Internet Addiction and Psychosocial Maladjustment: Avoidant Coping and Coping Inflexibility as Psychological Mechanisms

Abstract

This 6 month prospective study systematically tested some multivariate models that advanced the understanding of the psychological mechanisms underlying Internet addiction and psychosocial maladjustment. On the basis of previous theories, avoidant coping and coping inflexibility were proposed as underlying mechanisms. Participants were 271 Chinese undergraduates (75% women, M_age=20.49) who took part in both phases of this study. Structural equation modeling was performed to obtain the best fit models for both the cross-sectional and the prospective data. The cross-sectional model testing revealed statistically significant mediating effects for both avoidant coping (β=0.149 [95% CI 0.071–0.226], p=0.002) and coping flexibility (β=0.048 [95% CI 0.013–0.081], p=0.032). The prospective model testing further showed that avoidant coping mediated the relationship between Internet addiction and Time 2 psychosocial maladjustment (β=0.141 [95% CI 0.065–0.216], p=0.005), as well as that between coping flexibility and Time 2 psychosocial maladjustment (β=−0.096 [95% CI −0.161 to −0.031], p=0.015). This study was the first to establish theory-driven models, which unveiled an inflexible, avoidant coping style as psychological mechanisms that explained the link between Internet addiction and psychosocial maladjustment.

Introduction

Problematic Internet use is common, especially among young people in Asia.¹ Internet addiction has raised concerns among practitioners, parents, teachers, and the mass media worldwide because it poses threats to the psychosocial adjustment of a notable portion of users.^2,3 Studies have focused on scrutinizing the adjustment problems faced by users with Internet addiction,^4,5 but scant attempts have been made to explore systematically the underlying mechanisms that accounted for their psychosocial maladjustment. Also, most of these studies were cross-sectional in nature, and the longitudinal effects of psychological risk factors of Internet addiction remained largely unexplored. To fill these knowledge gaps, the present study aimed to formulate and test several theory-driven models that specify psychological mechanisms for explicating the link between Internet addiction and psychosocial maladjustment. A two-phase prospective study was conducted in order to perform both cross-sectional and prospective model testing.

Internet addiction and psychosocial maladjustment

A meta-analysis has indicated that Internet addiction is related to emotional disorders including depression and anxiety.⁶ Apart from emotional problems, individuals with Internet addiction also frequently encounter interpersonal problems.⁷ The rich-get-richer hypothesis states that online communication may benefit extroverts who are endowed with abundant offline social resources.⁸ Such psychological benefits, however, are absent for introverts because Internet use results in the allocation of inadequate time for social interactions,⁹ thus shrinking their already scarce social resources. As individuals with Internet addiction were found to report lower levels of extraversion,¹⁰ their excessive Internet use may reduce opportunities to polish their inadequate social skills in the real-life social milieu, thus increasing their susceptibility to social maladjustment. The following hypothesis was tested:

H1: Internet addiction is positively related to psychosocial maladjustment.

Mediating role of avoidant coping

Studies have revealed that individuals with Internet addiction report higher levels of stress than their counterparts without this addictive problem.^11,12 According to the transactional theory of coping,¹³ some individuals who consider the event a challenge tend to confront and tackle the problem directly (i.e., approach coping), whereas those who consider the same event an uncontrollable threat tend to refrain from facing it (i.e., avoidant coping). It was predicted that individuals with Internet addiction are characterized by an avoidant coping style. This notion stems from studies revealing that these individuals tend to be more dissatisfied with their real life,² and that they may have a greater tendency to escape from the self, which unveils one's wish to avoid pondering the painful past and to engage in distracting tasks.¹⁴ Unlike the real world, everything in the virtual world seems alterable and controllable,¹⁵ and individuals with Internet addiction may rely more on online activities to get away from their real-life stress.¹⁶ Taken together, it was predicted that these individuals tend to use avoidant coping more than those without Internet addiction. The following hypothesis was therefore tested:

H2: Internet addiction is positively related to avoidant coping.

Transactional theory states that adaptive coping strategies should be distinguished from maladaptive ones. Approach coping strategies (e.g., active coping, planning) are generally deemed adaptive in tackling the problems directly, thus fostering psychosocial adjustment.¹⁷ In contrast, avoidant coping strategies (e.g., denial, self-blame) are generally considered maladaptive because the problems remain intact rather than disappearing, thus eliciting psychosocial maladjustment in the long run.¹⁸ In light of previous findings that revealed the mediating effects of coping between addictive behaviors (e.g., alcohol use) and social anxiety among university students,¹⁹ it was predicted that greater use of avoidant coping may account for the link between Internet addiction and psychosocial maladjustment. The following hypotheses were tested:

H3: Avoidant coping is positively linked with psychosocial maladjustment.

H4: Avoidant coping mediates the link between Internet addiction and psychosocial maladjustment.

Mediating role of coping inflexibility

Transactional theory also states that coping represents an ongoing evolving process. Flexible deployment of coping strategies is thus essential for adjustment to the ever-changing environment. Coping flexibility (vs. inflexibility) refers to the capability of modifying one's strategies to meet specific situational demands.²⁰ It was predicted that individuals may display coping inflexibility, as indicated by consistent deployment of avoidant coping across situations. This prediction stems from recent neurobiological evidence, which reveals that individuals with Internet addiction have impaired decision making abilities.²¹ Such an impairment interferes with their processing of feedback-related information and judgment of the appropriateness of their responses, both of which are cognitive mechanisms essential for flexible coping to take place.²² It is thus reasonable to infer an inverse link between Internet addiction and coping flexibility, and excessive Internet use may imply a manifestation of inflexible deployment of avoidant coping regardless of situational characteristics. The following hypotheses were tested:

H5: Internet addiction is inversely associated with coping flexibility.

H6: Coping flexibility is inversely associated with avoidant coping.

Individuals with Internet addiction may keep engaging in online activities despite encountering devastating consequences of their online behaviors.²¹ Such inflexibility is maladaptive due to an incapability of detecting the contingency between their behaviors and event outcomes.²³ According to the theory of learned helplessness,²⁴ such a perception of lacking control over the situation hinders attempts to adapt to the changing environment, and ultimately induces a state of learned helplessness and depression. In the coping literature, results provide support for the beneficial role of coping flexibility by indicating that individuals who display greater flexibility generally experience higher levels of psychosocial well-being, whereas those who are less flexible generally encounter more psychosocial problems.^25,26 In this light, it was predicted that coping inflexibility may account for the relationship between Internet addiction and psychosocial maladjustment. The following hypotheses were tested:

H7: Coping flexibility is inversely linked with psychosocial maladjustment.

H8: Coping flexibility mediates the link between Internet addiction and psychosocial maladjustment.

In summary, a set of hypotheses was derived from previous theories and findings. Several multivariate models were constructed to test these hypotheses, specifically the proposed mediating roles of avoidant coping and coping flexibility between Internet addiction and psychosocial maladjustment. Structural equation modeling was performed to obtain the best fit models for both the cross-sectional and prospective data.

Method

Research design

This study contained two phases conducted 6 months apart. Adopting a prospective design, all the psychosocial variables were assessed in the initial phase, but only the criterion variables were assessed in the follow-up phase.

Participants

A total of 294 Chinese undergraduates voluntarily took part in both phases. After discarding the data with missing or incorrect identity codes, the final sample comprised 271 Chinese participants (59 men, 204 women, 8 did not specify), with an average age of 20.49 years (SD=1.22 years). Participants who completed both phases and those who dropped out did not differ in any Time 1 measures.

Procedures

This study obtained prior ethical approval from the university's Institutional Review Board. Participants had to sign a consent form before filling out a set of questionnaires (see below). In the follow-up session, those who indicated interest received the questionnaires. All the participants were thanked and debriefed at the end of the study.

Measures

Internet addiction was assessed by two popular instruments: the Internet Addiction Test and the Online Cognition Scale.^27,28 The latter examined four types of cognition regarding Internet addiction: emotional discomfort, diminished impulse control, distraction, and social comfort (reverse scored).

Avoidant coping style was assessed by four subscales of the Brief COPE: denial, substance use, behavioral disengagement, and self-blame.²⁹ The Coping Flexibility Scale was employed to measure two coping flexibility processes: evaluation and adaptation.³⁰

To assess psychosocial maladjustment, three common criterion measures were included: Emotional and Social Loneliness Scale, negative affect subscale of the Positive Affect and Negative Affect Schedule, and Interaction Anxiousness Scale.^31–33

All the measures were in Chinese, the native language of the participants. These measures were validated in Chinese samples and were found to display good psychometric properties.^34,35

Results

A priori statistical analysis plan

To test the proposed models, structural equation modeling (SEM) was performed using SPSS Statistics for Windows v19.0 (IBM Corp., Armonk, NY) and MPlus 6.0. We evaluated the level of overall data fit of each tested model using the following fit indices³⁶: comparative fit index (CFI), root mean square error of approximation (RMSEA), standardized root mean square residual (SRMR), and Tucker–Lewis index (TLI).

The candidate models were then compared using the Akaike information criterion (AIC) and Bayesian information criterion (BIC). According to the general rule of thumb, the preferred model would be the one with the lowest AIC and BIC values.³⁷ A more robust test of the mediating effects of the preferred model was carried out using the bootstrap estimation procedures, which involved repeated sampling of the dataset and estimating the indirect effect in each resampled dataset.³⁸ The indirect effect was significant at the 0.05 level if the 95% confidence intervals (CI) for the average estimates excluded zero.³⁹

Cross-sectional model testing

Before performing cross-sectional model testing, the study first examined whether there were sex differences. The preliminary analyses revealed no overall sex differences, and thus all the analyses were performed based on the pooled sample. Table 1 presents the descriptive statistics of the study variables.

Table 1.

Descriptive Statistics for Time 1 and Time 2 Variables

Variable	M	SD	α
Young's Internet Addiction Scale	35.59	10.02	0.87
Online Cognition Scale (Internet addiction)
Emotional discomfort	19.29	5.66	0.71
Distraction	22.19	7.78	0.84
Diminished impulse control	35.46	7.87	0.70
Social comfort (reverse scored)	42.95	11.49	0.86
Brief COPE (avoidant coping)
Denial	3.83	1.43	0.60
Substance use	2.93	1.53	0.86
Behavioral disengagement	4.36	1.36	0.67
Self-blame	5.70	1.33	0.64
Coping Flexibility Scale
Evaluation coping process	14.55	1.98	0.73
Adaptive coping process	15.70	2.69	0.85
Psychosocial maladjustment indicators
Loneliness—Time 1	43.70	7.55	0.88
Negative affect—Time 1	22.80	7.04	0.90
Social anxiety—Time 1	40.50	7.81	0.81
Loneliness—Time 2	41.10	7.84	0.83
Negative affect—Time 2	22.64	6.57	0.87
Social anxiety—Time 2	42.90	7.40	0.73

Figure 1 depicts the standardized parameters of all the cross-sectional models, and Table 2 presents the fit indices of each of these models. The upper panel of Table 2 shows that Models 1 and 2 both had a good fit of the baseline data. However, the models significantly differed from each other (Δχ²[1]=22.90, p<0.01; AIC difference=20.90, BIC difference=16.67), indicating that the additional path from Internet addiction to Time 1 psychosocial maladjustment had a considerable contribution to the explanatory power of Model 2 (see Fig. 1b).

FIG. 1.

Results from cross-sectional model testing. (a) Model 1; (b) Model 2; (c) Model 3. IA1, Young's Internet Addiction Test; IA2, Emotional Discomfort subscale of Online Cognition Scale; IA3, Diminished Impulse Control subscale of Online Cognition Scale; IA4, Distraction subscale of Online Cognition Scale; IA5, Social Comfort subscale of Online Cognition Scale (reverse scored); CF1, Evaluation Coping Process subscale of Coping Flexibility Scale; CF2, Adaptive Coping Process subscale of Coping Flexibility Scale; AC1, Denial subscale of Brief COPE; AC2, Substance Use subscale of Brief COPE; AC3, Behavioral Disengagement subscale of Brief COPE; AC4, Self-Blame subscale of Brief COPE; PM-T1, psychosocial maladjustment assessed at Time 1; PM1, Emotional and Social Loneliness Scale; PM2, Negative Affect subscale of Positive Affect and Negative Affect Schedule; PM3, Interaction Anxiousness Scale.

Table 2.

Fit Indices of Candidate Models

Model	χ ²	df	CFI	TLI	RMSEA	SRMR	AIC	BIC
Cross-sectional model testing
1	267.56	73	0.91	0.93	0.073	0.062	36,344.59	36,538.92
2	244.66	72	0.92	0.94	0.069	0.055	36,323.69	36,522.25
3	233.73	71	0.92	0.94	0.067	0.048	36,314.76	36,517.54
Prospective model testing
4	226.89	71	0.94	0.92	0.061	0.054	33,453.19	33,663.11
5	230.25	73	0.94	0.92	0.061	0.058	33,452.57	33,653.75

df, degrees of freedom; CFI, comparative fit index; TLI, Tucker–Lewis index; RMSEA, root mean square error of approximation; SRMR, standardized root mean square residual; AIC, Akaike information criterion; BIC, Bayesian information criterion.

The study further examined Model 3 that contained an additional path from coping flexibility to avoidant coping (see Fig. 1c). Results showed that this additional path was statistically significant (β=−0.24, p=0.002), and this model also showed a good data fit (see the upper panel of Table 2). When Model 3 with Model 2 were compared, the additional path substantially contributed to the explanatory power of Model 3 (Δχ² [1]=10.93, p<0.01; AIC difference=8.93, BIC difference=4.71). Model 3 was the preferred cross-sectional model due to its smallest AIC and BIC values.

The statistically significant mediating effects of avoidant coping and coping flexibility identified in Model 3 were then tested using the bootstrap estimation procedures. As shown in the upper panel of Table 3, Internet addiction was significantly related to Time 1 psychosocial maladjustment. Moreover, both avoidant coping and coping flexibility were significant mediators, with statistically significant effects comprising (a) the mediating role of avoidant coping between Internet addiction and Time 1 psychosocial maladjustment; (b) the mediating role of coping flexibility between Internet addiction and Time 1 psychosocial maladjustment; and (c) the mediating role of avoidant coping between coping flexibility and Time 1 psychosocial maladjustment.

Table 3.

Standardized Effects and 95% Confidence Intervals for Models 3 and 5

		95% CI
Path	Estimated effect	Lower bound	Upper bound
Model 3 (Cross-sectional model testing)
Direct effect
IA→PM-T1	0.291^***	0.176	0.397
Indirect effect
IA→AC→PM-T1	0.149^**	0.071	0.226
IA→CF→PM-T1	0.048^*	0.013	0.081
IA→CF→AC	0.038	0.002	0.073
CF→AC→PM-T1	−0.104^**	−0.165	−0.044
IA→CF→AC→PM-T1	0.017	0.001	0.033
Model 5 (Prospective model testing)
Indirect effect
IA→AC→PM-T2	0.141^**	0.065	0.216
IA→CF→AC	0.037^*	0.008	0.011
CF→AC→PM-T2	−0.096^*	−0.161	−0.031
IA→CF→AC→PM-T2	.001	-.015	-.029

p<0.05; ^**p<0.01.

CI, confidence intervals; IA, Internet addiction; PM-T1, psychosocial maladjustment assessed at Time 1; AC, avoidant coping; CF, coping flexibility; PM-T2, psychosocial maladjustment assessed at Time 2.

Prospective model testing

A prospective model (Model 4) was constructed based on the preferred model (Model 3) obtained from the cross-sectional analyses. As shown in Figure 2a, the results indicated that the path from Internet addiction to Time 2 psychosocial maladjustment, as well as that from coping flexibility to Time 2 psychosocial maladjustment, were not statistically significant.

FIG. 2.

Results from prospective model testing. (a) Model 4; (b) Model 5. PM-T2, psychosocial maladjustment assessed at Time 2; PM1, Emotional and Social Loneliness Scale assessed at Time 2; PM2, Negative Affect subscale of Positive Affect and Negative Affect Schedule assessed at Time 2; PM3, Interaction Anxiousness Scale assessed at Time 2. All scales and subscales were assessed at Time 1 unless specified. Dash arrows indicate statistically nonsignificant paths.

Then another model (Model 5) was built by omitting these two nonsignificant paths (see Fig. 2b). Although the two models similarly showed a good data fit (see the lower panel of Table 2), Model 5 had a slight improvement over Model 4 (AIC difference=−0.62, BIC difference=−9.36), as indicated by smaller AIC and BIC values. Model 5 was the preferred prospective model because it was more parsimonious with all the nonsignificant paths removed.

The lower panel of Table 3 displays the indirect effects and their respective 95% CI for Model 5. Results from the bootstrap estimation procedures revealed that the mediating effect of avoidant coping between Internet addiction and Time 2 psychosocial maladjustment was statistically significant. In addition, the indirect effect of avoidant coping between coping flexibility and Time 2 psychosocial maladjustment was also statistically significant.

Discussion

The present study is the first to explain the intricate relationships between Internet addiction and psychosocial maladjustment using both cross-sectional and prospective model testing. New findings indicate that individuals with Internet addiction are characterized by a maladaptive coping style that comprises both avoidant coping and coping inflexibility. First, the positive link between Internet addiction and avoidant coping indicates that Chinese young adults with Internet addiction may have a greater propensity to handle stress using avoidant coping. More important, avoidant coping is a psychological mechanism that explains the link between Internet addiction and psychosocial maladjustment. These findings imply that individuals with Internet addiction are more likely to find ways to escape from real-life difficulties rather than to handle stress actively.

The new findings also indicate that inflexible coping style is another mediator that accounts for the link between Internet addiction and psychosocial maladjustment. From a biological perspective, the greater tendency to cope inflexibly with environmental changes may be related to the greater brain activation in bilateral precuneus among individuals with Internet addiction.⁴⁰ Taken together, the inverse link between avoidant coping and coping flexibility indicates that individuals with Internet addiction are characterized by an inflexible, avoidant coping style. The psychosocial maladjustment among individuals with Internet addiction can be explained by their deployment of avoidant coping regardless of the characteristics of stressful events. Such an inflexible, avoidant coping style bears some resemblance to learned helplessness, a psychological condition in which individuals perceive having no control over their environment due to prolonged repeated exposure to aversive events.⁴¹ Hence, the present results provide evidence that the classic theory of learned helplessness,²⁴ formulated a few decades ago, may also be applicable to explicate the contemporary problem of Internet addiction.

Results from the prospective model testing document that only avoidant coping, but not coping inflexibility, can explain the link between Internet addiction and subsequent psychosocial maladjustment. Results further show that avoidant coping also accounts for the link between coping inflexibility and subsequent psychosocial maladjustment. These findings suggest that coping inflexibility per se cannot account for the maladaptive aspect of Internet addiction, and coping inflexibility predicts subsequent psychosocial maladjustment due to the deployment of avoidant coping. Such results are consistent with theories of coping flexibility,²⁰ which distinguishes two types of coping inflexibility: (a) active inflexible coping, which refers to the indiscriminant use of approach coping regardless of situational demands, and (b) passive inflexible coping, which refers to the indiscriminant use of avoidant coping regardless of situational demands. Previous studies have indicated that active and passive inflexible coping are related to dissimilar psychosocial outcomes, with more desirable outcomes associated with an active inflexible coping style and more undesirable ones associated with a passive inflexible coping style.^42,43 In line with these findings, the prospective findings of the current study specify that individuals with Internet addiction are characterized by a passive (avoidant) inflexible coping style but not an active inflexible one.

The present findings may have practical implications for mental health professionals working with clients with Internet addiction. Specifically, these new findings demonstrate that strengthening flexible coping—through reducing avoidant coping and enhancing approach coping—may foster psychosocial adjustment among this group of clients. Before delivering training sessions to achieve these overarching goals, practitioners should enhance the clients' awareness of their inflexible, avoidant coping style and its maladaptiveness in preventing them from facing real-life stressful events. Given that all coping strategies have their unique goals and characteristics,^44,45 the clients should then learn to distinguish the specific characteristics of an array of stressful events and to deploy situation-appropriate coping strategies to meet the specific demands of diverse stressful situations.⁴⁶

There are, however, some limitations of this study. This study only recruited young adults because individuals from this age cohort are the most vulnerable to Internet addiction,⁴⁷ and many of the participants were residing in urban cities. Previous findings have revealed differences in the susceptibility to Internet addiction between rural and urban youngsters,⁴⁸ and further studies should evaluate the generalization of the present findings to samples with distinct demographic characteristics such as adults from rural areas and hard-to-reach hidden youth.⁴⁹ Also, other addictive behaviors that are usually clustered with Internet addiction, such as smoking and compulsive overeating, were not studied here. These addictive behaviors together with Internet addiction may have conjoint effects on psychosocial maladjustment. Last but not least, this study examined psychosocial maladjustment as a criterion, but it is equally possible that it is an antecedent that leads to Internet addiction. Future studies should test the feasibility of this alternative model.

Despite such limitations, this study contributes to the literature by being the first to examine some psychological mechanisms, which explain the role of avoidant coping style and coping flexibility in accounting for the relationship between Internet addiction and psychosocial maladjustment. Unlike the scattered results in the coping literature, the present study examined both coping style and process in a more comprehensive manner. More important, it is also the first attempt to adopt both a cross-sectional and a prospective design that systematically test several candidate multivariate models, thus providing a useful theoretical framework that may lay a foundation for designing intervention programs to treat Internet addiction. The different patterns of findings yielded by the cross-sectional and prospective model testing provide further support for the value of adopting more than one time point in Internet addiction research.

Footnotes

Acknowledgments

Preparation of this manuscript was supported by the Hong Kong Research Grants Council's General Research Fund (17400714) and Seed Funding Program for Basic Research (201411159152) to Cecilia Cheng; and the National Social Science Foundation of China Research Grant (15CSH052) and Priority Academic Program Development of Jiangsu Higher Education Institutions to Peizhen Sun.

Author Disclosure Statement

No competing financial interests exist.

References

Mak

, Lai

, Watanabe

, et al. Epidemiology of Internet behaviors and addiction among adolescents in six Asian countries. Cyberpsychology, Behavior, & Social Networking, 2014; 17:720–728.

Cheng

, Li

. Internet addiction prevalence and quality of (real) life: a meta-analytic review of 31 nations across seven world regions. Cyberpsychology, Behavior, & Social Networking, 2014; 17:755–760.

Al‐Gamal

, Alzayyat

, Ahmad

. Prevalence of Internet addiction and its association with psychological distress and coping strategies among university students in Jordan. Perspectives in Psychiatric Care, 2015; doi: 10.1111/ppc.12102.

Burnay

, Billieux

, Blairy

, Larøi

. Which psychological factors influence Internet addiction? Evidence through an integrative model. Computers in Human Behavior, 2015; 43:28–34.

Floros

, Siomos

, Stogiannidou

, et al. The relationship between personality, defense styles, Internet addiction disorder, and psychopathology in college students. Cyberpsychology, Behavior, & Social Networking, 2014; 17:672–676.

, Zhang

, Tsang

, et al. The association between Internet addiction and psychiatric co-morbidity: a meta-analysis. BMC Psychiatry, 2014; 14:183.

Caplan

. A social skill account of problematic Internet use. Journal of Communication, 2005; 55:721–736.

Kraut

, Kiesler

, Boneva

, et al. Internet paradox revisited. Journal of Social Issues, 2002; 58:49–74.

Valkenburg

, Peter

. Online communication and adolescent well-being: testing the stimulation versus the displacement hypothesis. Journal of Computer-Mediated Communication, 2007; 12:1169–1182.

10.

Van der Aa

, Overbeek

, Engels

, et al. Daily and compulsive Internet use and well-being in adolescence: a diathesis-stress model based on Big Five personality traits. Journal of Youth & Adolescence, 2009; 38:765–76.

11.

Tang

, Yu

, Du

, et al. Prevalence of Internet addiction and its association with stressful life events and psychological symptoms among adolescent Internet users. Addictive Behaviors, 2014; 39:744–747.

12.

Yan

, Li

, Sui

. The relationship between recent stressful life events, personality traits, perceived family functioning and Internet addiction among college students. Stress & Health, 2014; 30:3–11.

13.

Lazarus

, Folkman

. Transactional theory and research on emotions and coping. European Journal of Personality, 1987; 1:141–169.

14.

Kwon

J-H

, Chung

C-S

, Lee

. The effects of escape from self and interpersonal relationship on the pathological use of Internet games. Community Mental Health Journal, 2011; 47:113–121.

15.

Boellstorff

. (2008) Coming of age in Second Life: an anthropologist explores the virtually human. Princeton, NJ: Princeton University.

16.

Snodgrass

, Lacy

, Dengah Ii

HJF

, et al. A vacation from your mind: problematic online gaming is a stress response. Computers in Human Behavior, 2014; 38:248–260.

17.

Holahan

, Moos

, Schaefer

. (1996) Coping, stress resistance, and growth: conceptualizing adaptive functioning. In Zeidner

, Endler

, eds. Handbook of coping: theory, research, applications. New York: Wiley, pp. 24–43.

18.

Vaillant

. Ego mechanisms of defense and personality psychopathology. Journal of Abnormal Psychology, 1994; 103:44–50.

19.

Lewis

, Hove

, Whiteside

, et al. Fitting in and feeling fine: conformity and coping motives as mediators of the relationship between social anxiety and problematic drinking. Psychology of Addictive Behaviors, 2008; 22:58–67.

20.

Cheng

, Lau

, Chan

. Coping flexibility and psychological adjustment to stressful life changes: a meta-analytic review. Psychological Bulletin, 2014; 140:1582–1607.

21.

Dong

, Hu

, Lin

, et al. What makes Internet addicts continue playing online even when faced by severe negative consequences? Possible explanations from an fMRI study. Biological Psychology, 2013; 94:282–289.

22.

Cheng

, Cheung

MWL

. Cognitive processes underlying coping flexibility: differentiation and integration. Journal of Personality, 2005; 73:859–886.

23.

Cheng

. Dialectical thinking and coping flexibility: a multimethod approach. Journal of Personality, 2009; 77:471–494.

24.

Seligman

. (1975) Helplessness: on depression, development and death. San Francisco, CA: Freeman.

25.

Burton

, Yan

, Pat‐Horenczyk

, et al. Coping flexibility and complicated grief: a comparison of American and Chinese samples. Depression & Anxiety, 2012; 29:16–22.

26.

Gan

, Shang

, Zhang

. Coping flexibility and locus of control as predictors of burnout among Chinese college students. Social Behavior & Personality, 2007; 35:1087–1098.

27.

Young

. Internet addiction: the emergence of a new clinical disorder. CyberPsychology & Behavior, 1998; 1:237–244.

28.

Davis

, Flett

, Besser

. Validation of a new scale for measuring problematic Internet use: Implications for pre-employment screening. CyberPsychology & Behavior, 2002; 5:331–345.

29.

Carver

. You want to measure coping but your protocol's too long: consider the Brief COPE. International Journal of Behavioral Medicine, 1997; 4:92–100.

30.

Kato

. Development of the Coping Flexibility Scale: evidence for the coping flexibility hypothesis. Journal of Counseling Psychology, 2012; 59:262–273.

31.

Russell

, Cutrona

, Rose

, et al. Social and emotional loneliness: an examination of Weiss's typology of loneliness. Journal of Personality & Social Psychology, 1984; 46:1313–1321.

32.

Leary

. Social anxiousness: the construct and its measurement. Journal of Personality Assessment, 1983; 47:66–75.

33.

Watson

, Clark

, Tellegen

. Development and validation of brief measures of positive and negative affect: the PANAS scales. Journal of Personality & Social Psychology, 1988; 54:1063–1070.

34.

Yang

. (1997) Measures of personality and social psychological attitudes. Vol. 1. Taipei, Taiwan: Yuan-Liou. [in Chinese].

35.

Lai

, Mak

, Watanabe

, et al. Psychometric properties of the Internet addiction test in Chinese adolescents. Journal of Pediatric Psychology, 2013; 38:794–807.

36.

Byrne

. (2006) Structural equation modeling with EQS: basic concepts, applications, and programming. 2nd ed. Mahwah, NJ: Lawrence Erlbaum.

37.

Satorra

, Bentler

. Ensuring positiveness of the scaled difference chi-square test statistic. Psychometrika, 2010; 75:243–248.

38.

MacKinnon

, Lockwood

, Williams

. Confidence limits for the indirect effect: distribution of the product and resampling methods. Multivariate Behavioral Research, 2004; 39:99–128.

39.

Shrout

, Bolger

. Mediation in experimental and nonexperimental studies: new procedures and recommendations. Psychological Methods, 2002; 7:422–445.

40.

Dong

, Lin

, Zhou

, et al. Cognitive flexibility in Internet addicts: FMRI evidence from difficult-to-easy and easy-to-difficult switching situations. Addictive Behaviors, 2014; 39:677–683.

41.

Peterson

, Maier

, Seligman

MEP

. (1993) Learned helplessness: a theory for the age of personal control. New York: Oxford University.

42.

Cheng

, Chan

, Chio

, et al. Being active or flexible? Role of control coping on quality of life among patients with gastrointestinal cancer. Psycho-Oncology, 2012; 21:211–218.

43.

Gan

, Zhang

, Wang

, et al. The coping flexibility of neurasthenia and depressive patients. Personality & Individual Differences, 2006; 40:859–871.

44.

Bonanno

, Burton

. Regulatory flexibility: an individual differences perspective on coping and emotion regulation. Perspectives on Psychological Science, 2013; 8:591–612.

45.

Cheng

. Cognitive and motivational processes underlying coping flexibility: a dual-process model. Journal of Personality & Social Psychology, 2003; 84:425–438.

46.

Cheng

, Kogan

, Chio

. The effectiveness of a new, coping flexibility intervention as compared with a cognitive–behavioural intervention in managing work stress. Work & Stress, 2012; 26:272–288.

47.

Dayan

, Bernard

, Olliac

, et al. Adolescent brain development, risk-taking and vulnerability to addiction. Journal of Physiology, 2010; 104:279–286.

48.

, Zhang

, Lu

, et al. Internet addiction among elementary and middle school students in China: a nationally representative sample study. Cyberpsychology, Behavior, & Social Networking, 2014; 17:111–116.

49.

Wong

PWC

, Li

TMH

, Chan

, et al. The prevalence and correlates of severe social withdrawal (hikikomori) in Hong Kong: a cross-sectional telephone-based survey study. International Journal of Social Psychiatry, 2015; 61:330–342.