Compensatory Internet Use Model
Motives to Play
The Present Study
Method
Study Selection
Sstudy | Sample size | Gender distribution (%) | Age range and mean age (sd) | Sample characteristics | Operationalization of IGD | Operationalization of escape/avoidance |
---|---|---|---|---|---|---|
Király et al. (2015) | 3186 | 89.74% male | Range = 14–54; Mage = 21.1 (5.9) | Hungarian MMORPGs1 and MOFPSs2 players | Problematic Online Gaming Questionnaire | Motives for Online Gaming Questionnaire |
Bányai et al. (2019) | 4284 | 89.89% male | Range = 14–58; Mage = 23.08 (6.57) | Esports3 and recreational Hungarian speaker gamers | Ten-item Internet Gaming Disorder Test (IGDT-10) | Motives for Online Gaming Questionnaire |
Schneider et al. (2018) | 823 | 49% male | Range = 12–19; Mage = 14.2 (1.4) | Australian school students | Internet Gaming Disorder Checklist | BriefCOPE4 |
Šporčić and Glavak-Tkalić (2018) | 509 | N/R | Range = 18–40; Mage = 23.14 (4.66) | Croatian FPS5 games, role-playing games, adventures, MOBA6, and MMORPG1 games | Internet Gaming Disorder Scale | Motives for Online Gaming Questionnaire |
Loton et al. (2016) | 552 | N/R | Range = 18–59; Mage = 24.90 (7.28) | Fluent English speaker gamers | Addiction Engagement Questionnaire | Brief Approach-Avoidance Questionnaire |
Milani et al. (2018) | 612 | 47.2% male | Range = 9–19; Mage = 13.94 (2.44) | Italian school students | Video Game Addiction Questionnaire Internet Addiction Test | Children’s Coping Strategies Checklist-Revised1 |
Li et al. (2016) | 651 | 46.6% male | Range = 18–22; Mage = 20.29 (1.39) | Chinese university students: 31 were identified as online game addicts and 623 were non-addicts | Internet Addiction Scale Online Game Cognitive Addiction Scale | Coping Style Questionnaire |
Biegun et al. (2020) | 651 | 47.2% male | Range = N/R; Mage = 21 (N/R) | Canadian University Students | Video Gaming Test | Motives for Online Video Game Questionnaire |
Maroney et al. (2019) | 2261 | 88.6% male | Range = 18–64; Mage = 23.78 (5.47) | North and South American, European, African, Australasian FPS5, and MMRPG1 players | The Problematic Video Game Playing Test | Video Game Motivation Scale- Escapism Subscale The Video Game Uses and Gratification Instrument (VGUGS |
von der Heiden et al. (2019) | 2734 | 88.5% | Range = 13–65; Mage = 23.06 (5.91) | German gamers | Scale for Assessment of Internet and Computer Game Addiction | BriefCOPE4 |
Rosenkranz et al. (2017) | 5667 | 50.3% male | Range = 12–19; Mage = 15.44 (1.73) | German Internet Gamers and Generalized Internet Users | Scale for Assessment of Internet and Computer Game Addiction | BriefCOPE4 |
702 | 89% male | Range = 14–60; Mage = N/R | English speaker WoW7 gamers | Excessive online gaming was assessed by five negative outcomes of excessive online gaming that were adapted for WoW7 from previous studies | Escapism was measured with five items from Yee’s (2007) gaming motivations inventory | |
Kaczmarek and Drążkowski (2014) | 1056 | 93.3% male | Range = 12–49; Mage = 18.64 (5.12) | Polish MMORPG1 players | N/R | Motivations of Play in MMORPGs Scale |
Billieux et al. (2013) | 690 | 87.10% male | Range = 18–66; Mage = 26.22 (8.14) | French-speaking WoW7 players | Internet Addiction Test | Motivation to Play in Online Games Questionnaire |
Dreier et al. (2017) | 3967 | 54.5% male | Range = 12–18; N/R | German school students | Scale for Assessment of Internet and Computer Game Addiction | BriefCOPE4 |
Kim et al. (2016) | 3041 | 59.9% male | Range = 20–49; N/R | Hungarian, English, Italian, Norwegian, Slovenian, French, Romanian, and Korean speaker gamers | Criteria IGD DSM-5 | Motives for Online Gaming Questionnaire |
Kwon et al. (2011) | 1136 | 60.9% male | N/R; Mage = 14.01 (0.51) | Korean school students | Internet Addiction Scale | Escape from Self Scale |
Dauriat et al. (2011) | 696 | 93.1% male | Range = 13–54; Mage = 25.83 (7.36) | French, English, and Italian MMORPGs1 players | MMORPG1 Addiction Scale | The motivations to play were assessed with the question, “What importance do you give to the following aspects of the game?” in regard to 14 items |
Ramos-Diaz et al. (2018) | 821 | 98% male | N/R; Mage = 20 (3.69) | Peruvian gamers (79.3% MOBA6 gamers) | Ten-Item Internet Gaming Disorder Test (IGDT-10) | Motives for Online Gaming Questionnaire |
7757 | 49.9% male | Range = 13–18; N/R | Swedish school students (89.4% MMORPGs1 players) | N/R Seven-item screening measure | 15 affective statements, extracted from a qualitative study | |
Király et al. (2017) | 5222 | 92.6% male | Range = 14–64; Mage = 22.2 (6.4) | Hungarian gamers | Ten-Item Internet Gaming Disorder Test (IGDT-10) | Motives for Online Gaming Questionnaire |
Zhengchuan Xu et al. (2012) | 623 | 56% male | Range = 12–18; Mage = about 15 N/R | Chinese school students | Addiction Engagement Questionnaire | 17 items asses 4 motivations factors (Advancement, Mechanics, Relationship and Escapism) |
Billieux et al. (2015) | 1057 | 87.67% male | Range = 18–66; Mage = 26.01 (7.96) | French-speaking WoW7 players | Internet Addiction Test | Motivation to Play in Online Games Questionnaire |
Kirby et al. (2014) | 565 | 79.6% male | Range = 18–58; Mage = 24.6 (N/R) | English-speaking WoW7 players | WoW7 Specific Problematic Usage/Engagement Questionnaire | Motivation to Play in Online Games Questionnaire |
Results
Country of Origin
Participants’ Characteristics
Operationalization of Problematic Gaming
Operationalization of Escapism/Avoidance
Risk of Bias
Study | Main findings | Study limitations1 | Risk of bias2 |
---|---|---|---|
Király et al. (2015) | MMORPG players scored significantly higher on fantasy, escape, Problematic Online Game Questionnaire, the Global Severity Index, and recreation scale. Psychiatric symptoms are directly and indirectly (via escape and competition) negatively associated with POG3. The results of the present study suggest that in contrast to playing to escape everyday problems, gaming can also be used as an adaptive coping strategy for stress release or tension reduction without necessarily leading to problematic use. | Limited representativeness and generalizability of gamer population due to the self-reported nature of the Hungarian-only data. The gender distribution was highly imbalanced with more male gamers the female gamers. | Sample bias due to self-selecting participants and lack of a significant female sample. |
Bányai et al. (2019) | Among both esport and recreational gamer groups, escapism motive appeared to be the most powerful predictor of gaming disorder. Both recreational and esport gamers did not differ significantly in the mediation effect of escapism but the motivation itself may have different effects among different populations. | Limited representativeness and generalizability of gamer population due to the self-reported nature of the Hungarian-only data. The gender distribution was highly imbalanced with more male gamers the female gamers. The categorization of esport and recreational gamers was not supported by rigorous empirical methods. | Sample bias due to self-selecting participants and lack of a significant female sample. |
Schneider et al. (2018) | Denial and behavioral disengagement (avoidant coping strategies) were significantly positively correlated with IGD. Denial coping was a significant predictor of IGD. Adolescents who had a tendency toward cognitive and behavioral avoidance may be at greater risk of developing IGD. This suggests that games are often used to play to escape from reality and to relieve negative mood states. | Limited representativeness and generalizability of gamer population due to the self-reported nature of the Australian-only data. Limitations of the COPE measure in its assumption that individuals report a general tendency toward specific strategies, rather than adopting different coping styles more dynamically to deal with different problems. There was small number of students who were not administered the COPE questionnaire due to time constraints. | Sample bias due to self-selecting participants. Measurement bias due to limitations of the COPE. |
Šporčić and Glavak-Tkalić (2018) | Escapism, fantasy, coping, sociability, and competition were positively correlated with IGD. The highest correlation was between IGD and the Escapism motive. Escape and self-concept clarity were significant predictors of problematic online gaming. The findings suggest that excessive playing of online videogames is to avoid real-life difficulties, which are escapism motives may be related to problematic gaming. Additionally, self-concept clarity was both directly and indirectly (via escapism motive) associated with IGD. | Limited representativeness and generalizability of gamer population due to the self-reported nature of the Croatian-only data. | Sample bias due to self-selecting participants. |
Loton et al. (2016) | Individuals with higher levels of video game addiction reported a greater use of avoidance-oriented coping, resignation, and withdrawal. These strategies appeared to increase the levels of distress, thwarted goal attainment, symptoms of burnout, and negative emotions. Video game addiction and high levels of engagement were associated with more use of resignation and withdrawal coping. Findings demonstrated that coping is an important partial mediator in the relationship between video game addiction and mental health, but does not explain the entire relationship. | Limited representativeness and generalizability of gamer population due to the gamers identified themselves as problematic gamers. | Sample bias due to self-selecting participants. |
Milani et al. (2018) | Children and adolescents affected by IGD and Internet addiction (both at pre-clinical and clinical levels) were characterized by poorer quality of interpersonal relationships, poorer coping strategies, anxiety/depression, withdrawal, social problems, thought problems, attention problems, aggressive behavior, internalization, and externalization. Problematic video gamers preferred to adopt dysfunctional coping strategies as distraction and avoidance. Sub-clinical condition tended to have better coping strategies. Risk factors for gaming addiction were being male and preferring avoidant coping strategies when in stressful situations. | Limited representativeness and generalizability of gamer population due to mainly youth from dense urban areas and self-reported nature of the Italian-only data. | Sample bias due to self-selecting participants and sample selection from dense urban areas. |
Li et al. (2016) | The Coping Style Questionnaire showed that the IGD group showed a greater propensity toward Avoidance Coping Strategies than the non-IGA group, and the IGA group’s scores were significantly higher for neuroticism. Stressful life events may be an important predictor of IGA. The magnitude of the ACSs mediation effect was dependent upon neuroticism. | Limited representativeness and generalizability of Chinese college students and due to the nature of the Chinese-only data. There is diversity in students’ places of origin, major and cultural circumstances in the three universities and the possible cluster effect by university should be taken into account. | Sample bias due to different sociodemographic characteristics. |
Biegun et al. (2020) | The desire to escape the negatives of real life was the strongest predictor of problematic online videogame play. | Limited representativeness and generalizability of gamer population due to the self-reported nature of the Canadian-only data. Sample includes both average users and problematic video game users. Self-reported responses | Sample bias due to self-selecting participants. |
Maroney et al. (2019) | In FPS players, both escapism and social interaction were significantly correlated with PVGU. Both depression and loneliness had significant indirect effects on Problematic Video Game Use mediated via escapism, but their effects were not significantly mediated via social interaction. Social anxiety had a significant indirect effect on PVGU via both escapism and social interaction. For MMORPG players, both escapism and social interaction were significantly correlated with PVGU. Depression-problem gaming and social anxiety-problem gaming relationships were only partially mediated by escapism motives. | Limited representativeness and generalizability of gamer population due to the self-reported data. The gender distribution was highly imbalanced with more male gamers the female gamers. Participants also had to identify as either FPS or MMORPG players but many individuals may play both types. The measure of escapism-related gaming motivations has marginal reliability and no validity data available and it is not clear how constructs used in non-online research (e.g., loneliness) translate into the online environment. | Sample bias due to self-selecting participants and lack of a significant female sample. |
von der Heiden et al. (2019) | Behavioral disengagement (avoidance coping strategy) showed the strongest positive relation to potential video game use, followed by denial and self-distraction. Distracting oneself was clearly associated with a high level of psychological symptoms (tolerance, craving, loss of control, emotion regulation, withdrawal, and unsuccessful attempts to cut back). These gamers further reported low self-esteem and low life satisfaction, loneliness, a preference for solitude, shyness, a lack of self-efficacy and social support, and poor achievement in school. | Limited representativeness and generalizability of gamer population due to the self-selected sample in which potentially problematic video players were overrepresented and self-reported nature of the German-only data. Limitations of the COPE measure in its assumption that individuals report a general tendency toward specific strategies, rather than adopting different coping styles more dynamically to deal with different problems. | Sample bias due to self-selecting participants and lack of a significant female sample. |
Rosenkranz et al. (2017) | Problematic Internet gamers employed dysfunctional coping strategies, such as behavioral disengagement and denial, more frequently than non-problematic IGs. | Limited representativeness and generalizability of gamer population due to the self-reported nature of the German-only data. The sample of IG could also include users who would not classify themselves primarily as IG despite indicating the highest possible answer. Cutoff criterion was based upon self-reporting and might not be reliable enough for diagnosis. Limitations of the COPE measure in its assumption that individuals report a general tendency toward specific strategies, rather than adopting different coping styles more dynamically to deal with different problems. | Sample bias due to self-selecting participants. Measurement bias due to operationalization of the sample of IG and cutoff criterion for classifying problematic users. |
Kardefelt-Winther (Kardefelt-Winther 2014a) | Escapism was a positive and significant predictor of higher negative outcomes scores. Higher levels of stress were associated with negative outcomes of gaming, but this was an indirect effect explained by the escapism motive. This suggests that psychological characteristics may be more usefully explored as part of a chain of events rather than a focal point. Escapist behavior may therefore better compensate for the negative feelings stemming from a lack of social interaction offline. Social anxiety and loneliness were both significantly correlated with escapism. | Limited representativeness and generalizability of gamer population due to using only WoW player base. Author presents some reliability testing bit, no validity testing, and no exploratory or confirmatory factor analysis. | Sample bias due to self-selecting participants and lack of a significant female sample. Measurement bias due to lack of rigorously or psychometrically validated measures. |
Kardefelt-Winther (Kardefelt-Winther 2014b) | Among players with high levels of stress, high escapism resulted in more negative outcomes. Among players with low level of stress, higher escapism resulted in fewer negative outcomes. There was a strong interaction effect between stress and escapism when predicting negative outcomes among the high negative outcomes group. With low self-esteem, higher escapism resulted in more negative outcomes. With high self-esteem, higher escapism results in fewer negative outcomes. There was a strong interaction effect between self-esteem and escapism when predicting negative outcomes in the negative outcomes group. The study did not find significant interaction effects between escapism and stress or between escapism and self-esteem in the group of low negative outcomes. | Limited representativeness and generalizability of gamer population due to using only WoW player base. Author presents some reliability testing bit, no validity testing, and no exploratory or confirmatory factor analysis. | Sample bias due to self-selecting participants and lack of a significant female sample. Measurement bias due to lack of rigorously or psychometrically validated measures. |
Kaczmarek and Drążkowski (2014) | The findings suggested the specific pathways through which escapist motivations undermine the well-being of MMORPG players through depletion of offline social support. Escapism predicted more time spent in the virtual environment and a stronger belief that the MMORPG environment is as realistic as the real world. Escapism, gaming time, and realism beliefs independently predicted lower offline social support, a resource known to facilitate well-being. | Limited representativeness and generalizability of gamer population due to the self-reported nature of the Polish-only data. Lack of measurements of IGD. | Sample bias due to self-selecting participants and lack of significant female samples. |
Billieux et al. (2013) | Achievement and escapism motives are central in problematic MMORPG use. The dysfunctional use of MMORPGs most often resulted in either an uncontrolled drive to look for achievement in the game, or a maladaptive strategy used to cope with negative emotions (e.g., depression, anxiety, or boredom). The escapism motive was unrelated to actual in-game behaviors. | Limited representativeness and generalizability of gamer population due to the self-reported data. The gender distribution was highly imbalanced with more male gamers the female gamers. Considered players to have a unique main avatar, although the authors cannot exclude the possibility that some participants had more than one character that they considered as main characters. | Sample bias due to self-selecting participants and lack of significant female samples. |
Dreier et al. (2017) | IGD was related to a higher level of perceived stress and to specific coping strategies such as media-related coping, denial, self-distraction, self-blame, and behavior disengagement. Media-related coping and behavioral disengagement was predictors for IGD. There was a significant association with a large proportion of children and adolescents meeting IGD criteria being classified as “Whales”: subjects that spend the highest amounts of money and time on the game The capability of solving problematic in-game situation by spending money might be associated with the coping mechanisms of vulnerable free-to-play gamers. This opportunity supports the vulnerable gamer’s attitude to escape problems by applying media-related coping strategies. Dysfunctional coping was reported to play a key role in developing IGD. | Limited representativeness and generalizability of gamer population due to the self-reported nature of the German-only data. Limitations of the COPE measure in its assumption that individuals report a general tendency toward specific strategies, rather than adopting different coping styles more dynamically to deal with different problems. The BriefCOPE subscales had poor internal reliability: self-distraction (α = .45) and active coping (α = .49) as did the positive reframing subscale (α = .52). A validation for their adapted BriefCOPE is required. | Measurement bias due to lack of validation of BriefCOPE. |
Kim et al. (2016) | The risk group of IGD used online gaming to escape from reality or specific issues (coping, fantasy, competition) more than the healthy controls. Also, the risk group showed higher score in Motives for online gaming questionnaire in the use of online gaming as a coping mechanism compared to the healthy controls. | Limited representativeness of Korean population and limited generalizability to other countries. The diagnostic criteria of IGD in DSM-5 used in this study were based upon self-reporting. | Sample bias due to self-selecting participants. |
Kwon et al. (2011) | IGA scores were most significantly correlated with escape and negative mood. The influence exerted on IGA by the tendency to escape was more important than exerted by negative interpersonal relationships. IGA was utilized as a way to escape from oneself an increase in self-discrepancy tended to aggravate not only a negative evaluation of oneself but also a negative mood, from which an attempt to escape arose as strong means for eliminating a negative mood. The scale Hostility of Perceived Parent-Child Relationship (Parent) and Perceived Parental Supervision Scale were taken as correlated and exhibited direct effects on negative mood and Escape. | Limited representativeness and generalizability of gamer population due to the self-reported nature of the Korean-only data. | Sample bias due to self-selecting participants. |
Dauriat et al. (2011) | The results showed that achievement and escapism were the best “predictors” of the MMORPG addiction scale. | Limited representativeness and generalizability of gamer population due to the self-reported data. The gender distribution was highly imbalanced with more male gamers the female gamers. | Sample bias due to self-selecting participants and lack of significant female samples. Lack of clinical validated measures |
Ramos-Diaz et al. (2018) | Peruvian gamers who play for fantasy motives may do it because they are trying out new identities in gaming which may serve as an escape from real-life problems. Escape and fantasy were the best predictors of IGD. | Limited representativeness and generalizability of gamer population due to the self-reported nature of the Peruvian-only data. The gender distribution was highly imbalanced with more male gamers the female gamers. | Sample bias due to self-selecting participants and lack of significant female samples. |
Hellström et al. (2015) | The highest incidence for depressive, musculoskeletal, and psychosomatic symptoms were found among weekday gamers, especially those who played for high escape motives and spent an excessive amount of time gaming. | Limited representativeness and generalizability of gamer population due to the self-reported nature of the adolescent Swedish-only data. | Sample bias due to self-selecting participants. |
Hellström et al. (2012) | Gamers that played to escape and for gaming status showed a higher risk of negative consequences (such as forgot to eat, do not have time for friends, troubles with parents due to gaming, or less sleep due to gaming) and whereas those who played for fun or social motives had a reduced risk of negative consequences. Motives are more important than the time spent. | Limited representativeness and generalizability of gamer population due to the self-reported nature of the Swedish-only data. Despite the large population, the subgroups of some variables were still rather small which may have affected the result. | Sample bias due to self-selecting participants. |
Király et al. (2017) | Escape motives are more pronounced in problematic online gaming than in intense users. Escape was strongly associated with both problematic gaming and gaming time. The only substantial difference between gaming time and problematic use in term of motives was Escape which was the only motive consistently associated with problematic gaming (but not with gaming time). | Limited representativeness and generalizability of gamer population due to the self-reported nature of the Hungarian-only data. The gender distribution was highly imbalanced with more male gamers the female gamers. Instruments used to assess problematic gaming and motives were validated in convenience samples, and were based on correlations with similar self-report measures lacking clinical validation (IGDT-10) or comparison with actual in-game behaviors. | Sample bias due to self-selecting participants and lack of significant female samples. Lack of clinical validation of measures of IGD-10. |
Zhengchuan Xu et al. (2012) | Findings suggest the need for mastering game mechanics, relationships, and escapism increase online game playing and need for relationships and escapism increase online game addiction. | Limited representativeness and generalizability of gamer population due to the self-reported nature of the Chinese-only data. Considered only a limited set of functional need (need for relationship, escapism, mastering the machines, and advancement). | Sample bias due to self-selecting participants. |
Billieux et al. (2015) | Three different clusters of problematic gamers were determined. Members of the first cluster were named “unregulated escapers” and were characterized by low achievement, high escapism motives, poor self-esteem and high impulsive traits. For them, excessive use can be conceptualized as a maladaptive coping strategy serving to overcome negative affect and/or negative life events. The third cluster included individuals mainly motivated by achievement and escapism, with high level of self-esteem. For them the “virtual” self has overwhelmed the real self, implying a potential denial of real life and an over-involvement in virtual life. These results emphasize that problem online gaming is highly heterogeneous and depend on a wide range of psychological factors. | Limited representativeness and generalizability of gamer population due to the self-reported data. The gender distribution was highly imbalanced with more male gamers the female gamers. | Sample bias due to self-selecting participants and lack of significant female samples. |
Kirby et al. 2014 | Findings suggest that when gamers play for escape motive, there is a greater negative impact of playing time spent on psychological well-being. | Limited representativeness and generalizability of gamer population due to the self-reported data. The gender distribution was highly imbalanced with more male gamers the female gamers. | Sample bias due to self-selecting participants and lack of significant female samples. |