statistically insignificant. The multiple regression results, in general, provide support for Hypothesis 2.

WEEKIU is associated positively and significantly with DIU but negatively and significantly with HIU. In other words, the more one uses the Internet, the more one is likely to exhibit antisocial Internet use. On the other hand, WEEKIU is associated positively and significantly with NIU. It seems that heavy use of the Internet by Hong Kong students is also associated with prosocial Internet use. While heavy use of the Internet tends to be associated with antisocial Internet use, in general, it is also associated positively with prosocial Internet use in Hong Kong participants.

HIU is the highest in the prosocial daily behavior group and the lowest in the delinquent baily behavior group. Further, DIU is the lowest in the prosocial daily behavior group and the highest in the delinquent daily behavior group. In general, prosocial adolescents tend to be more prosocial and less antisocial in Internet use than antisocial adolescents. Similarly, delinquent adolescents tend to be more antisocial and less prosocial in Internet use than prosocial adolescents. Since the prosocial behavior group and the delinquent behavior group are classified based on their real-life behavior, this finding in some sense provides some support to the positive association hypothesis (i.e., Hypothesis 2), which hypothesizes that there is a positive association between the social behavior in the cyber world and the real-life world.

Concluding remarks

It is now clear that our two hypotheses are supported by the data. Integrating all the findings together, it is also clear that (a) prosocial Internet use is associated positively with prosocial daily social behavior, and (b) delinquent Internet use is likewise associated positively with delinquent daily social behavior. While there is no doubt that Internet use is a daily social behavior, the present findings indicate that the behavioral pattern on the Internet is in some sense similar to that in real daily life. In addition, the Internet behavior measured by the major indices of the AIUQ is closely related with the relevant social behavioral indices measured by the ABQ. In this regard, the predictive validity of the major indices of AIUQ is substantiated by the present findings.

Educational implications

The educational implications of the present study are twofold. First, since Internet behavior is quite similar to social behavior in real life, Internet ethics should be our major concern in the education of our children. Second, care should also be taken with the emotion and interpersonal relationships associated with Internet behavior in our school education.

Limitations and directions for future research

Although the present study provides good empirical data regarding the relationship of Internet use behavior to prosocial and antisocial daily social behavior in Chinese adolescents in Hong Kong, there are some limitations in the present study. First, the use of the self-report method in the study of the prosocial and antisocial Internet behavior has some limitations. For example, research participants may not always answer honestly. Nevertheless, this method has been found to be useful and acceptable, especially in finding out the correlations among a large number of variables.^26,28,44 Second, there are no data on the personality, mental disorder, and individual differences in academic achievement in the present study. This kind of data may shed some light on the explanation of the causes of the prosocial and antisocial Internet behavior, and future studies should include this kind of data. Third, the present study is a cross-sectional study, and further research should include a longitudinal study to explore the causes and effects among the variables. Fourth, it would be interesting to conduct cross-cultural studies in other countries to substantiate the findings in the present study. Finally, it is important to conduct more research to justify that the virtual (cyber) world is not virtual, and behavior in the cyber world is more or less the same as that in the real daily-life world. For example, is the motivation underlying the Internet behavior the same as that underlying the real-life social behavior? To what extent is the Internet behavior related to or even determined by one's morality or personality?

Footnotes

Acknowledgments

This research was funded by faculty research grants from the Hong Kong Baptist University.

Disclosure Statement

No competing financial interests exist.

References

Chou

, Hsiao

. Internet addiction, usage, gratification, and pleasure experience: The Taiwan college students' case. Computers & Education, 2000; 35:65–80.

Lin

SSJ

, Tsai

. Sensation seeking and internet dependence of Taiwanese high school adolescents. Computers in Human Behavior, 2002; 18:411–26.

Morahan-Martin

, Schumacher

. Incidence and correlates of pathological Internet use among college students. Computers in Human Behavior, 2000; 16:13–29.

Wang

. Internet dependency and psychosocial maturity among college students. International Journal of Human–Computer Studies, 2001; 55:919–38.

Internet addiction in Adolescents: 42 hours weekly online. Singtao Post. Dec. 16, 2005.

Mitchell

. Internet addiction: Genuine diagnosis or not? The Lancet, 2000; 355:632.

Treuer

, Fabian

, Furedi

. Internet addiction associated with features of impulse control disorder: Is it a real psychiatric disorder? Journal of Affective Disorder, 2001; 66:283.

Griffiths

. Does Internet and computer “addiction” exist? Some case study evidence. CyberPsychology & Behavior, 2000; 3:211–18.

Beard

, Wolf

. Modification in the proposed diagnostic criteria for Internet addiction. CyberPsychology & Behavior, 2001; 4:377–83.

10.

Widyanto

, Griffiths

. “Internet addiction”: A critical review. International Journal of Mental Health & Addiction, 2006; 4:31–51.

11.

Davis

. A cognitive-behavioral model of pathological Internet use. Computers in Human Behavior, 2001; 17:187–95.

12.

Caplan

. Problematic Internet use and psychosocial well-being: Development of a theory-based cognitive-behavior measurement instrument. Computers in Human Behavior, 2002; 18:553–75.

13.

Greenberg

, Lewis

, Dodd

. Overlapping addictions and self-esteem among college men and women. Addictive Behaviors, 1999; 24:565–71.

14.

Chak

, Leung

. Shyness and locus of control as predictors of Internet addiction and Internet use. CyberPsychology & Behavior, 2004; 7:559–70.

15.

Niemz

, Griffiths

, Banyard

. Prevalence of pathological Internet use among university students and correlations with self-esteem, the General Health Questionnaire (GHQ), and disinhibition. CyberPsychology & Behavior, 2005; 6:562–70.

16.

Young

. 1999. Internet addiction: Symptom, evaluation, and treatment. VandeCreek

, Jackson

. Innovations in clinical practice: A source book, 17. Sarasota, FL: Professional Resources, 19–31.

17.

Engelberg

, Sjoberg

. Internet use, social skills, and adjustment. CyberPsychology & Behavior, 2004; 7:41–7.

18.

Swickert

, Hittner

, Harris

et al. Relationships among Internet use, personality, and social support. Computers in Human Behavior, 2002; 18:437–51.

19.

Eisenberg

, Fabes

. 1998. Prosocial development. Damon

series Eisenberg

vol.Handbook of child psychology, 5th 3. New York: John Wiley, 701–78.

20.

Radke-Yarrow

, Zahn-Waxler

, Chapman

. 1983. Children's prosocial dispositions and behavior. Mussen

series Hetherington

volHandbook of child psychology: 4. Socialization, personality, and social development, 4th. New York: John Wiley, 469–545.

21.

Staub

, Bar-Tal

. Karylowski

et al. 1984. Development and maintenance of prosocial behavior. New York: Plenum Press.

22.

Coie

, Dodge

. 1998. Aggression and antisocial behavior. Damon

series Eisenberg

volHandbook of child psychology, 5th 3. New York: John Wiley, 779–862.

23.

Gibbons

, Krohn

. 1986. Delinquent behavior, 4th. Englewood Cliffs, NJ: Prentice-Hall.

24.

Rutter

, Giller

. 1983. Juvenile delinquency: Trends and perspectives. Harmondsworth, England: Penguin.

25.

Widom

. Does violence beget violence? A critical examination of the literature. Psychological Bulletin, 1989; 106:3–28.

26.

, Shek

DTL

, Cheung

et al. The relation of prosocial and antisocial behavior to personality and peer relationships of Hong Kong Chinese adolescents. Journal of Genetic Psychology, 1996; 157:255–66.

27.

, Shek

DTL

, Cheung

et al. Parental, peer, and teacher influences on the social behavior of Hong Kong Chinese adolescents. Journal of Genetic Psychology, 2000; 161:65–78.

28.

Hindelang

, Hirschi

, Weis

. 1981. Measuring delinquency. Beverly Hills, CA: Sage.

29.

Douglas

, Mills

, Niang

et al. Internet addiction: Meta-synthesis of qualitative research for the decade 1996–2006. Computers in Human Behavior, 2008; 24:3027–44.

30.

Kaltiala-Heino

, Lintonen

, Rimpela

. Internet addiction? Potentially problematic use of the internet in a population of 12–18 year-old adolescents. Addiction Research & Theory, 2004; 12:89–96.

31.

Kraut

, Patterson

, Lundmark

et al.

Internet paradox: A social technology that reduces social involvement and psychological well-being?

American Psychologist, 1998; 53:1017–31.

32.

Shapira

, Goldsmith

, Keck

et al. Psychiatric features of individuals with problematic Internet use. Journal of Affective Disorders, 2000; 57:267–72.

33.

Yellowlees

, Marks

. Problematic internet use or Internet addiction. Computers in Human Behavior, 2007; 23:1447–53.

34.

Amichai-Hamburger

, Furnham

. The positive net. Computers in Human Behavior, 2007; 23:1033–45.

35.

Silverman

. The Internet and relational theory. American Psychologist, 1999; 54:780–781.

36.

. The relation of sex-role stereotypes to prosocial and antisocial behavior of Chinese adolescents. Journal of Genetic Psychology, 2005; 166:189–201.

37.

. Adolescent behavior questionnaire: An introduction. Rev.. Unpublished manuscript: Hong Kong Baptist University, 2002.

38.

, Leung

. Altruistic orientation in children: Construction and validation of the Child Altruism Inventory. International Journal of Psychology, 1991; 26:745–59.

39.

, Shek

DTL

, Cheung

. Prosocial Orientation Questionnaire. Unpublished test: The Chinese University of Hong Kong, 1992.

40.

Cheung

, Ma

, Shek

DTL

. Conceptions of success: Their correlates with prosocial orientation and behavior in Chinese adolescents. Journal of Adolescence, 1998; 21:31–42.

41.

, Cheung

, Shek

DTL

. The relation of prosocial orientation to peer interactions, family social environment and personality of Chinese adolescents. International Journal of Behavioral Development, 2007; 31:12–18.

42.

. The relation of moral orientation and moral judgment to prosocial and antisocial behavior of Chinese adolescents. International Journal of Psychology, 2003; 38:101–11.

43.

Cheung

. 1996. Gender role development. Lau

. Growing up the Chinese way: Chinese child and adolescent development. Hong Kong: The Chinese University Press, 44–67.

44.

Cheung

, Ng

AMC

. Social factors in adolescent deviant behavior in Hong Kong: An integrated theoretical approach. International Journal of Comparative & Applied Criminal Justice, 1998; 12:29–45.

The Relation of Internet Use to Prosocial and Antisocial Behavior in Chinese Adolescents

Abstract

Abstract

Introduction

Prosocial and antisocial behavior

Prosocial and antisocial Internet behavior

Hypotheses

Method

Participants

Instruments

Adolescent Internet Use Questionnaire

Adolescent Behavior Questionnaire 37

Prosocial Orientation Questionnaire 39

Procedures

Results

Reliability analysis

Correlation analysis

Multiple regression analysis

Classification of social behavior groups

Discussion

Concluding remarks

Educational implications

Limitations and directions for future research

Footnotes

Acknowledgments

Disclosure Statement

References

Adolescent Behavior Questionnaire³⁷

Prosocial Orientation Questionnaire³⁹