The Effects of Social Environments on Time Spent Gaming: Focusing on the Effects of Communities and Neighborhoods

Abstract

This study examined the impact of community and neighborhood on time spent computer gaming. Computer gaming for over 20 hours a week was set as the cutoff line for “engaged use” of computer games. For the analysis, this study analyzed data for about 1,800 subjects who participated in the Korean Children and Youth Panel Survey. The main findings are as follows: first, structural community characteristics and neighborhood social capital affected the engaged use of computer games. Second, adolescents who reside in regions with a higher divorce rate or higher residential mobility were likely to exhibit engaged use of computer games. Third, adolescents who highly perceive neighborhood social capital exhibited lower possibility of engaged use of computer games. Based on these findings, practical implications and directions for further study are suggested.

Introduction

In 2016, the game market worldwide was worth $99.6 billion, and there were 2.2 billion game players.¹ Adolescents form a greater portion of game players compared with other age groups.^2,3 This study adopted “computer game” as a term to refer to games on various platforms such as video game (console game), computer online game, mobile game, and virtual reality (VR) game. Studies on the computer game use among adolescents have been implemented and many of them deal with the addiction to it.^4,5 This study considered that there is no standardized concept or tool to measure addiction to computer gaming and focused instead on time spent computer gaming. Although time spent gaming may not in itself adequately identify problematic use,⁶ those who meet the criteria of internet gaming disorder reportedly use computer games more than other groups do and play games longer each time.^7,8 In the sense that time spent gaming is not only observable but also measurable, it would be an important indicator for the prevention of and intervention in problematic use of computer games.

Prior studies on computer games mainly focus on the problematic use of or addiction to it. Therefore, their findings may not be applied precisely to this study examining factors affecting time spent gaming. Even with this limitation, they suggest that psychological characteristics (e.g., extraversion, empathy), sociodemographic characteristics (e.g., gender, maternal employment), or proximal relationship (e.g., parent–child relationship, affiliation with deviant peers)^5,9 would also explain time spent gaming. However, from the socioecological perspective,¹⁰ it can be said that previous studies mainly dealt with limited parts of environments surrounding game users.

The socioecological model¹⁰ explains that the development of human beings is affected by their environments. Environments consist of different kinds of levels: microsystems, mesosystems, exosystems, and macrosystems. Microsystems are applied to the immediate physical and social environment of individuals; mesosystems are made up of interactions between systems; exosystems correspond to broader social, political, and economic conditions that affect not only the structure and availability of microsystems but also the way microsystems function even though they do not interact directly with individuals; and macrosystems refer to general beliefs and attitudes that affect exosystems.

Reflecting the limitations of preceding studies, this study examined mesosystems and exosystems that affect adolescents' time spent gaming and especially considered the influence of neighborhood and communities. Previous studies show that neighborhood and communities affect adolescents' problematic behaviors,¹¹ teenage pregnancies,¹² drinking,¹³ suicide,¹⁴ and depression, and anxiety.¹⁵ Even though there are controversies whether the use of computer game is related to the development of adolescents in negative or positive ways,^16,17 we can expect that neighborhood and communities would affect the use of computer game, especially time spent gaming, in certain ways.

As mechanisms through which neighborhoods and communities affect time spent playing computer games, this study focused on the neighborhood social capital and the structural characteristics of communities. Social capital refers to insubstantial resources produced in “relations among persons that facilitate action.”¹⁸ It is expected that there are differences in the interpersonal relationships generated in the neighborhoods, and they depend on the structural characteristics of communities surrounding them. This study examined whether the structural characteristics of community and neighborhood social capital affect adolescents' time spent gaming and their effects were still valid after controlling for psychological characteristics, sociodemographic characteristics, and proximal relationship.

Research Methods

Sample

This study used the Korean Children and Youth Panel Survey (KCYPS), which has been conducted annually since 2010 by the National Youth Policy Institute (www.nypi.re.kr). The sample was obtained through a multistage, stratified, nationwide sampling of children and adolescents, and consisted of three panels: elementary first year, elementary fourth year, and middle school first year. Among them, we used the fifth-year survey of the middle school first-year panel (i.e., the 2014 survey, when middle school students became sophomores in high school; 2,342 adolescents from 230 communities). Given the nature of the data as a multilevel, we excluded adolescents who resided in communities of less than 10 KCYPS adolescents to increase validity. This procedure left us with a total of 1,806 adolescents from 182 communities for our analysis.

Measures

Dependent variable: the engaged use of computer games

Previous research distinguished behaviors beyond the time of simple computer game usage in various ways. For instance, there are big gaps between engaged players and addicted players in terms of game time, such that engaged players played around 20 hours and addicted player played around 30 hours in a week.^8,19 Prior studies suggest that game usage of 20 hours or so can indicate the borderline of the engaged use. Hence, this study measured whether users played over 20 hours in a week; if so, this was called “engaged use.”

The KCYPS asks adolescents how many hours they spend on playing games on computers or other electronic devices (e.g., console game) per day on weekdays and weekends. We calculated time spent on gaming per week and applied the cutoff line of 20 hours. It was found that a total of 147 adolescents (8.14 percent) played more than 20 hours a week.

Independent variable

Neighborhood social capital

The KCYPS modified the ADD Health survey²⁰ to fit the Korean situation, and explored and measured neighborhood social capital. Five questions were used, such as “I usually feel safe in my neighborhood” and “I would continue to live in this community.” A four-point scale for each item, from “very likely” to “not at all,” was used. We created an index of neighborhood social capital (Cronbach's α = 0.71), after recoding and adding them to generate neighborhood social capital, with higher scores indicating a higher positive level.

Structural community characteristics

Community in this study corresponds to a local autonomy government, and as of 2016, there exist 243 local autonomy governments in Korea. As the structural characteristics of local society, previous studies mainly deal with poverty, ethnic heterogeneity, and residential mobility. Therefore, we included them in the analysis as follows: the percent of welfare recipients, percent of foreign population, and percent of moving population. The moving population was calculated by adding the moving-in and moving-out populations. In addition, the percent of children in the population was added, considering that the ratio of children younger than 18 years old to the total population would determine child- and adolescent-related policies or infrastructures. Moreover, we added the variable of total population to reflect survey results that the percent of those with problematic use of computer games differed according to the size of the local autonomy government where the respondents resided.²¹

We also included crude divorce rates because (a) divorce or separation increases residential mobility,²² (b) the absence of adult supervision and monitoring⁵ affects not only adolescents in the household but also those in the proximal neighborhood through the lack of collective supervision and monitoring, and (c) the proportion of single-parent families is one of the structural characteristics that weakens the network of local society.¹⁸ Data for structural community characteristics were obtained from the Korean Statistical Information Service (http://kosis.kr).

Control variables

Control variables identified by previous studies^4,5,21,23,24 as factors that might influence adolescents' game usage were added as follows: gender, dual-income couple or not, intact family or not, annual household income (log, one unit equals to about US$10), parenting behaviors, adolescents' emotional problems, relationship with peers, and relationships with teachers. The KCYPS originally measured parenting behaviors as supervision (two items), affection (four items), overexpectation (four items), interruption (three items), and rational explanation (nine items). We used original items from the KCYPS as it is, because these items measured different dimensions of each parenting behavior and failed to be summarized into one or two factors. Cronbach's alphas were 0.83, 0.81, 0.68, 0.71, and 0.78, respectively.

For emotional problems, 36 items were used to measure five dimensions (attention, aggression, somatic symptoms, withdrawal, and depression), and the five dimensions appeared as one factor based on factor analysis (Cronbach's α = 0.94). In the KCYPS, relationships with peers and relationships with teachers were measured using five items each, such as “I get on with my classmates” and “I feel comfortable talking to my teacher” (Cronbach's αs = 0.68 and 0.79, respectively).

Analysis

Analyses of this research were implemented based on the following steps. First, a descriptive analysis was performed for independent and control variables. Second, a multilevel logit regression was carried out to investigate how community structural characteristics influence the engaged use of computer games. Third, to examine how neighborhood social capital influences the engaged use of computer games, a multilevel logit regression was implemented by adding neighborhood social capital into the previous regression model. Fourth, we also added control variables into the previous model to see whether neighborhoods and communities exert influences, controlling for sociodemographic characteristics and relationship quality. There arose no multicollinearity problems among the variables in the models. We performed multilevel analyses, considering that, based on the result of a likelihood-ratio test, a multilevel logit regression model was more suitable for the data than a standard logit regression model. Stata S.E. 13.0 was used for all the analyses.

Results

Descriptive analysis

Table 1 shows results of descriptive analysis of independent variables, such as structural community characteristics and neighborhood social capital. There were large differences across 182 communities, especially in the percent of welfare recipients and foreign population. The average of neighborhood social capital perceived by individual adolescents was set as the middle of the range. Table 2 shows the descriptive statistics of the control variables. Sixty-four percent of adolescents came from dual-income-couple households and 84 percent came from intact families.

Table 1.

Descriptive Analysis of Independent Variables

	Mean	SD	Min.	Max.
Structural community characteristics (n = 182)
Total population	217,056.6	222,539.7	10,264	1,174,228
Percent of child population (out of 100)	19.10	3.70	10.34	29.28
Percent of welfare recipients (out of 100)	3.29	1.57	0.80	14.64
Percent of foreign population (out of 100)	2.03	1.71	0.31	10.22
Crude divorce rates (out of 1,000)	2.21	0.34	1.07	3.14
Percent of moving population (out of 100)	27.11	5.18	16.25	39.55
Neighborhood social capital (n = 1,806)	16.39	2.91	6	24

SD, standard deviation.

Table 2.

Descriptive Analysis of Control Variables (n = 1,806)

Variables (range)	Mean or n	SD or %	Variables (range)	Mean or n	SD or %
Male (1/0)	923	51.11	Parenting behaviors
Both parents working (1/0)	1,159	64.17	Supervision (0–6)	4.45	1.23
Intact family (1/0)	1,517	84.00	Affection (0–12)	7.97	2.17
Annual family income, in about US$10 (log; 4.61–9.90)	8.30	0.58	Overexpectation (0–12)	5.80	2.24
Emotional problems (0–101)	37.51	15.20	Interruption (0–9)	3.61	1.94
Relationship with peers (0–12)	8.59	1.58	Rational explanation (0–9)	5.29	1.74
Relationship with teachers (0–15)	9.48	2.83

Social environments explaining the engaged use of computer games

Table 3 shows how structural community characteristics and neighborhood social capital influence the engaged use of computer games. In model 1, where structural community characteristics were put in, we could see that the higher the crude divorce rates (B = 1.03, p < 0.01), the higher the percent of the moving population (B = 0.05, p < 0.10), the higher the log odds of the engaged use of computer games. The variable of neighborhood social capital was added in model 2. The result shows that even if adolescents reside in communities with the same structural characteristics, adolescents with high perception of neighborhood social capital were less likely to be engaged players of computer games (B = −0.08, p < 0.01).

Table 3.

Multilevel Logit Regression: Social Environments Explaining the Engaged Use of Computer Games

	Model 1	Model 2	Model 3
	B (SE)	B (SE)	B (SE)
Structural community characteristics
Total population (in 10,000)	−0.002 (0.004)	−0.01 (0.004)	−0.005 (0.005)
Percent of child population	0.06 (0.04)	0.02 (0.04)	0.003 (0.04)
Percent of welfare recipients	0.01 (0.11)	−0.03 (0.11)	−0.10 (0.12)
Percent of foreign population	0.01 (0.07)	−0.02 (0.07)	−0.02 (0.08)
Crude divorce rates	1.03 (0.33)^**	0.95 (0.33)^**	0.95 (0.39)^*
Percent of moving population	0.05 (0.03)⁺	0.04 (0.03)⁺	0.06 (0.03)^*
Neighborhood social capital	—	−0.08 (0.03)^**	−0.08 (0.03)^*
Control
Male	—	—	1.15 (0.22)^***
Both parents working	—	—	0.03 (0.19)
Intact family	—	—	−0.45 (0.24)⁺
Family income (log)	—	—	−0.42 (0.16)^*
Parenting: supervision	—	—	−0.09 (0.08)
Parenting: affection	—	—	0.05 (0.06)
Parenting: overexpectation	—	—	−0.02 (0.05)
Parenting: interruption	—	—	0.03 (0.06)
Parenting: rational explanation	—	—	0.05 (0.07)
Emotional problems	—	—	0.01 (0.01)
Relationship with peers	—	—	−0.09 (0.06)
Relationship with teachers	—	—	0.01 (0.04)
Constant	−7.23 (1.47)	−4.78 (1.54)	−2.48 (2.35)
Community: var(_cons)	0.16 (0.12)	0.13 (0.11)	0.18 (0.15)
Wald χ²	18.13^**	22.29^**	74.15^***

p < 0.05; ^**p < 0.01; ^***p < 0.001; ⁺p < 0.10.

In model 3, we added control variables that were believed to be associated with use of computer games. In the case of adolescents with similar characteristics in terms of demographic, psychological, and relational perspectives, the log odds of engaged use of computer game got higher for adolescents residing in areas with higher divorce rates or higher mobility, or those perceiving neighborhood social capital being low (B = 0.95, p < 0.05; B = 0.06, p < 0.05; B = −0.08, p < 0.05, respectively).

Discussion

This study examined the influence of community and neighborhoods on time spent computer gaming. Computer gaming for over 20 hours a week was set as the cutoff line and was called “engaged use” of computer games. The following summarizes the findings and implications of an analysis of the influence of community and neighborhood on the engaged use of computer games.

First, structural community characteristics and neighborhood social capital had an influence on the engaged use of computer games. The influence of community and neighborhood was significant even after controlling for the microsystems discussed in preceding studies⁴ and some mesosystems that include schools and peer relationships.²⁴ This shows that distal environments, such as the community and neighborhood, influence human behaviors related to the use of computer games as much as proximal environments do.

Second, adolescents that reside in regions with higher divorce rates or residential mobility had a higher probability of engaged use of computer games. This shows that when there are two adolescents with the same characteristics based on the controlled variables (e.g., demographics), the one residing in a region with a higher divorce rate or residential mobility is more likely to play computer games over 20 hours a week compared with the other not residing in such a region. In addition, research shows that structural community characteristics have an influence on the engaged use of computer games even controlling for neighborhood social capital. This means that structural community characteristics not only have an indirect influence through neighborhood social capital but also have direct influence on the engaged use of computer games.

The direct effects of structural community characteristics can be explained as follows. Prior studies that studied adolescents on the individual level²³ mentioned that the adolescents of single-parent families are highly likely to exhibit problematic use of computer games due to the lack of adult supervision and monitoring. Even though they deal with problematic use rather than time spent gaming, current gaming culture, where multiple players gather together or connect online, makes it predictable that adolescents residing in the communities lacking general supervision and monitoring are likely to be engaged players. Due to maternal employment and change of family structure, providing care for younger children is considered a social responsibility; however, there is no agreement on the idea that society should be responsible for the supervision and monitoring of adolescents. Even with positive functions of computer games, such as increasing life satisfaction,²⁵ we need to train youth to exercise self-control for the proper use of computer games. In addition, social services that can provide adult supervision and monitoring for after-school activities, including the use of computer games, need to be provided.

Third, adolescents who highly perceived neighborhood social capital were less likely to play computer games over 20 hours a week. The findings of this study can be interpreted as follows; divorce rates and residential mobility significantly affected the engaged use of computer games through a partial mediation of neighborhood social capital. Local communities with high divorce rates are known to have a weak neighborhood social network because single parents do not have enough time and energy to participate in community activities or build solidarity with community members.²⁶ Considering that the social network is one of the factors composing social capital and that neighborhood social capital provides collective supervision and monitoring, we can understand that adolescents residing in the communities with higher divorce rates tend to have a higher possibility of being engaged players, due to lack of neighborhood social capital.

This study contributes to the research in the field of computer game in the sense that it analyzed the distal environments of neighborhood and community beyond prior studies that focused on the demographics or social relationships as factors affecting computer game use. For example, it shows that regions with high divorce rates or residential mobility should be considered first when trying to establish systems for social service or expand services to guide proper use of computer games. Also, it would be effective to enhance social capital through neighborhood solidarity when developing programs or services for adolescents' use of computer games.

Despite these contributions, this study has the following limitations. First, it is necessary to apply caution to the findings, since the time spent computer gaming itself does not determine problematic use or addiction. Second, the cutoff line for engaged use of computer games was set at 20 hours a week based on prior studies, but it has yet to be determined if 20 hours defines excessive use of computer game. Third, the time spent computer gaming mentioned in this study does not separate time spent on computer online games and mobile games. Considering that using computers as a platform for game raises the risk of problematic use or addiction,²¹ the choice of platform to play games would provide more critical information. In this study, however, it was impossible to apply the time of use on each platform due to the limitation of resources provided by the KCYPS. Fourth, game genre was not included in the study since it was not collected in the panel. Given that certain genres (e.g., Massively Multiplayer Online Role-Playing Games) are more likely to lead to the excessive use than other genres,²⁷ it needs to be considered in the future study.

Footnotes

Acknowledgments

This research was supported by Post-Doctor Research Program (2014) through Incheon National University (INU), Incheon, Republic of Korea.

Author Disclosure Statement

No competing financial interests exist.

References

Newzoo. (2017) Global games market 2016. http://resources.newzoo.com/global-games-market-report (accessed December 1, 2017).

KISDI. (2014) KISDI STAT report 14-04-01. Korea: Korea Information Society Development Institute. http://m.kisdi.re.kr/mobile/news/news_view.m?seq=29398 (accessed February 18, 2018).

Lenhart

, Kahne

, Middaugh

, et al. (2008) Teens, video games, and civic. Pew Research Center. www.pewinternet.org/2008/09/16/teens-video-games-and-civics (accessed February 18, 2018).

Griffiths

, Kuss

, King

. Video game addiction: past, present and future. Current Psychiatry Reviews, 2012; 8:308–318.

Toker

, Baturay

. Antecedents and consequences of game addiction. Computers in Human Behavior, 2016; 55:668–679.

Pawlikowski

, Altstotter-Gleich

, Brand

. Validation and psychometric properties of a short version of Young's Internet Addiction Test. Computers in Human Behavior, 2013; 29:1212–1223.

Bargeron

, Hormes

. Psychosocial correlates of internet gaming disorder: psychopathology, life satisfaction, and impulsivity. Computers in Human Behavior, 2017; 68:388–394.

Lehenbauer-Baum

, Fohringer

. Towards classification criteria for Internet Gaming Disorder: debunking differences between addiction and high engagement in a German sample of World of Warcraft players. Computers in Human Behavior, 2015; 45:345–351.

Collins

, Freeman

. Do problematic and non-problematic video game players differ in extraversion, trait empathy, social capital and prosocial tendencies?. Computers in Human Behavior, 2013; 29:1933–1940.

10.

Bronfenbrenner

. (1979) The ecology of human development: experiments by nature and design. Cambridge, MA: Harvard University Press.

11.

Kalff

, Kroes

, Vles

JSH

, et al. Neighbourhood level and individual level SES effects on child problem behaviour: a multilevel analysis. Journal of Epidemiology and Community Health, 2001; 55:246–250.

12.

Sellström

, Bremberg

. The significance of neighborhood context to child and adolescent health and well-being: a systematic review of multilevel studies. Scandinavian Journal of Public Health, 2006; 34:544–554.

13.

Bendtsen

, Damsgaard

, Tolstrup

, et al. Adolescent alcohol use reflects community-level alcohol consumption irrespective of parental drinking. Journal of Adolescent Health, 2013; 53:368–373.

14.

Thorlindsson

, Bernburg

. Community structural instability, anomie, imitation and adolescent suicidal behavior. Journal of Adolescence, 2009; 32:233–245.

15.

Aneshensel

, Sucoff

. The neighborhood context of adolescent mental health. Journal of Health and Social Behavior, 1996; 37:293–310.

16.

Ferguson

. Do Angry Birds make for angry children? A meta-analysis of video game influence on children's and adolescents' aggression, mental health, prosocial behavior, and academic performance. Perspectives on Psychological Science, 2015; 10:646–666.

17.

Shokouhi-Moqhaddam

, Khezri-Moghadam

, Javanmard

, et al. A study of the correlation between computer games and adolescent behavioral problems. Addiction & Health, 2013; 5:43–50.

18.

Coleman

. Social capital in the creation of human capital. American Journal of Sociology, 1988; 94:95–120.

19.

Charlton

, Danforth

IDW

. Distinguishing addiction and high engagement in the context of online game playing. Computers in Human Behavior, 2007; 23:1531–1548.

20.

Carolina Population Center (2009) Add Health. Chapel Hill, NC: Carolina Population Center. www.cpc.unc.edu/projects/addhealth (accessed December 1, 2017).

21.

Jun

, Kim

, Lee

. (2016) Report on the excess immersion in computer games 2015. Kocca. 15–52. Korea Creative Content Agency. www.kocca.kr/cop/bbs/view/B0000147/1828050.do?menuNo=200904 (accessed February 18, 2018).

22.

Feijten

, van Ham

. Residential mobility and migration of the divorced and separated. Demographic Research, 2007; 17:623–654.

23.

Rehbein

, Baier

. Family-, media-, and school-related risk factors of video game addiction: a 5-year longitudinal study. Journal of Media Psychology: Theories, Methods, and Applications, 2013; 25:118–128.

24.

, Turel

, Yuan

. Online game addiction among adolescents: motivation and prevention factors. European Journal of Information Systems, 2012; 21:321–340.

25.

Przybylski

. Electronic gaming and psychosocial adjustment. Pedatrics, 2014; 134:1–7.

26.

Knoester

, Haynie

. Community context, social integration into family, and youth violence. Journal of Marriage and Family, 2005; 67:767–780.

27.

Kuss

. Internet gaming addiction: current perspectives. Psychology Research and Behavior Management, 2013; 6:125–137.