Peer Group Identification as Determinant of Youth Behavior and the Role of Perceived Social Support in Problem Gambling

Vol.:(0123456789) Journal of Gambling Studies (2019) 35:15–30 https://doi.org/10.1007/s10899-018-9813-8 1 3 ORIGINAL PAPER Peer Group Identification asDeterminant ofYouth Behavior andtheRole ofPerceived Social Support inProblem Gambling IinaSavolainen1 · AnuSirola1· MarkusKaakinen1· AtteOksanen1 Published online: 21 November 2018 © The Author(s) 2018 Abstract Gambling opportunities have increased rapidly during recent years. Previous research shows that gambling is a popular activity among youth, which may contribute to problem gambling. This study examined how social identification with online and offline peer groups associates with youth problem gambling behavior and if perceived social support buffers this relationship. Data were gathered with an online survey with 1212 American and 1200 Finnish participants between 15 and 25years of age. Measures included the South Oaks Gambling Screen for problem gambling, and items for peer group identification and perceived social support. It was found that youth who identify strongly with offline peer groups were less likely to engage in problem gambling, while strong identification with online peer groups had the opposite effect. We also found that the associations between social identification and problem gambling behavior were moderated by perceived social support. Online peer groups may be a determinant in youth problem gambling. Focusing on offline peer groups and increasing social support can hold significant potential in youth gambling prevention. Keywords Social identification· Social support· Problem gambling behavior· Youth * Iina Savolainen iina.sav[email protected] Anu Sirola [email protected] Markus Kaakinen [email protected] Atte Oksanen [email protected] 1 Faculty ofSocial Sciences, University ofTampere, 33100Tampere, Finland 16 Journal of Gambling Studies (2019) 35:15–30 1 3 Introduction Over the past decade, gambling has increased its popularity as a recreational activity (Molinaro etal. 2018; Orford 2010), particularly among individuals between 11 and 25years of age, or, youth (Calado etal. 2017; UNESCO 2017). Even though gambling is illegal for underaged youth, new gambling technologies have made different forms of gambling widespread and much easier for even the youngest individuals to access (Blinn-Pike etal. 2010; Canale etal. 2016; Elton-Marshall etal. 2016). Above all, Internet gambling has transformed the traditional gambling landscape by offering convenient, instant, and constant access to novel gambling forms (Gainsbury etal. 2015; Griffiths and Parke 2010). Among a sample of ≥ 18-year-old college student problem gamblers, Petry and GonzalezIbanez (2015) found that nearly half (49%) had gambled on the Internet during the month prior to the study.The sampleincluded onlythose studentswho hadscored more than 3 points on a combined gambling measure, consisting of the South Oaks Gambling Screen (SOGS) and money spent on gambling during the past 2 months. More recently, Molinaro etal. (2018) reported that 22.6% of 16-year-old students had gambled during the past year, 16.2% of them online. Past research further indicates that the prevalence rate of gambling engagement is considerably high among youth and predominantly focused on private betting on skill-based games (Elton-Marshall etal. 2016; Volberg etal. 2010). An extensive review study from the turn of the decade found that, compared to 11% of adults, 28% of youth reported having bet on games of skill, such as card games, in the past year (Volberg etal. 2010). A more recent review consisting of 44 studies on gambling among 11to 24-year-olds, concluded that up to 12.3% of youth within that age range qualify as problem gamblers across five continents (Calado etal. 2017). Gambling activities can provide individuals with many subjective benefits, such as excitement, entertainment, and a perceived sense of acquiring wealth without much effort (Derevensky and Gilbeau 2015; Kim etal. 2017). However, both recreational and problematic gambling alike are associated with several psychosocial, physical, and mental health problems (Fröberg etal. 2015; Kong etal. 2013). Past research has found associations between gambling engagement and substance abuse (Calado etal. 2017; Kessler etal. 2008), increased financial difficulties (Raisamo etal. 2013), and poor school performance, as well as damaged social relationships (Raisamo etal. 2013; Splevins etal. 2010). Problem gambling is a growing global issue that may further manifest in a range of mental health problems, such as depression, anxiety, mood difficulties, and aggression (Lloyd etal. 2010; Yip etal. 2011). As these issues can become increasingly prevalent when the onset of gambling occurs prior to adulthood (Kong etal. 2013), new research perspectives are needed to explain youth gambling behavior and motives, as well as other underlying mechanisms. Social Identity asaDeterminant ofBehavior Social relationships are recognized as a key determinant of overall well-being (Baumeister and Leary 1995; Thoits 2011) and behavior (Cruwys etal. 2015; Holt-Lunstad 2010) for people in general, but especially for young individuals in the 15–25 age group (Best etal. 2014a; Dishion and Tipsord 2011; Tarrant 2002). One possible linkage between social 17 Journal of Gambling Studies (2019) 35:15–30 1 3 relationships and subsequent behavior is social identification, which is operationalized as the subjective sense of belonging to a certain group (Buckingham etal. 2013; Cruwys etal. 2017; Jetten etal. 2014). Social identification, as introduced by Tajfel and Turner (1979), refers to a process in which an individual’s identity is partly determined by his or her connectedness to desired social groups. These “in-groups” provide individuals with a sense of belonging and purpose, which have been shown to have significant outcomes in terms of personal capital and guide behavioral choices beyond normative peer influence (Best etal. 2014b; Frings and Albery 2015; Mawson etal. 2015). According to the social identity theory (SIT), becoming a member of an “in-group” consisting of peers or similar others is beneficial for an individual (Tajfel and Turner 1979), as it enhances self-esteem, positive self-concept, and contributes to decision-making processes (Buckingham etal. 2013). Social psychological research has consistently recognized the impact that social identities have in shaping not only individuals’ beliefs, but various behaviors as well, ranging from positive health promotion to negative health-destructive behaviors (Jetten etal. 2012, 2014; Oyserman etal. 2007). Once social identity with a desired group is established, the individual is more motivated to behave in accordance with the groups’ perceived norms (Turner 1991; Marino etal. 2016). These social identity effects may be even more pronounced among youth who are still constructing their identities (Becht etal. 2017). Kobus (2003) found that adolescents between ages 11 and 20 were more likely to engage in smoking behavior when their peer group identity was salient. Oyserman etal. (2007) concluded that, across seven experiments, race-related social identities were associated with either health-promotion or unhealthy behavior, depending on the social group with which the participants identified. Congruently, health-promoting behaviors were associated with belonging to the white and middle-class in-group identity, while unhealthy behaviors were associated with racial minority identities. A study by Foster etal. (2014) on college student gambling found that gambling behavior was associated with a stronger social identity with other gambling students, rather than the student body in general. Social identity was also found to moderate the association between perceived college norms and gambling behavior, as gambling students were more likely to perceive it as a normative behavior among their peers (Foster etal. 2014). Peer relationships have a heightened importance among adolescents and young adults, and most adolescents report that they belong to a peer group (Chow etal. 2011; Flynn etal. 2017; Tarrant 2002). Due to the changing structure of the social world, individuals can now find meaningful groups with which to identify in both offline (Sussman etal. 2007) and online (Davis 2012; Mikal etal. 2016) environments. Ever since the introduction and exponential growth in popularity of the first social networking services, such as MySpace, Bebo, and Facebook—and, later on, Instagram, Twitter, and Snapchat—people have been connecting with ever-widening circles of other users and creating associations with individuals around the globe (Livingstone 2008; Tsitsika etal. 2014; Young 2011). Social media is particularly attractive to young users: About 90% of young adults in the United States use social media platforms and more than half of them visit the sites daily (Villanti etal. 2017; Lin etal. 2016). Research has found that social media services are used mainly for communication with friends from the past and present (Davis 2012; Neira etal. 2014), sharing thoughts and interesting content (Kaplan and Haenlein 2010), and for self-disclosure and expression (Best etal. 2014a; Livingstone 2008). Research has also suggested that through online networks, individuals can more easily come in contact with like-minded others and share mutual ideas and content (Aiello etal. 2012; Sirola etal. 2018). Despite the fact that online and offline social networks tend to overlap, online social 18 Journal of Gambling Studies (2019) 35:15–30 1 3 ties have been identified as an independent source of social connectedness that are not reducible to the ones originating from face-to-face interactions (Cole etal. 2017). Moreover, online social groups are valued as being equally important as those offline and building one’s identity through online venues and groups seems to be a growing modern norm (Borca etal. 2015; Lehdonvirta and Räsänen 2011). Given its increasing popularity, youth gambling generates a new type of question in terms of social identities and the gambling phenomenon. Past research has been able to identify several individual, familial, and contextual factors associated with youth gambling (Buckle etal. 2013; Dussault etal. 2017), yet little research has examined whether social identity functions as an implicit mechanism reducing or inducing the behavior. It has been systematically reported that youth gambling and problem gambling are associated with the male gender (Splevins etal. 2010), young age, lower education level (Gainsbury etal. 2015; Hing etal. 2017), family-induced gambling experiences (Gupta and Derevensky 1997; Volberg etal. 2010) and having a positive attitude toward gambling (Dixon etal. 2016). These common features of gamblers and problem gamblers might suggest that gambling youth are socially motivated to either participate in the behavior or assimilate with a desired peer group via the behavior. Perceived Social Support As discussed above, social identities’ impact on health and well-being is considerable. Equally important to health and well-being is the support derived from these social relationships (Cohen and Wills 1985; Dussault etal. 2016). These social mechanisms commonly go together, especially in psychology and health literature, where their impact is widely documented in patient and addiction recovery outcomes (Buckingham etal. 2013; Dingle etal. 2015a, b). For instance, Best etal. (2014a) found that rethinking identity processes allowed therapeutic community patients to recognize the social resources accessible for them while in alcohol and drug addiction. In the study, the patients explored potential transitional identities from “user” to “therapeutic community member.” Through this rethinking practice, the patients were better able to gain a sense of belonging and support. These were later associated with improved health outcomes, such as life-satisfaction and abstinence from drinking at follow-up (Best etal. 2014a). In another study, Wu etal. (2016) found that social support was significantly associated with lower levels of Internet addiction among adolescents. These results were further supported by a meta-analysis showing evidence that adolescents and young adults with low support are at a higher risk of becoming addicted to Internet use (Lei etal. 2018). In terms of gambling behavior, research suggests that lack of perceived social support is a risk factor in youth gambling. One study found that youth who were at-risk or probable pathological gamblers also reported that they felt a lack of social support (Hardoon etal. 2004). Similarly, Petry and Weiss (2009) concluded that social support received from family members and friends significantly mediated both short-term and long-term gambling outcomes among pathological gamblers. Previous research on perceived social support as a resource for recovery has further found that young individuals estimate their total personal, social, and recovery capital to be lower than that of older individuals, suggesting that these resources continue to strengthen over time and in conjunction with stronger social identities (Mawson etal. 2015). Given that social support received from meaningful groups can have significant outcomes in terms 19 Journal of Gambling Studies (2019) 35:15–30 1 3 of decision making and the consequences that follow, youth are a particularly vulnerable group to the effects of perceived low social support. To this effect, earlier research findings indicate that strong social belonging and support can buffer youths against harmful online experiences (Kaakinen etal. 2018; Minkkinen etal. 2016). Notably, it was only the offline social relations that were found to have a buffering effect, while online social relations did not. In line with this finding, offline and online social connections have been reported to correlate with online risk behavior in inverse ways; while social ties offline are linked to a decreased likelihood of riskbehavior,online relationships seem to have a reverse, increasing effect (Kaakinen etal. 2018). Considering the amplitude of harmful content and risky groups to which youth are exposed to online every day, investigating young individuals’ perceived social support capital is important and has extensive implications. Social identities, in a form of meaningful group memberships and social support, are both social psychological explanations as to why social ties are important for human well-being (Jetten etal. 2014). More specifically, it has been suggested that social identities might contribute to well-being by making social support more accessible to individuals (Haslam etal. 2005). There is, however, a gap in research literature that assesses whether the positive outcomes of social identification are actualized when there is a lack of perceived social support. This would further guide our understanding of social identity dynamics and well-being. The Current Study In the current study, youth between the ages of 15 and 25 are the population of interest and investigation, as these years include distinct developmental periods characterized by identity uncertainty and exploration (Archer 1982; Mawson etal. 2015). During these times, youth experiment with the diverse identities salient to them via different personal and social contexts online and offline. Using social identity theory as a theoretical framework, this cross-cultural study seeks to provide a supplementary explanation to youth problem gambling. The aim of this research is to investigate and compare how social identification with peer groups online and offline is associated with problem gambling behavior among American and Finnish youths. More specifically, this research seeks to determine whether social identification is related to higher social support and if this relationship may safeguard youth from engaging in problem gambling, as theory indicates. While the beneficial outcomes of social identification and social support are widely reported (Haslam etal. 2005; Jetten etal. 2014), more research is needed on whether social identity is related to reduced destructive behavior (i.e., problem gambling) if perceived social support is absent. The two countries were chosen because they are both technologically advanced Western countries while culturally diverse. Within both the United States and Finland, youth engage in gambling despite the existing gambling laws that aim to restrict such activity among underage individuals (i.e., those under 18years of age in Finland and those under 21years of age in several states of the United States). Cross-cultural research can provide deeper insight in understanding youth gambling. It is also needed to establish whether these social psychological phenomena exist in different cultural contexts. To our knowledge, no previous research has examined whether social identities of youth steer problem gambling behavior and if this possible effect is consistent across developed Western countries. Consequently, this study aims to contribute to the existing body of research by focusing on 20 Journal of Gambling Studies (2019) 35:15–30 1 3 inspecting online and offline social identities as pathways to problem gambling behavior among youth, as well as by further examining whether perceived social support can moderate this connection. Within both samples, we hypothesized that strong identification with an offline peer group is associated with lower engagement in problem gambling behavior (H1), while strong identification with an online peer group is associated with higher engagement in problem gambling behavior (H2). It was expected that high perceived social support is associated with less problem gambling behavior (H3). It was further hypothesized that perceived social support moderates the association between social identification and problem gambling behavior (H4). Method Participants Participants were recruited from a volunteer pool administered by Survey Sampling International (SSI). The samples consisted of a total of 1212 American participants aged 15–25 (M = 20.05, SD = 3.19, 50.17% female) and 1200 Finnish young people aged 15–25 (M = 21.29, SD = 2.85, 50.00% female). Both samples were demographically balanced in terms of age, gender, and living area. The AcademicEthics Committee of the Tampere Regionapproved the research before implementation. Participation in the study was fully voluntary and all respondents were informed of the aims of the study prior to participation. The participants were aware that they could withdraw from the study at any time. Participation in the study did not inflict any harm on the participants. Survey Data were collected with the YouGamble online survey conducted from March to April 2017 in Finland and in January 2018 in the United States. Both datasets were collected with LimeSurvey software by using identical survey formats. The surveys were optimized for both computers and mobile devices. The average survey response time was 14min and 49s in the United States and 15min 30s in Finland. The original survey was in Finnish. It was translated into English and back-translated again to ensure consistency and accurate matching of the survey items. Some questions were slightly modified to better fit the cultural setting in each country. The surveys were fully anonymous and included measures for all target variables, including gambling behavior, perceived social support, and identification with a primary peer group. Measures The South Oaks Gambling Screen (SOGS) was used to measure the frequency and intensity of problem gambling behavior. The SOGS is regularly used in studies to screen for pathological gambling behavior (Lesieur and Blume 1987; Salonen etal. 2017). Some of the test items were slightly modified to accommodate for cultural variations in gambling. The scale had good internal consistency in both the American (α = .90) and Finnish (α = .89) sample. The items were summed up to a continuous scale measuring the level of engagement in problem gambling behavior. Earlier studies have identified problems with the SOGS 21 Journal of Gambling Studies (2019) 35:15–30 1 3 cut-off scores which may lead to biased estimates of gambling problems and high rates of false positives (Battersby etal. 2002; Stinchfield 2002). To account for this, the SOGS was used as a continuous measure in our analyses, measuring the intensity of problem gambling behavior, instead of categorizing respondents to problem gamblers. Using the SOGS measure as a continuous variable also responds tothe methodological criticisms raised from categorizing continuous outcome variables (Altman 2014). The suggested SOGS cut-off scores; 0–2 = no problem gambling, 3–7 = at-risk gambling, and ≥ 8 = probablepathologicalgambling, were used to provide descriptive statistics (Goodie etal. 2013). Identification with a primary peer group consisting of friends or an online community was assessed with a survey item that inquired about the subjective sense of belonging to a primary peer group. The item was written as: “How strongly do you feel you belong to the following?” The group-type options provided for this inquiry included “a friendship group” and “an online community.” Answers could be provided on a scale ranging from 1 (no belonging at all) to 10 (very strong belonging). Perceived social support was assessed with a survey item asking about the support an individual receives from close ones. The item asked: “Do you feel you receive support from your close ones when you need it?” Three answer choices wereprovided for the item: 1 = Rarely, 2 = Sometimes, 3 = Often. This question was then turned into a dummy variable (0 = rarely; 1 = sometimes or often). Statistical Analysis Descriptive statistics for all continuous variables were calculated as means (M) and standard deviations (SD), and as frequencies (n) and relational proportions (%) for categorical variables. This information is presented in detail in Table1. In order to compare differences Table 1 Descriptive statistics. Continuous variables reported as means (M) and standard deviations (SD), categorical variables as frequencies (n) and relational proportions (%) The South Oaks Gambling Screen (SOGS) cut-off scores used were: no problem gambling (0–2), at risk gambling (3–7) and probablepathological gambling (≥ 8) Variable United States Finland M SD Range M SD Range Problem gambling 1.27 2.55 0–20 1.59 2.56 0–20 Identification w/offline peers 6.72 2.62 1–10 6.83 2.49 1–10 Identification w/online peers 5.38 2.69 1–10 5.04 2.61 1–10 Age 20.05 3.19 15–25 21.29 2.85 15–25 Categorical variables Coding n% Coding n% Gender Male 604 49.83 Male 600 50 Female 608 50.17 Female 600 50 Perceived social support Weak 212 17.49 Weak 112 9.33 Strong 1000 82.51 strong 1088 90.67 SOGS cut-off score 0–2 1011 83.42 0–2 946 78.83 3–7 157 12.95 3–7 210 17.50 ≥ 8 44 3.63 ≥ 8 44 3.67 22 Journal of Gambling Studies (2019) 35:15–30 1 3 between the two countries and statistically test the hypotheses, linear regression analysis was conducted. Two separate regression models were conducted for both countries to analyze the direct effects of social identification and support on problem gambling behavior (see Table2). Interaction analysis was conducted with regression analysis while treating perceived social support as a moderator. This approach was also applied separately for both countries to allow for better observation of the ways in which the effects vary by a given country. These results are reported in Table3. We conducted the moderation analysis by first testing the statistical significances of both interaction terms in the regression models. Secondly, the slope difference analysis suggested by Robinson etal. (2013) was used to analyze a difference in the independent variables’ regression coefficients over the values of the dichotomous moderator variable. The slope difference test is less conservative than interaction term testing and, thus, is less prone to Type II errors (failing to reject a false null hypothesis). In the interaction analysis, social identification variables were mean-centered in both samples to avoid multicollinearity. Gender and age were treated as controls in all models. Table 2 Main effects of the model predicting problem gambling in the United States (N = 1212) and Finland (N = 1200) Statistically significant results (p < .05) boldfaced. Gender was represented as a dummy variable with code 1 serving as the male reference group. Perceived social support measured as a dummy variable with code 1 serving asthe “strong support” reference group Variable US Finland B SE p β B SE p β Identifying w/offline peers −0.07 0.03 0.034 − 0.07 − 0.11 0.03 0.001 − 0.11 Identifying w/online peers 0.17 0.03 0.000 0.17 0.01 0.03 0.868 0.01 Perceived social support − 0.71 0.20 0.000 − 0.10 − 0.10 0.26 0.709 − 0.011 Age 0.16 0.02 0.000 0.19 0.01 0.03 0.718 0.01 Gender −0.75 0.15 0.000 −0.14 −1.22 0.15 0.000 −0.24 Table 3 Regression results of the moderation analyses for problem gambling in the United States (N = 1212) and Finland (N = 1200) Statistically significant results (p < .05) boldfaced. Gender was represented as a dummy variable with code 1 serving as the male reference group. Perceived social support measured as a dummy variable with code 1 serving as the“strong support” reference group Variable US Finland B SE p β B SE p β Identification with offline peers 0.05 0.07 0.523 0.05 0.08 0.09 0.408 0.07 Identification with online peers 0.17 0.07 0.015 0.17 0.25 0.10 0.011 0.26 Perceived social support −0.86 0.22 0.000 − 0.12 2.19 0.56 0.000 0.25 Perceived social support × identific w/offline peers −0.15 0.08 0.078 − 0.12 − 0.22 0.10 0.024 − 0.26 Perceived social support × identific w/online peers 0.00 0.08 0.984 0.00 − 0.27 0.10 0.009 − 0.31 Age 0.15 0.02 0.000 0.18 0.01 0.03 0.794 0.01 Gender − 0.75 0.15 0.000 − 0.14 − 1.18 0.15 0.000 − 0.23 23 Journal of Gambling Studies (2019) 35:15–30 1 3 Results According to our models, identification with a primary offline peer group had a significant association with lower problem gambling behavior (see Table2). This was true both in the United States (B = − .07, SE = .03, t(1211) = − 2.12, p = .034) and in Finland (B = − .11, SE = .03, t(1199) = − 3.34, p = .001). Identifying with a primary online peer group was associated with higher engagement in problem gambling behavior, but the effect was significant only among U.S. youths (B = .17, SE = .03, t(1211) = 5.68, p < .001). This finding supports our hypothesis that strong identification with online peers is associated with higher gambling behavior. High perceived social support was associated with lower problem gambling behavior, but this direct effect was also significant only among the U.S. sample (B = − .71 SE = .20, t(1211) = − 3.53, p < .001). Of our covariates, only gender was associated with engaging in problem gambling behavior in both countries, as male respondents reported significantly higher rates of problem gambling behavior than females in the U.S. (B = − .75, t(1211) = − 5.8, p < .001) and in Finland (B = − 1.22, SE = .15, t(1199) = − 8.13, p < .001). Age, in turn, was not associated with problem gambling behavior in Finland (B = .01, SE = .03, t(1199) = .26, p = .718), but was associated with it in the United States (B = .16, SE = .02, t(1211) = 6.72, p < .001). In terms of our moderation analysis (Table3), it was found that high perceived social support significantly moderated the association between engagement in problem gambling behavior and peer group identification with both offline (B = − .22, SE = .10, t(1199) = − 2.27, p = .024) and online (B = − .27, SE = .10, t(1199) = − 2.61, p = .009) peers among Finnish youths. With the U.S. sample, the interaction term between high perceived social support, social identification, and problem gambling behavior was not significant in either offline (B = − .15, SE = .08, t(1211) = − 1.76, p = .078) or online (B = .00, SE = .08, t(1211) = − .02, p = .984) peer groups. According to the slope difference test, the slope for social identification with an offline peer group differed significantly between those who report perceived social support and those who do not (F(1,1204) = 16.50, p < .001). In case of online identification, such a difference did not exist (F(1,1204) = 0.00, p = .964). Notably, identification with offline peer groups was only associated with decreased problem gambling behavior if respondents also reported at least some degree of social support. Discussion This study examined the effects of peer group identification on problem gambling behavior among American and Finnish youth. Our data suggested that both U.S. and Finnish youth identify with peers online and offline and engage in gambling activities to a similar degree. Our multivariate results, in turn, found varying but significant relationships between the examined variables within both samples. Social identification with an offline peer group was associated with lower levels of problem gambling behavior among both U.S. and Finnish youths. In the United States, identification with an online peer group was associated with higher levels of problem gambling behavior. Also, among U.S. youths, social support was associated with lower levels of problem gambling behavior. In both the U.S. and Finnish samples, perceived social support moderated the association between problem gambling behavior and social identification offline, as the negative association existed only if respondents also reported perceived social support. In the 30 Journal of Gambling Studies (2019) 35:15–30 1 3 Salonen, A. H., Rosenström, T., Edgren, R., Volberg, R., Alho, H., & Castrén, S. (2017). Dimensions of the South Oaks Gambling Screen in Finland: A cross-sectional population study. Scandinavian Journal of Psychology, 58, 228–237. https ://doi.org/10.1111/sjop.12357 . Sampasa-Kanyinga, H., & Lewis, R. F. (2015). Frequent use of social networking sites is associated with poor psychological functioning among children and adolescents. Cyberpsychology, Behavior, and Social Networking, 18, 380–385. https ://doi.org/10.1089/cyber .2015.0055. Savolainen, I., Kaakinen, M., Sirola, A., & Oksanen, A. (2018). Addictive behaviors and psychological distress among adolescents and emerging adults: A mediating role of peer-group identification. Addictive Behaviors Reports. https ://doi.org/10.1016/j.abrep .2018.03.002. Sirola, A., Kaakinen, M., & Oksanen, A. (2018). Excessive gambling and online gambling communities. Journal of Gambling Studies, 34(4), 1313–1325. https ://doi.org/10.1007/s1089 9-018-9772-0. Splevins, K., Mireskandari, S., Clayton, K., & Blaszczynski, A. (2010). Prevalence of adolescent problem gambling, related harms and help-seeking behaviours among an Australian population. Journal of Gambling Studies, 26, 189–204. https ://doi.org/10.1007/s1089 9-009-9169-1. Stinchfield, R. (2002). Reliability, validity, and classification accuracy of the South Oaks Gambling Screen (SOGS). Addictive Behaviors, 27, 1–19. https ://doi.org/10.1016/S0306 -4603(00)00158 -1. Sussman, S., Pokhrel, P., Ashmore, R. D., & Brown, B. B. (2007). Adolescent peer group identification and characteristics: A review of the literature. Addictive Behaviors, 32(8), 1602–1627. https ://doi. org/10.1016/j.addbe h.2006.11.018. Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G. Austin & S. Worchel (Eds.), The Social Psychology of Intergroup Relations (pp. 33–47). Monterey, CA: Brooks Cole. Tarrant, M. (2002). Adolescent peer groups and social identity. Social Development, 11, 110–123. https :// doi.org/10.1111/1467-9507.00189 . Thoits, P. (2011). Mechanisms linking social ties and support to physical and mental health. Journal of Health and Social Behavior, 52, 145–161. Tsitsika, A. K., Tzavela, E. C., Janikian, M., Ólafsson, K., Iordache, A., Schoenmakers, T. M., etal. (2014). Online social networking in adolescence: Patterns of use in six European countries and links with psychosocial functioning. Journal of Adolescent Health, 55, 141–147. Turner, J. C. (1991). Social influence. Pacific Grove, CA: Open University Press. United Nations Educational, Scientific and Cultural Organization. (2017). About Youth. Retrieved from http://www.unesc o.org/new/en/socia l-and-human -scien ces/theme s/youth /about -youth /. Villanti, A. C., Johnson, A. L., Ilakkuvan, V., Jacobs, M. A., Graham, A. L., & Rath, J. M. (2017). Social media use and access to digital technology in US young adults in 2016. Journal of Medical Internet Research. https ://doi.org/10.2196/jmir.7303. Volberg, R. A., Gupta, R., Griffiths, M. D., Olason, D. T., & Delfabbro, P. (2010). An international perspective on youth gambling prevalence studies. International Journal of Adolescent Medicine and Health, 22, 3–38. Westberg, K., Beverland, M. B., & Thomas, S. L. (2017). The unintended normalization of gambling: Family identity influences on the adoption of harmful consumption practices. Journal of Macromarketing, 37, 426–443. https ://doi.org/10.1177/02761 46717 72097 9. Wu, X. S., Zhang, Z. H., Zhao, F., Wang, W. J., Li, Y. F., Bi, L., etal. (2016). Prevalence of Internet addiction and its association with social support and other related factors among adolescents in China. Journal of Adolescence, 52, 103–111. https ://doi.org/10.1016/j.adole scenc e.2016.07.012. Yip, S. W., White, M. A., Grilo, C. M., & Potenza, M. N. (2011). An exploratory study of clinical measures associated with subsyndromal pathological gambling in patients with binge eating disorder. Journal of Gambling Studies, 27, 257–270. https ://doi.org/10.1007/s1089 9-010-9207-z. Young, K. (2011). Social ties, social networks and the Facebook experience. International Journal of Emerging Technologies and Society, 9, 20. Yücel, M., Whittle, S., Youssef, G. J., Kashyap, H., Simmons, J. G., Schwartz, O., etal. (2015). The influence of sex, temperament, risk-taking and mental health on the emergence of gambling: A longitudinal study of young people. International Gambling Studies, 15, 108–123. https ://doi.org/10.1080/14459 795.2014.10003 56. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.