Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

How coping styles, cognitive distortions, and attachment predict problem gambling among adolescents and young adults

How coping styles, cognitive distortions, and attachment predict problem gambling among... FULL-LENGTH REPORT Journal of Behavioral Addictions 6(4), pp. 648–657 (2017) DOI: 10.1556/2006.6.2017.068 First published online October 26, 2017 How coping styles, cognitive distortions, and attachment predict problem gambling among adolescents and young adults 1 2 1 FILIPA CALADO *, JOANA ALEXANDRE and MARK D. GRIFFITHS International Gaming Research Unit, Psychology Department, Nottingham Trent University, Nottingham, United Kingdom Department of Psychology, ISCTE – University Institute of Lisbon, Lisbon, Portugal (Received: March 26, 2017; revised manuscript received: September 4, 2017; accepted: October 1, 2017) Background and aims: Recent research suggests that youth problem gambling is associated with several factors, but little is known how these factors might influence or interact each other in predicting this behavior. Consequently, this is the first study to examine the mediation effect of coping styles in the relationship between attachment to parental figures and problem gambling. Methods: A total of 988 adolescents and emerging adults were recruited to participate. The first set of analyses tested the adequacy of a model comprising biological, cognitive, and family variables in predicting youth problem gambling. The second set of analyses explored the relationship between family and individual variables in problem gambling behavior. Results: The results of the first set of analyses demonstrated that the individual factors of gender, cognitive distortions, and coping styles showed a significant predictive effect on youth problematic gambling, and the family factors of attachment and family structure did not reveal a significant influence on this behavior. The results of the second set of analyses demonstrated that the attachment dimension of angry distress exerted a more indirect influence on problematic gambling, through emotion-focused coping style. Discussion: This study revealed that some family variables can have a more indirect effect on youth gambling behavior and provided some insights in how some factors interact in predicting problem gambling. Conclusion: These findings suggest that youth gambling is a multifaceted phenomenon, and that the indirect effects of family variables are important in estimating the complex social forces that might influence adolescent decisions to gamble. Keywords: adolescent gambling, attachment, cognitive distortions, coping styles, youth gambling Morris, 1998), which addresses individual risk factors, as INTRODUCTION well as interpersonal and community factors that create the conditions for the development of youth gambling problems Gambling is an activity that occurs in almost all cultures and (Shead, Deverensky, & Gupta, 2010). across all age periods (Griffiths, 1995). However, the current At the individual level, most research has consistently generation of youth represents a vulnerable age group, given found that gender is a risk factor for adolescent gambling they have grown up in an era where gambling opportunities problems. In fact, gambling is much more common are widespread (Gupta & Derevensky, 2000). While for among males than females (Kristiansen & Jensen, 2014), most adolescents, gambling is an enjoyable and harmless and males are more vulnerable to develop gambling-related activity, for a small minority, gambling can become problems (Bastiani et al., 2013; Dodig, 2013; Olason et al., problematic with severe negative consequences (Calado, 2011). Alexandre, & Griffiths, 2017). Therefore, there is a need In addition, at the individual level, some empirical research to study the risk factors underlying youth problem gambling has examined cognitive distortions (e.g., Ariyabuddhiphongs, to provide a more comprehensive description of this phe- 2013; Griffiths, 1994; Tang & Wu, 2012). According to nomenon and its onset. In addition, knowledge about risk some research, adolescent problem gamblers have erroneous factors is critical to identify the signs of youth problem beliefs about the independence of random gambling events and gambling, which can be used to improve assessment tools tend to overestimate their chances of winning (Delfabbro, and develop effective preventive initiatives. Lahn, & Grabosky, 2006; Froberg, 2006; Turner, Macdonald, Researchers have devoted substantial attention to ado- Bartoshuk, & Zangeneh, 2008). lescent gambling and its associated risk factors. Problem gambling is a multifaceted rather than unitary phenomenon (Griffiths, 2011), and consequently, many factors may come * Corresponding author: Filipa Calado; International Gaming into play in various ways and at different levels that Research Unit, Psychology Department, Nottingham Trent contribute to the acquisition, development, and maintenance University, 50 Shakespeare Street, Nottingham NG1 4FQ, United of gambling-related problems. These factors can be concep- Kingdom; Phone: +44 115 941 8418; E-mail: filipa.calado2013@ tualized using an ecological model (Bronfrenbrenner & my.ntu.ac.uk This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium for non-commercial purposes, provided the original author and source are credited. ISSN 2062-5871 © 2017 The Author(s) A model to predict youth problematic gambling Moreover, copying styles, which can be conceptualized Seeley, & Rohling, 2004). In fact, little empirical attention as the way in which people deal with life circumstances, and has been given to the relationship between family socio- regarded as a function of personality and experience (Shead demographic characteristics and adolescent gambling beha- et al., 2010), are also an important risk factor for the viors. This is despite the fact that they appear to be important acquisition and maintenance of youth gambling problems. variables in studying the context of gambling behavior Such coping styles have been categorized into those because ecological models of health behavior recognize intended to directly act on the stressor (i.e., task-oriented family demographic characteristics as determinants of or problem-focused coping) and those intended to regulate health behavior (Flay & Petraitis, 1994). emotional states associated with or resulting from stressful Although there is a growing body of literature on risk and life events (i.e., emotion-oriented coping; Endler & Parker, protective factors, there is still a lack of consensus regarding 1990; Folkman & Lazarus, 1985). For instance, a study the relative weight of each factor in contributing to problem conducted by Gupta, Derevensky, and Marget (2004) gambling among youth (Shead et al., 2010). In fact, it is still reported that adolescents who gamble excessively exhibit unclear which variables (biological, cognitive, and family) coping styles that are more emotion-based. Bergevin, Gupta, play a more significant influence in the development of Derevensky, and Kaufman (2006) also found that students youth problematic gambling. Further research is needed to aged between 11 and 20 years with gambling-related pro- clarify the complex functional relationships between specif- blems used less task-focused coping and more avoidance- ic variables and to incorporate the individual and family focused coping strategies. predictors into a comprehensive and testable etiological Some research also places importance on attitudes in model. Consequently, this study tested a model in which predicting adolescents’ gambling (e.g., Jackson, Dowling, biological, cognitive, and family variables are integrated, Thomas, Bond, & Patton, 2008; Moore & Ohtsuka, 1999). weighting the contribution of each factor, and provided In an Australian sample of 505 adolescents, Delfabbro and further insights into the mechanisms of these variables, by Thrupp (2003) found that more frequent gambling was examining how these variables can interact and influence associated with more pro-gambling attitudes. Similarly, a each other in the development of youth problematic gam- study carried out by Wood and Griffiths (2004) with 1,195 bling behavior. adolescents aged between 11 and 15 years showed that Two sets of analysis were conducted. In the first set of attitudes were an accurate predictor of adolescent gambling analyses, the predictive power of a set of variables on behavior when playing the National Lottery and scratch gambling was examined (including gender along with cards. Furthermore, a qualitative study conducted by cognitive, personality, and family factors), weighing the Calado, Alexandre, and Griffiths (2014) demonstrated that specific contribution of each predictor. This is in line with adolescents displayed a positive attitude toward gambling the conceptualization of gambling as a multifaceted behavior, and that gambling was associated with positive (rather than unitary) phenomenon and therefore attempted outcomes (e.g., more entertainment and a better life due to to overcome previous research that mainly examined the money won). these predictors separately. In addition, this study over- It has also been shown that adolescent gambling comes the lack of research concerning family variables, behaviors are associated with numerous family character- namely the role of attachment to parents or other attach- istics, which can be conceptualized as family com- ment figures in the emergence of youth gambling-related position and parent–adolescent relationship characteristics problems. It was hypothesized that biological, cognitive, (McComb & Sabiston, 2010). In fact, previous research and family variables would show different weights in has found that a low-quality attachment to parents or predicting problem gambling among young people. The other attachment figures have an influence in the initiation model hypothesized comprised the following. First, the of other adolescent risky behaviors, such as drug use starting model examined gender, and it was predicted that and delinquent behaviors (e.g., Kuntsche & Kuendig, male gender would show a significant positive effect on 2006; Miller, Jennings, Alvarez-Rivera, & Lanza-Kaduce, youth problem gambling. Second, the individual predic- 2009). Although not widely studied by gambling research- tors of cognitive distortions, attitudes, and coping were ers, there was some preliminary evidence that attachment added to the model. It was predicted that these variables plays an important role in adolescent gambling behaviors would show a significant predictive effect on youth (Magoon & Ingersoll, 2006), which highlights the need for problem gambling. Finally, the family variables of attach- further research on this specific variable. This study exam- ment and family structure were added to the model, and ined the effect of attachment to parents or other attachment based on previous literature, it was predicted that attach- figures in youth problem gambling in an attempt to over- ment to parents would show a significant influence on come the lack of attention to the influence of specific youth problem gambling, whereas family structure would family variables in this behavior (McComb & Sabiston, nothaveasignificant predictive effect on this behavior. 2010). Based on this first set of analyses, a new model was In addition to attachment, some researchers have also hypothesized examining how individual and family vari- noted that family composition, such as living with parents, ables influence each other in predicting youth problem might serve as a factor that might protect adolescents from gambling. The second set of analyses tested the hypothe- engaging in this risky behavior (Hayer & Griffiths, 2015). sized model, to provide further insights on the relation- On the other hand, other empirical studies have reported that ship between different types of variables that have a family configuration is not associated with adolescent predictive role in the emergence of youth problem gambling behavior (e.g., Langhinrichsen-Rohling, Rohde, gambling. Journal of Behavioral Addictions 6(4), pp. 648–657 (2017) | 649 Calado et al. METHOD (1989) and assesses different coping styles. This instrument comprises 14 subscales, each one measuring a different copying style: active coping, planning, positive reframing, Participants and procedure acceptance, humor, religion, using emotional support, using The participants comprised 988 adolescents and young instrumental support, self-distraction, denial, venting, sub- adults (59.2% males, 40.8% females; mean age = 19.8 stance use, behavioral disengagement, and self-blame. For years, SD = 2.0) attending high schools and the first year this study [as used previously by Reinecke (2009)], the of college in the Nottinghamshire area of the UK. The data scores of the active coping and planning subscales were were collected using standard questionnaires, completed on combined to form a single index for problem-focused a voluntary basis in the school or college. coping, whereas the scores of the self-distraction and the denial subscales were combined to form an index of emo- tion-focused coping. The items of the problem-focused Measures coping describe strategies that comprise problem-solving Sociodemographic information and gambling frequency. (e.g., “I concentrate my efforts on doing something about Sociodemographic data were collected on age, gender, and the situation I am in”) and the items for the emotion-focused family structure (participants had to indicate with whom coping describe strategies that are directed to the regulation they lived, i.e., if they lived with both birth parents, in a of emotions caused by the stressor (e.g., “I turn to work or single-parent family, or with other family members). Parti- other activities to turn my mind off things”). Participants cipants were also asked to indicate how often they had were instructed to respond how often they reacted in the gambled during the past year from 1 (“never”)to6 respective way when facing a problem on a Likert scale (“everyday”). from 0 (never) to 3 (frequently). Cronbach’s αs for the DSM-IV-Multiple Response-Juvenile (DSM-IV-MR-J). subscales in this study were .86 for the problem-focused The DSM-IV-MR-J is a psychometrically validated tool subscale and .72 for the emotion-focused subscale. developed by Fisher (2000) for assessing adolescent prob- Attitudes Towards Gambling Scale (ATGS). The ATGS- lem gambling among those who have gambled during the 8 is an instrument that was developed for the 2010 British past year. This instrument contains nine items, and assesses Gambling Prevalence Survey by Wardle et al. (2011)to a number of important variables related to youth problem assess people’s attitudes toward gambling. The scale com- gambling, such as progression and preoccupation, tolerance, prises eight items with responses given on a 5-point withdrawal, and loss of control. The response categories Likert scale. Higher scores indicate more positive attitudes comprise 1 = “never,” 2 = “once or twice,” 3 = “some- toward gambling. Cronbach’s α for the instrument in this times,” and 4 = “often.” However, although each item has study was .75. four response options, it receives a dichotomous scoring of 0 Adolescent Attachment Questionnaire (AAQ). The AAQ or 1 depending on the response choice (for instance, in the developed by West, Rose, Spreng, Sheldon-Keller, and item 1, if a person chooses the option “often,” he/she will Adam (1998) assesses adolescents’ perceptions of relation- receive a score of 1, but if he chooses any of the other ship security with a nominated adult attachment figure on options, he/she will receive a score of 0). Total score (range three continuous dimensions developed from Bowlby’sspe- 0–9) was calculated by summing up the scores of all nine cific ideas concerning the key characteristics of attachment items. Participants who obtain a score of 0 or 1 are classified relations. The first subscale (angry distress) comprises three as social gamblers, a score of 2 or 3 indicates at-risk items (e.g., “I get annoyed at my mum/dad, because it seems I gambling, and a score of 4 or more indicates problem have to demand his/her care and support”) and assesses anger gambling. Cronbach’s α for the instrument in this study toward attachment figures when attachment needs are frus- was .82. trated. The second subscale (availability) comprises three Gambling-Related Cognitions Scale (GRCS). The items (e.g., “Iamconfident that my mum/dad will listen to 23-item GRCS was developed by Raylu and Oei (2004)to me”) and is related to perceptions of the attachment figure as assess gambling-related erroneous cognitions. It comprises reliably responsive and available to the adolescent’sattach- five subscales answered on a 7-point Likert scale, each one ment needs. The third subscale (goal-corrected partnership) assessing a different type of cognitive distortion: gambling also comprises three items (e.g., “I feel for my mum/dad when expectancies (i.e., expected benefits from gambling), he/she is upset”) and reflects Bowlby’ concept that secure illusion of control (i.e., the perceived ability to control attachment bonds are characterized by an increasing sense of gambling outcomes), predictive control (i.e., the misattribu- empathy toward the attachment figure. Individuals respond to tion of cause–effect relationships to unlinked events), these nine items on a 5-point Likert scale ranging from 1 inability to stop gambling (i.e., the perceived inability to (strongly disagree) to 5 (strongly agree). In this instrument, stop gambling behavior), and interpretative bias (i.e., an items of the availability and goal-corrected partnership sub- error of assessment, such as attributing wins to personal scales are reversed, so that higher scores on the total scale abilities). Higher scores on the GRCS indicate higher levels indicate lower levels of attachment. The total scale showed a of irrational belief. Cronbach’s α for the instrument in this Cronbach’s α of .88. study was .94. Brief Coping Orientation to Problems Experienced First set of analyses (Brief COPE) Inventory. The “Brief COPE” developed by Carver (1997) is a short form of the original COPE Statistical analysis (1). For the first set of analyses, descrip- inventory developed by Carver, Scheier, and Weintraub tive statistics were performed to report the gambling habits 650 | Journal of Behavioral Addictions 6(4), pp. 648–657 (2017) A model to predict youth problematic gambling and gambling activities most played by participants. To distortions were added to the model (model 2). Comparing identify the predictive factors for at-risk and problem gam- models 1 and 2, the χ difference was significant, and bling, a series of hierarchical logistic regressions were justified the introduction of this variable. Model 2 correctly conducted using gender, cognitive distortions, attitudes, classified 86.9% of the respondents. In model 3, attitudes coping, attachment, and family structure as independent were added. Comparing models 2 and 3, the χ difference variables. In accordance with Potenza et al. (2011), the was not significant. Finally, the remaining variables were dependent variable in this logistic regression was the individually added in subsequent steps to the model, and in combination of at-risk and problem gambling, and was each step, the χ significance was verified. The results compared against non-problem gamblers (social and non- indicated that the percentage of correctly classified respon- problem gamblers). dents grew from 85.1 to 88.2 (Table 1). In the final model, the predictors were gender, cognitive distortions, attitudes, coping, attachment, and family structure (model 6). Ethics The specific weight of each predictor is reported in The study procedures were carried out in accordance with Table 2. In addition, cognitive distortions showed a positive the Declaration of Helsinki. Parental permission to partici- significant relationship with problematic gambling. Emotion- pate was provided for those students aged below 18 years focused coping showed a significant positive relationship and informed consent from all participants was obtained. with problematic gambling, whereas problem-focused Participants were requested not to write their names to coping exhibited a negative significant relationship. Atti- maintain anonymity. Finally, the students were offered the tudes, attachment, and family structure did not show a possibility of contact with the authors in case they had significant relationship with problematic gambling. questions or concerns regarding the study. The research team’s university ethics committee provided approval for Second set of analyses the study. The second set of analyses attempted to provide further insights in the relationship between the variables examined. RESULTS (PART 1) The first set of analyses demonstrated that the individual factors of gender, cognitive distortions, and coping had a significant predictive effect on youth problematic gambling Descriptive analysis of gambling habits and activities The results indicated that 79.4% of students had gambled during the past year. The most frequent gambling activities Table 2. Logistic regression analysis with problematic gambling reported by participants were sports betting (15.4% of behavior (problem gambling/at-risk gambling) as the respondents reported they gambled this activity often), dependent variable (N = 988) scratch cards (14.7%), and instant win games (10.7%). Predictors BSE Wald df p OR When questioned about online gambling, the most frequent gambling activities were sports betting (24.8%), gambling in Gender −1.37 0.28 24.16 1 <.001 0.26 social networking sites (7.2%), and blackjack (5.7%). Cognitive 1.18 0.13 89.1 1 <.001 3.26 On the basis of the DSM-IV-MR-J criteria (Fisher, distortions 2000), 20.4% of the participants were classified as non- Attitudes 0 0.02 0.00 1 .986 1 Problem-focused −0.12 0.04 7.645 1 <.01 0.89 gamblers, 64.6% as social gamblers, 8.8% as at-risk gam- coping blers, and 6.2% as problem gamblers. Emotion-focused 0.24 0.04 35.39 1 <.001 1.27 coping Model for predicting youth problematic gambling Attachment 0.05 0.16 0.09 1 .764 1.05 Living with father 0.19 0.47 0.17 1 .69 1.21 In the first regression analysis, the starting point was a Living with mother −0.58 0.31 3.5 1 .06 0.56 model in which gender was the only predictor of the Living with other −0.32 0.30 1.12 1 .29 0.73 dependent variable (model 1; Table 1). This model classified family members 85.1% of respondents. In a second step, the cognitive Table 1. Hierarchical logistic regression analysis with problematic gambling behavior (problem gambling/at-risk gambling) as the dependent variable (N = 988) Model −2log Correct classification (%) Model comparison −2log p Model 1: Gender 765.554 85.1 –– Model 2: Model 1 + cognitive distortions 591.336 86.9 Models 2–1 174.218 <.001 Model 3: Model 2 + attitudes 590.808 86.9 Models 3–2 0.528 .468 Model 4: Model 3 + coping 544.081 87.9 Models 4–3 46.728 <.001 Model 5: Model 4 + attachment 543.995 87.9 Models 5–4 0.085 .770 Model 6: Model 5 + household 539.369 88.2 Models 6–5 4.626 .201 Journal of Behavioral Addictions 6(4), pp. 648–657 (2017) | 651 Calado et al. (problem and at-risk gambling). The family predictors of structural relationships between the predictor, mediator, and attachment and family structure did not show a significant outcome variable. Considering that the outcome variable predictive effect on youth problematic gambling. These assessed by the DSM-IV-MR-J is a dichotomous variable, findings suggest that family variables do not have a direct the WLSMV estimator implemented in Mplus was used impact in the emergence and maintenance of gambling (Muthén & Muthén, 2015). The two-step approach to SEM behavior among young people, and contradict previous recommended by Anderson and Gerbing (1988) was used research on adolescent risky behaviors showing that a low by testing the measurement model in the first step and then level of attachment to parents or other attachment figures has the structural model in the second step. Conventional fit a significant direct effect on adolescents’ alcohol consump- indices, independent of the sample size, were used to tion (Kuntsche & Kuendig, 2006), substance use (Bahr, examine the goodness of fit of the model under analysis: Hoffman, & Yang, 2005; Kopak, Chen, Haas, & Gillmore, root mean square error of approximation (RMSEA), the 2012), and deviant behavior (Miller et al., 2009). comparative fit index (CFI), and the Tucker–Lewis Index In addition, lower levels of attachment have been asso- (TLI) (Vandenberg, 2006). ciated with more avoidant coping strategies (Seiffge-Krenke & Beyers, 2005), which showed a significant influence of RESULTS (PART 2) problem gambling in the first set of analyses. In fact, some authors have argued that attachment theory should be an important base for understanding the origin of coping The means, standard deviations, and correlations between all the variables that will be included in the model are strategies (e.g., Howard & Medway, 2004; Seiffge-Krenke, 2011). Adolescents securely attached to parents or to other outlined in Table 3. attachment figures may develop a more positive self- image in the long term, may be able to better manage SEM with latent constructs stressful experiences, and develop more appropriate coping styles, such as problem-focused coping (Blomgren, Svahn, The criteria for acceptable model fit for these goodness-of-fit indices were defined by CFI ≥ 0.90; TLI ≥ 0.90; RMSEA < Åström, & Rönnlund, 2016). Consequently, the second set of analysis tested a model 0.08 (Hu & Bentler, 1999). Therefore, in the first step, the measurement model showed a very good model fit: to see if coping strategies could mediate the relationship between attachment and gambling behavior and therefore CFI = 0.968; TLI = 0.963; RMSEA = 0.031. In the second step, the structural model was tested. To advance knowledge on the influence of specific variables in youth problem gambling. To further understand how these examine mediation, bootstrapping procedures were con- ducted to determine the indirect effect (Preacher & Hayes, variables can impact each other, the three attachment dimen- sions of angry distress, availability, and goal-corrected 2008). The bootstrapping procedure has advantages over Baron and Kenny’s(1986) traditional approach and Sobel’s partnership, comprising the three subscales of the AAQ are considered. Therefore, based on the above analysis and (1982) test, because it does not assume normality of the sampling distribution of the indirect effects, and it has higher previous literature showing a positive relationship between attachment and more healthy coping strategies, it was power while maintaining adequate control over type I error rate (MacKinnon, Lockwood, Hoffman, West, & Shets, hypothesized that (i) attachment dimensions would not have a significant direct effect on youth problem gambling; (ii) 2002; Preacher & Hayes, 2008). The bootstrap estimates were based on 1,000 bootstrap samples. An indirect effect the negative attachment dimension of angry distress would have a significant indirect effect on youth problem gam- was considered to be significant if its 95% bias-corrected and accelerated (BCa) bootstrap CIs from 1,000 bootstrap sam- bling, through emotion-focused coping style and problem- focused coping style; and (iii) the positive attachment ples exclude zero (Fritz, Taylor, & MacKinnon, 2012). The full structural model again showed a very good model fit: dimensions of availability and goal-corrected partnership would have a significant indirect effect on youth problem CFI = 0.969; TLI = 0.965; RMSEA = 0.030. The direct path from attachment dimensions to problem gambling, through emotion-focused and problem-focused coping style. gambling was non-significant. As mentioned above, the attachment dimensions of availability and goal-corrected Statistical analysis (2). In the second set of analyses, structural equation modeling (SEM) was used to assess the partnership were reversed, so higher scores on these Table 3. Correlations between all the attachment dimensions, copying styles, and problem gambling MSD 1234 5 6 1. Angry distress 1.90 0.93 _ 0.526** 0.385** −0.091** 0.205** 0.057 2. Availability 2.10 0.92 0.526** _ 0.616** −0.205** 0.093** 0.025 3. Goal-corrected partnership 1.81 0.75 0.385** 0.616** _ −0.244** −0.006 0.041 4. Problem-focused coping 7.44 2.92 −0.091** −0.205** −0.244** _ 0.049 −0.149** 5. Emotion-focused coping 4.34 2.86 0.205** 0.093** −0.006 0.049 _ 0.202** 6. Problem gambling 0.72 1.55 0.057 0.025 0.041 −0.149** 0.202** _ **Correlation is significant at the p < .01 level. 652 | Journal of Behavioral Addictions 6(4), pp. 648–657 (2017) A model to predict youth problematic gambling dimensions reveal low levels of availability and goal- (B = 0.24; SE = 0.15; p = .11). Therefore, these results corrected partnership, respectively. The direct path from indicate that emotion-focused coping fully mediated the emotion-focused coping to problem gambling was positive- relationship between angry distress and problem gambling. ly significant and the direct path from problem-focused Moreover, the results indicate that the specific indirect effect coping was negatively significant (Table 4). Therefore, the from angry distress to problem gambling via problem- first hypothesis was fully supported. focused coping was not significant (B = −0.01; 95% BCa In addition, results relating to the second hypothesis CI = −0.06, 0.02). Consequently, the second hypothesis was indicated that the total effect from the attachment dimen- partially confirmed (see Figure 1 for the significant indirect sion of angry distress to problem gambling was significant path from angry distress to problem gambling). (B = 0.35; SE = 0.12; p < .05). The results also showed that The results for the third hypothesis indicated that the total the total indirect effect was also significant (B = 0.12; 95% effect from availability to problem gambling was not signifi- BCa CI = 0.03, 0.19). Examining the indirect effect, the cant (B = −0.197; SE = 0.19; p = .31). The total indirect results indicated that the specific indirect effect from angry effect was not significant either (B = 0.03; 95% BCa distress to problem gambling through emotion-focused CI = −0.05, 0.16). With regard to the attachment dimension coping was significant (B = 0.13; 95% BCa CI = 0.06, of goal-corrected partnership, the total effect from this di- 0.20). However, as noted in Table 4, the direct effect from mension to problem gambling was non-significant (B = 0.07; angry distress to problem gambling was not significant SE = 0.14; p = .60). In addition, the total indirect effect from Table 4. Direct paths to all dependent variables in the study (unstandardized regression coefficients) BSE p Direct paths to problem gambling Angry distress problem gambling 0.24 0.15 .11 Availability problem gambling −0.22 0.27 .41 Goal-corrected partnership problem gambling 0.11 0.199 .58 Emotion-focused coping problem gambling 0.92 0.24 <.001 Problem-focused coping problem gambling −0.23 0.06 <.001 Direct paths to emotion-focused coping Angry distress emotion-focused coping 0.14 0.06 .023 Availability emotion-focused coping −0.00 0.08 .97 Goal-corrected partnership emotion-focused coping −0.07 0.05 .20 Direct paths to problem-focused coping Angry distress problem-focused coping 0.05 0.11 .62 Availability problem-focused coping −0.12 0.24 .62 Goal-corrected partnership problem-focused coping −0.12 0.20 .56 Note. The values represented in bold are statistically significant. Figure 1. Indirect effects from the three attachment dimensions to problem gambling. ns: non-significant. *p > .05. ***p > .001 Journal of Behavioral Addictions 6(4), pp. 648–657 (2017) | 653 Calado et al. goal-corrected partnership to problem gambling, via the two have an indirect effect on youth problematic gambling via coping styles was not significant either (B = −0.04; 95% BCa other individual variables. As previous research has CI = −0.14, 0.04). Therefore, these results suggest that there demonstrated that attachment has an influence on coping is no mediation of coping styles in the relationship between strategies (Seiffge-Krenke & Beyers, 2005), and that a the attachment dimensions of angry distress and goal- primary function of interpersonal attachment is the regu- corrected partnership and problem gambling. This model lation of emotions (McNally, Palfai, Levine, & Moore, accounted for a total of 22% of the variance in problem 2003), the second set of analyses examined if coping gambling, 14.6% of the variance in emotion-focused coping, styles mediated the relationship between attachment to and 4.3% in the variance of problem-focused coping. parents and problem gambling. To better understand the complex relationships between these variables, the second set of analyses examined the three DISCUSSION dimensions of attachment (angry distress, availability, and goal-corrected partnership). The findings indicated that This study explored the effect of individual and family none of these dimensions had a significant direct effect on predictors upon problem gambling, and examined the rela- problem gambling. However, the attachment dimension of tionship between some individual and family variables in angry distress exerted a significant indirect effect on problem the prediction of problem gambling among a student sam- gambling via emotion-focused coping, and which fully me- ple. The first set of analyses showed that a model composed diated the relationship between angry distress and problem of individual and family factors together adequately gambling. Problem-focused coping did not exert any media- explained youth problematic gambling. The second set of tion in the relationship between angry distress and problem analyses demonstrated that attachment dimensions do not gambling. In addition, the other attachment dimensions of have a significant direct effect on problem gambling among availability and goal-corrected partnership did not exert any young people, and that the attachment dimension of angry significant indirect effect on youth problem gambling via any distress exerted a significant indirect influence on youth coping style. Therefore, these results suggest that the attach- problem gambling. ment dimension of angry distress, characterized by feelings The findings of the first set of analyses are in line with of anger toward attachment figures when attachment needs other research (e.g., Griffiths, 2011), which asserts that are frustrated (West et al., 1998), has a major effect in the gambling is a multidimensional (rather than a unitary) emergence of emotion-focused coping, characterized by phenomenon and therefore many factors may come into strategies of regulating emotions caused by the stressor. This, play in the acquisition, development, and maintenance of in turn, will exert an influence in the development of gambling-related problems. Within this integrated perspec- gambling-related problems among young people. tive, the most significant variables for predicting youth Although there was some preliminary evidence that problem gambling were male gender, cognitive distortions, attachment plays a role in adolescent gambling behaviors and emotion-focused coping. These findings are also in line (Magoon & Ingersoll, 2006), these findings extend the with previous research showing that problematic gambling previous gambling literature by examining how different behavior is more prevalent among males (e.g., Bastiani attachment dimensions can exert an effect on youth problem et al., 2013; Griffiths, 2011, Skokauskas & Satkeviciute, gambling via coping styles, and by showing that the attach- 2007), associated with erroneous beliefs of randomness ment dimension of angry distress indirectly influenced this (e.g., Delfabbro et al., 2006; Griffiths, 1994) and with behavior via an emotion-focused coping style. emotion-focused coping (e.g., Bergevin et al., 2006). How- The findings of this study have some important implica- ever, in this study, attitudes did not show a significant effect tions for clinical practice and prevention. In fact, it seems on youth problematic gambling, contradicting previous that a low-quality relationship with parents or other attach- research highlighting the impact of this variable in predict- ment figures may lead youngsters to learn less suitable ing gambling-related problems among young people. There- strategies to deal with their life difficulties, such as using fore, more research into the effect of attitudes on youth gambling to escape from their problems (Wood & Griffiths, gambling is needed among diverse samples, to understand 2007). Therefore, in a clinical context with young problem the meaning of gambling-related attitudes in different cul- gamblers, therapists should assess the quality of the parent– tural contexts and its influence in the emergence of this child relationship, namely potential feelings of anger that behavior. adolescents might feel toward their parents when their needs The first set of analyses also showed that the family are unfulfilled, and must include parents or other attachment predictors of attachment to parents or other attachment figures in the therapeutic process. In addition, during pre- figures, and family structure did not show a significant ventive initiatives, parents should also be trained about their effect on youth problem gambling. These findings con- abilities and aptitudes that could foster the development of a firm other studies showing that family structure does not positive relationship with their children. have a significant impact on youth problem gambling Although this study has some strengths, such as a (e.g., Langhinrichsen-Rohling et al., 2004) but contra- relatively large sample size and the examination of previ- dicts other studies showing that attachment has a signifi- ously unexplored relationships between attachment dimen- cant effect on other adolescent risky behaviors (e.g., Bahr sions, coping styles, and problem gambling, it has also some et al., 2005). Consequently, these findings suggest that limitations. These should be kept in mind when interpreting attachment to parents or other attachment figures may the findings. Most importantly, this study exclusively 654 | Journal of Behavioral Addictions 6(4), pp. 648–657 (2017) A model to predict youth problematic gambling utilized self-report data, which are prone to various well- Ariyabuddhiphongs, V. (2013). Adolescent gambling: A narrative known biases, such as social desirability and memory recall review of behavior and its predictors. International Journal of biases. Second, the study was conducted among a student Mental Health and Addiction, 11(1), 97–109. doi:10.1007/ sample recruited in England and therefore generalizability to s11469-012-9401-6 other populations is limited. Third, this study used a cross- Bahr, S., Hoffman, J., & Yang, X. (2005). Parental and peer sectional design, and thus possible causal relationships influences on the risk of adolescent drug use. Journal of between variables cannot be inferred. Primary Prevention, 26(6), 529–551. doi:10.1007/s10935- Despite these limitations, this is the first study, as far as the 005-0014-8 authors are aware, to examine the mediation effect of coping Baron, R., & Kenny, D. (1986). The moderator-mediator variable styles in the relationship between attachment and problem distinction in social psychological research: Conceptual, stra- gambling. This confirms that the indirect effects of family tegic, and statistical considerations. Journal of Personality and variables are important in estimating the complex social Social Psychology, 51(6), 1173–1182. doi:10.1037/0022- forces that may influence adolescent decisions to gamble. 3514.51.6.1173 However, future studies should be conducted in other coun- Bastiani, L., Gori, M., Colasante, E., Siciliano, V., Capitanucci, D., tries in different contexts and with a wider range in age to Jarre, P., & Molinaro, S. (2013). Complex factors and beha- extend the present findings to other youth populations. viours in the gambling population of Italy. Journal of Gam- Moreover, there is a need for replication of these results in bling Studies, 29(1), 1–13. doi:10.1007/s10899-011-9283-8 longitudinal designs. Considering that youth problem gam- Bergevin, T., Gupta, R., Derevensky, J., & Kaufman, F. (2006). bling has several negative consequences, longitudinal studies Adolescent gambling: Understanding the role of stress and as well as investigations carried out in other contexts could be coping. Journal of Gambling Studies, 22(2), 195–208. doi:10. of great utility in minimizing these outcomes. 1007/s10899-006-9010-z Blomgren, A. S., Svahn, K., Åström, E., & Rönnlund, M. (2016). Coping strategies in late adolescence: Relationships to parental attachment and time perspective. Journal of Genetic Psychol- Funding sources: FC received a grant from FCT, Portu- ogy, 177(3), 85–96. doi:10.1080/00221325.2016.1178101 guese national funding agency for science, research and Bronfrenbrenner, U., & Morris, P. A. (1998). The ecology of technology (reference number SFRH/BD/119749/2016). developmental processes. In Lerner, R. (Ed.), Handbook of MDG has received funding for a number of research child psychology: Theoretical models of human development projects in the area of gambling education for young people, (5th ed., Vol. 1, pp. 993–1028). New York, NY: John Wiley. social responsibility in gambling, and gambling treatment Calado, F., Alexandre, J., & Griffiths, M. D. (2014). Mom, Dad it’s from the Responsibility in Gambling Trust, a charitable only a game! Perceived gambling and gaming behaviors body, which funds its research program based on donations among adolescents and young adults: An exploratory study. from the gambling industry. He also undertakes consultancy International Journal of Mental Health and Addiction, 12(6), for various gaming companies in the area of social respon- 772–794. doi:10.1007/s11469-014-9509-y sibility in gambling. Calado, F., Alexandre, J., & Griffiths, M. D. (2017). Prevalence of adolescent problem gambling: A systematic review of recent Authors’ contribution: FC and MDG designed the study. FC research. Journal of Gambling Studies, 33(2), 397–424. doi:10. collected and analyzed the data. All authors inputted into the 1007/s10899-016-9627-5 interpretation of the results. FC wrote the initial draft of the Carver, C. S. (1997). You want to measure coping but your paper and MDG added significantly to the first draft. All protocol’s too long: Consider the brief COPE. International authors then went through a number of iterative versions Journal of Behavioral Medicine, 4(1), 92–100. doi:10.1207/ before the final manuscript was submitted. All authors have s15327558ijbm0401_6 full access to all the data in the study and take responsibility Carver, C. S., Scheier, M. F., & Weintraub, J. K. (1989). Assessing for the integrity of the data and the accuracy of the data coping strategies: A theoretically based approach. Journal of analysis. Personality and Social Psychology, 56(2), 267–283. doi:10.1037/0022-3514.56.2.267 Conflict of interest: FC and JA declare no conflict of Delfabbro, P., Lahn, J., & Grabosky, P. (2006). It’s not what you interest. know, but how you use it: Statistical knowledge and adolescent problem gambling. Journal of Gambling Studies, 22(2), 179– Acknowledgements: The authors would like to thank 193. doi:10.1007/s10899-006-9009-5 Tadeusz Borejko for his assistance in participant recruitment Delfabbro, P., & Thrupp, L. (2003). The social determinants of and data collection. youth gambling in South Australian adolescents. Journal of Adolescence, 26(3), 313–330. doi:10.1016/S0140-1971(03) 00013-7 REFERENCES Dodig, D. (2013). Assessment challenges and determinants of adolescents’ adverse psychosocial consequences of gambling. Anderson, J. C., & Gerbing, D. W. (1988). Structural equation Kriminologija i Socijalna Integracija, 21, 1–29. modeling in practice: A review and recommended two-step Endler, N. S., & Parker, J. D. A. (1990). Coping Inventory for approach. Psychological Bulletin, 103(3), 411–423. doi:10. Stressful Situations (CISS): Manual. Toronto, Canada: Multi- 1037/0033-2909.103.3.411 Health Systems. Journal of Behavioral Addictions 6(4), pp. 648–657 (2017) | 655 Calado et al. Fisher, S. (2000). Developing the DSM-IV-MR-J criteria to iden- Kuntsche, E. N., & Kuendig, H. (2006). What is worse? A tify adolescent problem gambling in nonclinical populations. hierarchy of family-related risk factors predicting alcohol use Journal of Gambling Studies, 16(2–3), 253–273. doi:10.1023/ in adolescence. Substance Use and Misuse, 41(1), 71–86. A:1009437115789 doi:10.1080/10826080500368694 Flay, B. R., & Petraitis, J. (1994). The theory of triadic influence: A Langhinrichsen-Rohling, J., Rohde, P., Seeley, J. R., & Rohling, new theory of health behavior with implications for preventive M. L. (2004). Individual, family and peer correlates of adoles- interventions. Advances in Medical Sociology, 4, 19–44. cent gambling. Journal of Gambling Studies, 20(1), 23–46. Folkman, S., & Lazarus, R. A. (1985). If it changes it must be doi:10.1023/B:JOGS.0000016702.69068.53 a process: A study of emotion and coping during three MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., stages of a college examination. Journal of Personality and & Sheets, V. (2002). A comparison of methods to test media- Social Psychology, 48(1), 150–170. doi:10.1037/0022- tion and other intervening variable effects. Psychological 3514.48.1.150 Methods, 7(1), 83–104. doi:10.1037/1082-989X.7.1.83 Fritz, M. S., Taylor, A. B., & MacKinnon, D. P. (2012). Explana- Magoon, M. E., & Ingersoll, G. M. (2006). Parental modeling, tion of two anomalous results in statistical mediation analysis. attachment, and supervision as moderators of adolescent gam- Multivariate Behavioral Research, 47(1), 61–87. doi:10.1080/ bling. Journal of Gambling Studies, 22(1), 1–22. doi:10.1007/ 00273171.2012.640596 s10899-005-9000-6 Froberg, F. (2006). Gambling among young people. A knowledge McComb, J. L., & Sabiston, C. M. (2010). Family influences on review. Stockholm, Sweden: Swedish National Institute of adolescent gambling behavior: A review of the literature. Public Health. Journal of Gambling Studies, 26(4), 503–520. doi:10.1007/ Griffiths, M. D. (1994). The role of cognitive bias and skill in fruit s10899-010-9181-5 machine gambling. British Journal of Psychology, 85(3), 351– McNally, A. M., Palfai, T. P., Levine, R. V., & Moore, B. M. 369. doi:10.1111/j.2044-8295.1994.tb02529.x (2003). Attachment dimensions and drinking-related problems Griffiths, M. D. (1995). Adolescent gambling. London, UK: among young adults. The mediational role of coping motives. Routledge. Addictive Behaviors, 28(6), 1115–1127. doi:10.1016/S0306- Griffiths, M. D. (2011). Adolescent gambling. In Bradford Brown, 4603(02)00224-1 B. & Prinstein, M. J. (Eds.), Encyclopedia of adolescence (Vol. Miller, H. V., Jennings, W. G., Alvarez-Rivera, L. L., & Lanza- 3, pp. 11–20). San Diego, CA: Academic Press. Kaduce, L. (2009). Self-control, attachment, and deviance Gupta, R., Derevensky, J., & Marget, N. (2004). Coping strategies among Hispanic adolescents. Journal of Criminal Justice, employed by adolescents with gambling problems. Child and 37(1), 77–84. doi:10.1016/j.jcrimjus.2008.12.003 Adolescent Mental Health, 9(3), 115–120. doi:10.1111/j.1475- Moore, S. M., & Ohtsuka, K. (1999). Beliefs about control over 3588.2004.00092.x gambling among young people, and their relation to problem Gupta, R., & Derevensky, J. L. (2000). Adolescents with gambling gambling. Psychology of Addictive Behaviors, 13(4), 339–347. problems: From research to treatment. Journal of Gambling doi:10.1037/0893-164X.13.4.339 Studies, 16(2–3), 315–342. doi:10.1023/A:1009493200768 Muthén, L. K., & Muthén, B. O. (2015). Mplus user’s guide (7th Hayer, T., & Griffiths, M. D. (2015). The prevention and treatment ed.). Los Angeles, CA: Muthén & Muthén. of problem gambling in adolescence. In Gullotta, T. P. & Olason, D. T., Kristjansdottir, E., Einarsdottir, H., Haraldsson, H., Adams, G. (Eds.), Handbook of adolescent behavioural pro- Bjarnason, G., & Derevensky, J. L. (2011). Internet gambling blems: Evidence-based approaches to prevention and treat- and problem gambling among 13 to 18 year old adolescents in ment (2nd ed., pp. 539–558). New York, NY: Springer. Iceland. International Journal of Mental Health and Addiction, Howard, M. S., & Medway, F. J. (2004). Adolescents’ attachment 9(3), 257–263. doi:10.1007/s11469-010-9280-7 and coping with stress. Psychology in the Schools, 41(3), 391– Potenza, M. N., Wareham, J. D., Steinberg, M. A., Rugle, L., 402. doi:10.1002/pits.10167 Cavallo, D. A., Krishnan-Sarin, S., & Desai, R. A. (2011). Hu, L., & Bentler, P. M. (1999). Cutoff criteria in fix indexes in Correlates of at-risk/problem internet gambling in adolescents. covariance structure analysis: Conventional criteria versus new Journal of American Academy of Child and Adolescent Psy- alternatives. Structural Equation Modeling, 6(1), 1–55. doi:10. chiatry, 50(2), 150–159.e3. doi:10.1016/j.jaac.2010.11.006 1080/10705519909540118 Preacher, K., & Hayes, A. (2008). Asymptotic and resampling Jackson, A. C., Dowling, N., Thomas, S. A., Bond, L., & Patton, G. strategies for assessing and comparing indirect effects in (2008). Adolescent gambling behaviour and attitudes: A multiple mediator models. Behavior Research Methods, prevalence study and correlates in an Australian population. 40(3), 879–891. doi:10.3758/BRM.40.3.879 International Journal of Mental Health and Addiction, 6(3), Raylu, N., & Oei, T. P. (2004). The Gambling Related Cognitions 325–352. doi:10.1007/s11469-008-9149-1 Scale (GRCS): Development, confirmatory factor validation Kopak, A. M., Chen, A. C. C., Haas, S. A., & Gillmore, M. R. and psychometric properties. Addiction, 99(6), 757–769. (2012). The importance of family factors to protect against doi:10.1111/j.1360-0443.2004.00753.x substance use related problems among Mexican heritage and Reinecke, L. (2009). Games and recovery: The use of video and White youth. Drug and Alcohol Dependence, 124(1–2), 34–41. computer games to recuperate from stress and strain. Journal of doi:10.1016/j.drugalcdep.2011.12.004 Media Psychology, 21(3), 126–142. doi:10.1027/1864-1105. Kristiansen, S. G., & Jensen, S. M. (2014). Prevalence and 21.3.126 correlates of problematic gambling among Danish adolescents. Seiffge-Krenke, I. (2011). Coping with relationship stressors: A International Journal of Social Welfare, 23(1), 89–99. decade review. Journal of Research on Adolescence, 21(1), doi:10.1111/ijsw.12021 196–210. doi:10.1111/j.1532-7795.2010.00723.x 656 | Journal of Behavioral Addictions 6(4), pp. 648–657 (2017) A model to predict youth problematic gambling Seiffge-Krenke, I., & Beyers, W. (2005). Coping trajectories from Vandenberg, R. J. (2006). Statistical and methodological myths adolescence to young adulthood: Links to attachment state of and urban legends: Where, pray tell, did they get this idea? mind. Journal of Research on Adolescence, 15(4), 561–582. Organizational Research Methods, 9(2), 194–201. doi:10. doi:10.1111/j.1532-7795.2005.00111.x 1177/1094428105285506 Shead, N. W., Derevensky, J. L., & Gupta, R. (2010). Risk and Wardle, H., Moody, A., Spence, S., Orford, J., Volberg, R., protective factors associated with youth problem gambling. Jotangia, D., Griffiths, M., Hussey, D., & Dobbie, F. International Journal of Adolescent Medicine and Health, 22, (2011). The British gambling prevalence survey 2010. London, 39–58. UK: The Stationery Office. Skokauskas, N., & Satkeviciute, R. (2007). Adolescent pathologi- West,W., Rose,S.M., Spreng,S., Sheldon-Keller,A., &Adam, cal gambling in Kaunas, Lithuania. Nordic Journal of Psychi- K. (1998). Adolescent attachment questionnaire: A brief atry, 61(2), 86–91. doi:10.1080/08039480701226054 assessment of attachment in adolescence. Journal of Youth Sobel, M. (1982). Asymptotic confidence intervals for indirect and Adolescence, 27(5), 661–673. doi:10.1023/A:102289 effects in structural equation models. Sociological Methodology, 1225542 13, 347–355. doi:10.2307/270723 Wood, R. T., & Griffiths, M. D. (2004). Adolescent lottery and Tang, C. S.-K., & Wu, A. M. S. (2012). Gambling-related cognitive scratchcard players: Do their attitudes influence their gambling biases and pathological gambling among youths, young adults, behaviour? Journal of Adolescence, 27(4), 467–475. doi:10. and mature adults, in Chinese societies. Journal of Gambling 1016/j.adolescence.2003.12.003 Studies, 28(1), 139–154. doi:10.1007/s10899-011-9249-x Wood, R. T., & Griffiths, M. D. (2007). A qualitative investigation Turner, N. E., Macdonald, J., Bartoshuk, M., & Zangeneh, M. of problem gambling as an escape-based coping strategy. (2008). Adolescent gambling behaviour, attitudes, and gam- Psychology and Psychotherapy: Theory, Research and Prac- bling problems. International Journal of Mental Health and tice, 80(1), 107–125. doi:10.1348/147608306X107881 Addiction, 6(2), 223–237. doi:10.1007/s11469-007-9117-1 Journal of Behavioral Addictions 6(4), pp. 648–657 (2017) | 657 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Behavioral Addictions Pubmed Central

How coping styles, cognitive distortions, and attachment predict problem gambling among adolescents and young adults

Calado, Filipa; Alexandre, Joana; Griffiths, Mark D.
Journal of Behavioral Addictions , Volume 6 (4) – Oct 23, 2017
Download PDF
10 pages

Loading next page...
 
/lp/pubmed-central/how-coping-styles-cognitive-distortions-and-attachment-predict-problem-Ro31QnZf9g

References (115)

Publisher
Pubmed Central
Copyright
© 2017 The Author(s)
ISSN
2062-5871
eISSN
2063-5303
DOI
10.1556/2006.6.2017.068
Publisher site
See Article on Publisher Site

Abstract

FULL-LENGTH REPORT Journal of Behavioral Addictions 6(4), pp. 648–657 (2017) DOI: 10.1556/2006.6.2017.068 First published online October 26, 2017 How coping styles, cognitive distortions, and attachment predict problem gambling among adolescents and young adults 1 2 1 FILIPA CALADO *, JOANA ALEXANDRE and MARK D. GRIFFITHS International Gaming Research Unit, Psychology Department, Nottingham Trent University, Nottingham, United Kingdom Department of Psychology, ISCTE – University Institute of Lisbon, Lisbon, Portugal (Received: March 26, 2017; revised manuscript received: September 4, 2017; accepted: October 1, 2017) Background and aims: Recent research suggests that youth problem gambling is associated with several factors, but little is known how these factors might influence or interact each other in predicting this behavior. Consequently, this is the first study to examine the mediation effect of coping styles in the relationship between attachment to parental figures and problem gambling. Methods: A total of 988 adolescents and emerging adults were recruited to participate. The first set of analyses tested the adequacy of a model comprising biological, cognitive, and family variables in predicting youth problem gambling. The second set of analyses explored the relationship between family and individual variables in problem gambling behavior. Results: The results of the first set of analyses demonstrated that the individual factors of gender, cognitive distortions, and coping styles showed a significant predictive effect on youth problematic gambling, and the family factors of attachment and family structure did not reveal a significant influence on this behavior. The results of the second set of analyses demonstrated that the attachment dimension of angry distress exerted a more indirect influence on problematic gambling, through emotion-focused coping style. Discussion: This study revealed that some family variables can have a more indirect effect on youth gambling behavior and provided some insights in how some factors interact in predicting problem gambling. Conclusion: These findings suggest that youth gambling is a multifaceted phenomenon, and that the indirect effects of family variables are important in estimating the complex social forces that might influence adolescent decisions to gamble. Keywords: adolescent gambling, attachment, cognitive distortions, coping styles, youth gambling Morris, 1998), which addresses individual risk factors, as INTRODUCTION well as interpersonal and community factors that create the conditions for the development of youth gambling problems Gambling is an activity that occurs in almost all cultures and (Shead, Deverensky, & Gupta, 2010). across all age periods (Griffiths, 1995). However, the current At the individual level, most research has consistently generation of youth represents a vulnerable age group, given found that gender is a risk factor for adolescent gambling they have grown up in an era where gambling opportunities problems. In fact, gambling is much more common are widespread (Gupta & Derevensky, 2000). While for among males than females (Kristiansen & Jensen, 2014), most adolescents, gambling is an enjoyable and harmless and males are more vulnerable to develop gambling-related activity, for a small minority, gambling can become problems (Bastiani et al., 2013; Dodig, 2013; Olason et al., problematic with severe negative consequences (Calado, 2011). Alexandre, & Griffiths, 2017). Therefore, there is a need In addition, at the individual level, some empirical research to study the risk factors underlying youth problem gambling has examined cognitive distortions (e.g., Ariyabuddhiphongs, to provide a more comprehensive description of this phe- 2013; Griffiths, 1994; Tang & Wu, 2012). According to nomenon and its onset. In addition, knowledge about risk some research, adolescent problem gamblers have erroneous factors is critical to identify the signs of youth problem beliefs about the independence of random gambling events and gambling, which can be used to improve assessment tools tend to overestimate their chances of winning (Delfabbro, and develop effective preventive initiatives. Lahn, & Grabosky, 2006; Froberg, 2006; Turner, Macdonald, Researchers have devoted substantial attention to ado- Bartoshuk, & Zangeneh, 2008). lescent gambling and its associated risk factors. Problem gambling is a multifaceted rather than unitary phenomenon (Griffiths, 2011), and consequently, many factors may come * Corresponding author: Filipa Calado; International Gaming into play in various ways and at different levels that Research Unit, Psychology Department, Nottingham Trent contribute to the acquisition, development, and maintenance University, 50 Shakespeare Street, Nottingham NG1 4FQ, United of gambling-related problems. These factors can be concep- Kingdom; Phone: +44 115 941 8418; E-mail: filipa.calado2013@ tualized using an ecological model (Bronfrenbrenner & my.ntu.ac.uk This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium for non-commercial purposes, provided the original author and source are credited. ISSN 2062-5871 © 2017 The Author(s) A model to predict youth problematic gambling Moreover, copying styles, which can be conceptualized Seeley, & Rohling, 2004). In fact, little empirical attention as the way in which people deal with life circumstances, and has been given to the relationship between family socio- regarded as a function of personality and experience (Shead demographic characteristics and adolescent gambling beha- et al., 2010), are also an important risk factor for the viors. This is despite the fact that they appear to be important acquisition and maintenance of youth gambling problems. variables in studying the context of gambling behavior Such coping styles have been categorized into those because ecological models of health behavior recognize intended to directly act on the stressor (i.e., task-oriented family demographic characteristics as determinants of or problem-focused coping) and those intended to regulate health behavior (Flay & Petraitis, 1994). emotional states associated with or resulting from stressful Although there is a growing body of literature on risk and life events (i.e., emotion-oriented coping; Endler & Parker, protective factors, there is still a lack of consensus regarding 1990; Folkman & Lazarus, 1985). For instance, a study the relative weight of each factor in contributing to problem conducted by Gupta, Derevensky, and Marget (2004) gambling among youth (Shead et al., 2010). In fact, it is still reported that adolescents who gamble excessively exhibit unclear which variables (biological, cognitive, and family) coping styles that are more emotion-based. Bergevin, Gupta, play a more significant influence in the development of Derevensky, and Kaufman (2006) also found that students youth problematic gambling. Further research is needed to aged between 11 and 20 years with gambling-related pro- clarify the complex functional relationships between specif- blems used less task-focused coping and more avoidance- ic variables and to incorporate the individual and family focused coping strategies. predictors into a comprehensive and testable etiological Some research also places importance on attitudes in model. Consequently, this study tested a model in which predicting adolescents’ gambling (e.g., Jackson, Dowling, biological, cognitive, and family variables are integrated, Thomas, Bond, & Patton, 2008; Moore & Ohtsuka, 1999). weighting the contribution of each factor, and provided In an Australian sample of 505 adolescents, Delfabbro and further insights into the mechanisms of these variables, by Thrupp (2003) found that more frequent gambling was examining how these variables can interact and influence associated with more pro-gambling attitudes. Similarly, a each other in the development of youth problematic gam- study carried out by Wood and Griffiths (2004) with 1,195 bling behavior. adolescents aged between 11 and 15 years showed that Two sets of analysis were conducted. In the first set of attitudes were an accurate predictor of adolescent gambling analyses, the predictive power of a set of variables on behavior when playing the National Lottery and scratch gambling was examined (including gender along with cards. Furthermore, a qualitative study conducted by cognitive, personality, and family factors), weighing the Calado, Alexandre, and Griffiths (2014) demonstrated that specific contribution of each predictor. This is in line with adolescents displayed a positive attitude toward gambling the conceptualization of gambling as a multifaceted behavior, and that gambling was associated with positive (rather than unitary) phenomenon and therefore attempted outcomes (e.g., more entertainment and a better life due to to overcome previous research that mainly examined the money won). these predictors separately. In addition, this study over- It has also been shown that adolescent gambling comes the lack of research concerning family variables, behaviors are associated with numerous family character- namely the role of attachment to parents or other attach- istics, which can be conceptualized as family com- ment figures in the emergence of youth gambling-related position and parent–adolescent relationship characteristics problems. It was hypothesized that biological, cognitive, (McComb & Sabiston, 2010). In fact, previous research and family variables would show different weights in has found that a low-quality attachment to parents or predicting problem gambling among young people. The other attachment figures have an influence in the initiation model hypothesized comprised the following. First, the of other adolescent risky behaviors, such as drug use starting model examined gender, and it was predicted that and delinquent behaviors (e.g., Kuntsche & Kuendig, male gender would show a significant positive effect on 2006; Miller, Jennings, Alvarez-Rivera, & Lanza-Kaduce, youth problem gambling. Second, the individual predic- 2009). Although not widely studied by gambling research- tors of cognitive distortions, attitudes, and coping were ers, there was some preliminary evidence that attachment added to the model. It was predicted that these variables plays an important role in adolescent gambling behaviors would show a significant predictive effect on youth (Magoon & Ingersoll, 2006), which highlights the need for problem gambling. Finally, the family variables of attach- further research on this specific variable. This study exam- ment and family structure were added to the model, and ined the effect of attachment to parents or other attachment based on previous literature, it was predicted that attach- figures in youth problem gambling in an attempt to over- ment to parents would show a significant influence on come the lack of attention to the influence of specific youth problem gambling, whereas family structure would family variables in this behavior (McComb & Sabiston, nothaveasignificant predictive effect on this behavior. 2010). Based on this first set of analyses, a new model was In addition to attachment, some researchers have also hypothesized examining how individual and family vari- noted that family composition, such as living with parents, ables influence each other in predicting youth problem might serve as a factor that might protect adolescents from gambling. The second set of analyses tested the hypothe- engaging in this risky behavior (Hayer & Griffiths, 2015). sized model, to provide further insights on the relation- On the other hand, other empirical studies have reported that ship between different types of variables that have a family configuration is not associated with adolescent predictive role in the emergence of youth problem gambling behavior (e.g., Langhinrichsen-Rohling, Rohde, gambling. Journal of Behavioral Addictions 6(4), pp. 648–657 (2017) | 649 Calado et al. METHOD (1989) and assesses different coping styles. This instrument comprises 14 subscales, each one measuring a different copying style: active coping, planning, positive reframing, Participants and procedure acceptance, humor, religion, using emotional support, using The participants comprised 988 adolescents and young instrumental support, self-distraction, denial, venting, sub- adults (59.2% males, 40.8% females; mean age = 19.8 stance use, behavioral disengagement, and self-blame. For years, SD = 2.0) attending high schools and the first year this study [as used previously by Reinecke (2009)], the of college in the Nottinghamshire area of the UK. The data scores of the active coping and planning subscales were were collected using standard questionnaires, completed on combined to form a single index for problem-focused a voluntary basis in the school or college. coping, whereas the scores of the self-distraction and the denial subscales were combined to form an index of emo- tion-focused coping. The items of the problem-focused Measures coping describe strategies that comprise problem-solving Sociodemographic information and gambling frequency. (e.g., “I concentrate my efforts on doing something about Sociodemographic data were collected on age, gender, and the situation I am in”) and the items for the emotion-focused family structure (participants had to indicate with whom coping describe strategies that are directed to the regulation they lived, i.e., if they lived with both birth parents, in a of emotions caused by the stressor (e.g., “I turn to work or single-parent family, or with other family members). Parti- other activities to turn my mind off things”). Participants cipants were also asked to indicate how often they had were instructed to respond how often they reacted in the gambled during the past year from 1 (“never”)to6 respective way when facing a problem on a Likert scale (“everyday”). from 0 (never) to 3 (frequently). Cronbach’s αs for the DSM-IV-Multiple Response-Juvenile (DSM-IV-MR-J). subscales in this study were .86 for the problem-focused The DSM-IV-MR-J is a psychometrically validated tool subscale and .72 for the emotion-focused subscale. developed by Fisher (2000) for assessing adolescent prob- Attitudes Towards Gambling Scale (ATGS). The ATGS- lem gambling among those who have gambled during the 8 is an instrument that was developed for the 2010 British past year. This instrument contains nine items, and assesses Gambling Prevalence Survey by Wardle et al. (2011)to a number of important variables related to youth problem assess people’s attitudes toward gambling. The scale com- gambling, such as progression and preoccupation, tolerance, prises eight items with responses given on a 5-point withdrawal, and loss of control. The response categories Likert scale. Higher scores indicate more positive attitudes comprise 1 = “never,” 2 = “once or twice,” 3 = “some- toward gambling. Cronbach’s α for the instrument in this times,” and 4 = “often.” However, although each item has study was .75. four response options, it receives a dichotomous scoring of 0 Adolescent Attachment Questionnaire (AAQ). The AAQ or 1 depending on the response choice (for instance, in the developed by West, Rose, Spreng, Sheldon-Keller, and item 1, if a person chooses the option “often,” he/she will Adam (1998) assesses adolescents’ perceptions of relation- receive a score of 1, but if he chooses any of the other ship security with a nominated adult attachment figure on options, he/she will receive a score of 0). Total score (range three continuous dimensions developed from Bowlby’sspe- 0–9) was calculated by summing up the scores of all nine cific ideas concerning the key characteristics of attachment items. Participants who obtain a score of 0 or 1 are classified relations. The first subscale (angry distress) comprises three as social gamblers, a score of 2 or 3 indicates at-risk items (e.g., “I get annoyed at my mum/dad, because it seems I gambling, and a score of 4 or more indicates problem have to demand his/her care and support”) and assesses anger gambling. Cronbach’s α for the instrument in this study toward attachment figures when attachment needs are frus- was .82. trated. The second subscale (availability) comprises three Gambling-Related Cognitions Scale (GRCS). The items (e.g., “Iamconfident that my mum/dad will listen to 23-item GRCS was developed by Raylu and Oei (2004)to me”) and is related to perceptions of the attachment figure as assess gambling-related erroneous cognitions. It comprises reliably responsive and available to the adolescent’sattach- five subscales answered on a 7-point Likert scale, each one ment needs. The third subscale (goal-corrected partnership) assessing a different type of cognitive distortion: gambling also comprises three items (e.g., “I feel for my mum/dad when expectancies (i.e., expected benefits from gambling), he/she is upset”) and reflects Bowlby’ concept that secure illusion of control (i.e., the perceived ability to control attachment bonds are characterized by an increasing sense of gambling outcomes), predictive control (i.e., the misattribu- empathy toward the attachment figure. Individuals respond to tion of cause–effect relationships to unlinked events), these nine items on a 5-point Likert scale ranging from 1 inability to stop gambling (i.e., the perceived inability to (strongly disagree) to 5 (strongly agree). In this instrument, stop gambling behavior), and interpretative bias (i.e., an items of the availability and goal-corrected partnership sub- error of assessment, such as attributing wins to personal scales are reversed, so that higher scores on the total scale abilities). Higher scores on the GRCS indicate higher levels indicate lower levels of attachment. The total scale showed a of irrational belief. Cronbach’s α for the instrument in this Cronbach’s α of .88. study was .94. Brief Coping Orientation to Problems Experienced First set of analyses (Brief COPE) Inventory. The “Brief COPE” developed by Carver (1997) is a short form of the original COPE Statistical analysis (1). For the first set of analyses, descrip- inventory developed by Carver, Scheier, and Weintraub tive statistics were performed to report the gambling habits 650 | Journal of Behavioral Addictions 6(4), pp. 648–657 (2017) A model to predict youth problematic gambling and gambling activities most played by participants. To distortions were added to the model (model 2). Comparing identify the predictive factors for at-risk and problem gam- models 1 and 2, the χ difference was significant, and bling, a series of hierarchical logistic regressions were justified the introduction of this variable. Model 2 correctly conducted using gender, cognitive distortions, attitudes, classified 86.9% of the respondents. In model 3, attitudes coping, attachment, and family structure as independent were added. Comparing models 2 and 3, the χ difference variables. In accordance with Potenza et al. (2011), the was not significant. Finally, the remaining variables were dependent variable in this logistic regression was the individually added in subsequent steps to the model, and in combination of at-risk and problem gambling, and was each step, the χ significance was verified. The results compared against non-problem gamblers (social and non- indicated that the percentage of correctly classified respon- problem gamblers). dents grew from 85.1 to 88.2 (Table 1). In the final model, the predictors were gender, cognitive distortions, attitudes, coping, attachment, and family structure (model 6). Ethics The specific weight of each predictor is reported in The study procedures were carried out in accordance with Table 2. In addition, cognitive distortions showed a positive the Declaration of Helsinki. Parental permission to partici- significant relationship with problematic gambling. Emotion- pate was provided for those students aged below 18 years focused coping showed a significant positive relationship and informed consent from all participants was obtained. with problematic gambling, whereas problem-focused Participants were requested not to write their names to coping exhibited a negative significant relationship. Atti- maintain anonymity. Finally, the students were offered the tudes, attachment, and family structure did not show a possibility of contact with the authors in case they had significant relationship with problematic gambling. questions or concerns regarding the study. The research team’s university ethics committee provided approval for Second set of analyses the study. The second set of analyses attempted to provide further insights in the relationship between the variables examined. RESULTS (PART 1) The first set of analyses demonstrated that the individual factors of gender, cognitive distortions, and coping had a significant predictive effect on youth problematic gambling Descriptive analysis of gambling habits and activities The results indicated that 79.4% of students had gambled during the past year. The most frequent gambling activities Table 2. Logistic regression analysis with problematic gambling reported by participants were sports betting (15.4% of behavior (problem gambling/at-risk gambling) as the respondents reported they gambled this activity often), dependent variable (N = 988) scratch cards (14.7%), and instant win games (10.7%). Predictors BSE Wald df p OR When questioned about online gambling, the most frequent gambling activities were sports betting (24.8%), gambling in Gender −1.37 0.28 24.16 1 <.001 0.26 social networking sites (7.2%), and blackjack (5.7%). Cognitive 1.18 0.13 89.1 1 <.001 3.26 On the basis of the DSM-IV-MR-J criteria (Fisher, distortions 2000), 20.4% of the participants were classified as non- Attitudes 0 0.02 0.00 1 .986 1 Problem-focused −0.12 0.04 7.645 1 <.01 0.89 gamblers, 64.6% as social gamblers, 8.8% as at-risk gam- coping blers, and 6.2% as problem gamblers. Emotion-focused 0.24 0.04 35.39 1 <.001 1.27 coping Model for predicting youth problematic gambling Attachment 0.05 0.16 0.09 1 .764 1.05 Living with father 0.19 0.47 0.17 1 .69 1.21 In the first regression analysis, the starting point was a Living with mother −0.58 0.31 3.5 1 .06 0.56 model in which gender was the only predictor of the Living with other −0.32 0.30 1.12 1 .29 0.73 dependent variable (model 1; Table 1). This model classified family members 85.1% of respondents. In a second step, the cognitive Table 1. Hierarchical logistic regression analysis with problematic gambling behavior (problem gambling/at-risk gambling) as the dependent variable (N = 988) Model −2log Correct classification (%) Model comparison −2log p Model 1: Gender 765.554 85.1 –– Model 2: Model 1 + cognitive distortions 591.336 86.9 Models 2–1 174.218 <.001 Model 3: Model 2 + attitudes 590.808 86.9 Models 3–2 0.528 .468 Model 4: Model 3 + coping 544.081 87.9 Models 4–3 46.728 <.001 Model 5: Model 4 + attachment 543.995 87.9 Models 5–4 0.085 .770 Model 6: Model 5 + household 539.369 88.2 Models 6–5 4.626 .201 Journal of Behavioral Addictions 6(4), pp. 648–657 (2017) | 651 Calado et al. (problem and at-risk gambling). The family predictors of structural relationships between the predictor, mediator, and attachment and family structure did not show a significant outcome variable. Considering that the outcome variable predictive effect on youth problematic gambling. These assessed by the DSM-IV-MR-J is a dichotomous variable, findings suggest that family variables do not have a direct the WLSMV estimator implemented in Mplus was used impact in the emergence and maintenance of gambling (Muthén & Muthén, 2015). The two-step approach to SEM behavior among young people, and contradict previous recommended by Anderson and Gerbing (1988) was used research on adolescent risky behaviors showing that a low by testing the measurement model in the first step and then level of attachment to parents or other attachment figures has the structural model in the second step. Conventional fit a significant direct effect on adolescents’ alcohol consump- indices, independent of the sample size, were used to tion (Kuntsche & Kuendig, 2006), substance use (Bahr, examine the goodness of fit of the model under analysis: Hoffman, & Yang, 2005; Kopak, Chen, Haas, & Gillmore, root mean square error of approximation (RMSEA), the 2012), and deviant behavior (Miller et al., 2009). comparative fit index (CFI), and the Tucker–Lewis Index In addition, lower levels of attachment have been asso- (TLI) (Vandenberg, 2006). ciated with more avoidant coping strategies (Seiffge-Krenke & Beyers, 2005), which showed a significant influence of RESULTS (PART 2) problem gambling in the first set of analyses. In fact, some authors have argued that attachment theory should be an important base for understanding the origin of coping The means, standard deviations, and correlations between all the variables that will be included in the model are strategies (e.g., Howard & Medway, 2004; Seiffge-Krenke, 2011). Adolescents securely attached to parents or to other outlined in Table 3. attachment figures may develop a more positive self- image in the long term, may be able to better manage SEM with latent constructs stressful experiences, and develop more appropriate coping styles, such as problem-focused coping (Blomgren, Svahn, The criteria for acceptable model fit for these goodness-of-fit indices were defined by CFI ≥ 0.90; TLI ≥ 0.90; RMSEA < Åström, & Rönnlund, 2016). Consequently, the second set of analysis tested a model 0.08 (Hu & Bentler, 1999). Therefore, in the first step, the measurement model showed a very good model fit: to see if coping strategies could mediate the relationship between attachment and gambling behavior and therefore CFI = 0.968; TLI = 0.963; RMSEA = 0.031. In the second step, the structural model was tested. To advance knowledge on the influence of specific variables in youth problem gambling. To further understand how these examine mediation, bootstrapping procedures were con- ducted to determine the indirect effect (Preacher & Hayes, variables can impact each other, the three attachment dimen- sions of angry distress, availability, and goal-corrected 2008). The bootstrapping procedure has advantages over Baron and Kenny’s(1986) traditional approach and Sobel’s partnership, comprising the three subscales of the AAQ are considered. Therefore, based on the above analysis and (1982) test, because it does not assume normality of the sampling distribution of the indirect effects, and it has higher previous literature showing a positive relationship between attachment and more healthy coping strategies, it was power while maintaining adequate control over type I error rate (MacKinnon, Lockwood, Hoffman, West, & Shets, hypothesized that (i) attachment dimensions would not have a significant direct effect on youth problem gambling; (ii) 2002; Preacher & Hayes, 2008). The bootstrap estimates were based on 1,000 bootstrap samples. An indirect effect the negative attachment dimension of angry distress would have a significant indirect effect on youth problem gam- was considered to be significant if its 95% bias-corrected and accelerated (BCa) bootstrap CIs from 1,000 bootstrap sam- bling, through emotion-focused coping style and problem- focused coping style; and (iii) the positive attachment ples exclude zero (Fritz, Taylor, & MacKinnon, 2012). The full structural model again showed a very good model fit: dimensions of availability and goal-corrected partnership would have a significant indirect effect on youth problem CFI = 0.969; TLI = 0.965; RMSEA = 0.030. The direct path from attachment dimensions to problem gambling, through emotion-focused and problem-focused coping style. gambling was non-significant. As mentioned above, the attachment dimensions of availability and goal-corrected Statistical analysis (2). In the second set of analyses, structural equation modeling (SEM) was used to assess the partnership were reversed, so higher scores on these Table 3. Correlations between all the attachment dimensions, copying styles, and problem gambling MSD 1234 5 6 1. Angry distress 1.90 0.93 _ 0.526** 0.385** −0.091** 0.205** 0.057 2. Availability 2.10 0.92 0.526** _ 0.616** −0.205** 0.093** 0.025 3. Goal-corrected partnership 1.81 0.75 0.385** 0.616** _ −0.244** −0.006 0.041 4. Problem-focused coping 7.44 2.92 −0.091** −0.205** −0.244** _ 0.049 −0.149** 5. Emotion-focused coping 4.34 2.86 0.205** 0.093** −0.006 0.049 _ 0.202** 6. Problem gambling 0.72 1.55 0.057 0.025 0.041 −0.149** 0.202** _ **Correlation is significant at the p < .01 level. 652 | Journal of Behavioral Addictions 6(4), pp. 648–657 (2017) A model to predict youth problematic gambling dimensions reveal low levels of availability and goal- (B = 0.24; SE = 0.15; p = .11). Therefore, these results corrected partnership, respectively. The direct path from indicate that emotion-focused coping fully mediated the emotion-focused coping to problem gambling was positive- relationship between angry distress and problem gambling. ly significant and the direct path from problem-focused Moreover, the results indicate that the specific indirect effect coping was negatively significant (Table 4). Therefore, the from angry distress to problem gambling via problem- first hypothesis was fully supported. focused coping was not significant (B = −0.01; 95% BCa In addition, results relating to the second hypothesis CI = −0.06, 0.02). Consequently, the second hypothesis was indicated that the total effect from the attachment dimen- partially confirmed (see Figure 1 for the significant indirect sion of angry distress to problem gambling was significant path from angry distress to problem gambling). (B = 0.35; SE = 0.12; p < .05). The results also showed that The results for the third hypothesis indicated that the total the total indirect effect was also significant (B = 0.12; 95% effect from availability to problem gambling was not signifi- BCa CI = 0.03, 0.19). Examining the indirect effect, the cant (B = −0.197; SE = 0.19; p = .31). The total indirect results indicated that the specific indirect effect from angry effect was not significant either (B = 0.03; 95% BCa distress to problem gambling through emotion-focused CI = −0.05, 0.16). With regard to the attachment dimension coping was significant (B = 0.13; 95% BCa CI = 0.06, of goal-corrected partnership, the total effect from this di- 0.20). However, as noted in Table 4, the direct effect from mension to problem gambling was non-significant (B = 0.07; angry distress to problem gambling was not significant SE = 0.14; p = .60). In addition, the total indirect effect from Table 4. Direct paths to all dependent variables in the study (unstandardized regression coefficients) BSE p Direct paths to problem gambling Angry distress problem gambling 0.24 0.15 .11 Availability problem gambling −0.22 0.27 .41 Goal-corrected partnership problem gambling 0.11 0.199 .58 Emotion-focused coping problem gambling 0.92 0.24 <.001 Problem-focused coping problem gambling −0.23 0.06 <.001 Direct paths to emotion-focused coping Angry distress emotion-focused coping 0.14 0.06 .023 Availability emotion-focused coping −0.00 0.08 .97 Goal-corrected partnership emotion-focused coping −0.07 0.05 .20 Direct paths to problem-focused coping Angry distress problem-focused coping 0.05 0.11 .62 Availability problem-focused coping −0.12 0.24 .62 Goal-corrected partnership problem-focused coping −0.12 0.20 .56 Note. The values represented in bold are statistically significant. Figure 1. Indirect effects from the three attachment dimensions to problem gambling. ns: non-significant. *p > .05. ***p > .001 Journal of Behavioral Addictions 6(4), pp. 648–657 (2017) | 653 Calado et al. goal-corrected partnership to problem gambling, via the two have an indirect effect on youth problematic gambling via coping styles was not significant either (B = −0.04; 95% BCa other individual variables. As previous research has CI = −0.14, 0.04). Therefore, these results suggest that there demonstrated that attachment has an influence on coping is no mediation of coping styles in the relationship between strategies (Seiffge-Krenke & Beyers, 2005), and that a the attachment dimensions of angry distress and goal- primary function of interpersonal attachment is the regu- corrected partnership and problem gambling. This model lation of emotions (McNally, Palfai, Levine, & Moore, accounted for a total of 22% of the variance in problem 2003), the second set of analyses examined if coping gambling, 14.6% of the variance in emotion-focused coping, styles mediated the relationship between attachment to and 4.3% in the variance of problem-focused coping. parents and problem gambling. To better understand the complex relationships between these variables, the second set of analyses examined the three DISCUSSION dimensions of attachment (angry distress, availability, and goal-corrected partnership). The findings indicated that This study explored the effect of individual and family none of these dimensions had a significant direct effect on predictors upon problem gambling, and examined the rela- problem gambling. However, the attachment dimension of tionship between some individual and family variables in angry distress exerted a significant indirect effect on problem the prediction of problem gambling among a student sam- gambling via emotion-focused coping, and which fully me- ple. The first set of analyses showed that a model composed diated the relationship between angry distress and problem of individual and family factors together adequately gambling. Problem-focused coping did not exert any media- explained youth problematic gambling. The second set of tion in the relationship between angry distress and problem analyses demonstrated that attachment dimensions do not gambling. In addition, the other attachment dimensions of have a significant direct effect on problem gambling among availability and goal-corrected partnership did not exert any young people, and that the attachment dimension of angry significant indirect effect on youth problem gambling via any distress exerted a significant indirect influence on youth coping style. Therefore, these results suggest that the attach- problem gambling. ment dimension of angry distress, characterized by feelings The findings of the first set of analyses are in line with of anger toward attachment figures when attachment needs other research (e.g., Griffiths, 2011), which asserts that are frustrated (West et al., 1998), has a major effect in the gambling is a multidimensional (rather than a unitary) emergence of emotion-focused coping, characterized by phenomenon and therefore many factors may come into strategies of regulating emotions caused by the stressor. This, play in the acquisition, development, and maintenance of in turn, will exert an influence in the development of gambling-related problems. Within this integrated perspec- gambling-related problems among young people. tive, the most significant variables for predicting youth Although there was some preliminary evidence that problem gambling were male gender, cognitive distortions, attachment plays a role in adolescent gambling behaviors and emotion-focused coping. These findings are also in line (Magoon & Ingersoll, 2006), these findings extend the with previous research showing that problematic gambling previous gambling literature by examining how different behavior is more prevalent among males (e.g., Bastiani attachment dimensions can exert an effect on youth problem et al., 2013; Griffiths, 2011, Skokauskas & Satkeviciute, gambling via coping styles, and by showing that the attach- 2007), associated with erroneous beliefs of randomness ment dimension of angry distress indirectly influenced this (e.g., Delfabbro et al., 2006; Griffiths, 1994) and with behavior via an emotion-focused coping style. emotion-focused coping (e.g., Bergevin et al., 2006). How- The findings of this study have some important implica- ever, in this study, attitudes did not show a significant effect tions for clinical practice and prevention. In fact, it seems on youth problematic gambling, contradicting previous that a low-quality relationship with parents or other attach- research highlighting the impact of this variable in predict- ment figures may lead youngsters to learn less suitable ing gambling-related problems among young people. There- strategies to deal with their life difficulties, such as using fore, more research into the effect of attitudes on youth gambling to escape from their problems (Wood & Griffiths, gambling is needed among diverse samples, to understand 2007). Therefore, in a clinical context with young problem the meaning of gambling-related attitudes in different cul- gamblers, therapists should assess the quality of the parent– tural contexts and its influence in the emergence of this child relationship, namely potential feelings of anger that behavior. adolescents might feel toward their parents when their needs The first set of analyses also showed that the family are unfulfilled, and must include parents or other attachment predictors of attachment to parents or other attachment figures in the therapeutic process. In addition, during pre- figures, and family structure did not show a significant ventive initiatives, parents should also be trained about their effect on youth problem gambling. These findings con- abilities and aptitudes that could foster the development of a firm other studies showing that family structure does not positive relationship with their children. have a significant impact on youth problem gambling Although this study has some strengths, such as a (e.g., Langhinrichsen-Rohling et al., 2004) but contra- relatively large sample size and the examination of previ- dicts other studies showing that attachment has a signifi- ously unexplored relationships between attachment dimen- cant effect on other adolescent risky behaviors (e.g., Bahr sions, coping styles, and problem gambling, it has also some et al., 2005). Consequently, these findings suggest that limitations. These should be kept in mind when interpreting attachment to parents or other attachment figures may the findings. Most importantly, this study exclusively 654 | Journal of Behavioral Addictions 6(4), pp. 648–657 (2017) A model to predict youth problematic gambling utilized self-report data, which are prone to various well- Ariyabuddhiphongs, V. (2013). Adolescent gambling: A narrative known biases, such as social desirability and memory recall review of behavior and its predictors. International Journal of biases. Second, the study was conducted among a student Mental Health and Addiction, 11(1), 97–109. doi:10.1007/ sample recruited in England and therefore generalizability to s11469-012-9401-6 other populations is limited. Third, this study used a cross- Bahr, S., Hoffman, J., & Yang, X. (2005). Parental and peer sectional design, and thus possible causal relationships influences on the risk of adolescent drug use. Journal of between variables cannot be inferred. Primary Prevention, 26(6), 529–551. doi:10.1007/s10935- Despite these limitations, this is the first study, as far as the 005-0014-8 authors are aware, to examine the mediation effect of coping Baron, R., & Kenny, D. (1986). The moderator-mediator variable styles in the relationship between attachment and problem distinction in social psychological research: Conceptual, stra- gambling. This confirms that the indirect effects of family tegic, and statistical considerations. Journal of Personality and variables are important in estimating the complex social Social Psychology, 51(6), 1173–1182. doi:10.1037/0022- forces that may influence adolescent decisions to gamble. 3514.51.6.1173 However, future studies should be conducted in other coun- Bastiani, L., Gori, M., Colasante, E., Siciliano, V., Capitanucci, D., tries in different contexts and with a wider range in age to Jarre, P., & Molinaro, S. (2013). Complex factors and beha- extend the present findings to other youth populations. viours in the gambling population of Italy. Journal of Gam- Moreover, there is a need for replication of these results in bling Studies, 29(1), 1–13. doi:10.1007/s10899-011-9283-8 longitudinal designs. Considering that youth problem gam- Bergevin, T., Gupta, R., Derevensky, J., & Kaufman, F. (2006). bling has several negative consequences, longitudinal studies Adolescent gambling: Understanding the role of stress and as well as investigations carried out in other contexts could be coping. Journal of Gambling Studies, 22(2), 195–208. doi:10. of great utility in minimizing these outcomes. 1007/s10899-006-9010-z Blomgren, A. S., Svahn, K., Åström, E., & Rönnlund, M. (2016). Coping strategies in late adolescence: Relationships to parental attachment and time perspective. Journal of Genetic Psychol- Funding sources: FC received a grant from FCT, Portu- ogy, 177(3), 85–96. doi:10.1080/00221325.2016.1178101 guese national funding agency for science, research and Bronfrenbrenner, U., & Morris, P. A. (1998). The ecology of technology (reference number SFRH/BD/119749/2016). developmental processes. In Lerner, R. (Ed.), Handbook of MDG has received funding for a number of research child psychology: Theoretical models of human development projects in the area of gambling education for young people, (5th ed., Vol. 1, pp. 993–1028). New York, NY: John Wiley. social responsibility in gambling, and gambling treatment Calado, F., Alexandre, J., & Griffiths, M. D. (2014). Mom, Dad it’s from the Responsibility in Gambling Trust, a charitable only a game! Perceived gambling and gaming behaviors body, which funds its research program based on donations among adolescents and young adults: An exploratory study. from the gambling industry. He also undertakes consultancy International Journal of Mental Health and Addiction, 12(6), for various gaming companies in the area of social respon- 772–794. doi:10.1007/s11469-014-9509-y sibility in gambling. Calado, F., Alexandre, J., & Griffiths, M. D. (2017). Prevalence of adolescent problem gambling: A systematic review of recent Authors’ contribution: FC and MDG designed the study. FC research. Journal of Gambling Studies, 33(2), 397–424. doi:10. collected and analyzed the data. All authors inputted into the 1007/s10899-016-9627-5 interpretation of the results. FC wrote the initial draft of the Carver, C. S. (1997). You want to measure coping but your paper and MDG added significantly to the first draft. All protocol’s too long: Consider the brief COPE. International authors then went through a number of iterative versions Journal of Behavioral Medicine, 4(1), 92–100. doi:10.1207/ before the final manuscript was submitted. All authors have s15327558ijbm0401_6 full access to all the data in the study and take responsibility Carver, C. S., Scheier, M. F., & Weintraub, J. K. (1989). Assessing for the integrity of the data and the accuracy of the data coping strategies: A theoretically based approach. Journal of analysis. Personality and Social Psychology, 56(2), 267–283. doi:10.1037/0022-3514.56.2.267 Conflict of interest: FC and JA declare no conflict of Delfabbro, P., Lahn, J., & Grabosky, P. (2006). It’s not what you interest. know, but how you use it: Statistical knowledge and adolescent problem gambling. Journal of Gambling Studies, 22(2), 179– Acknowledgements: The authors would like to thank 193. doi:10.1007/s10899-006-9009-5 Tadeusz Borejko for his assistance in participant recruitment Delfabbro, P., & Thrupp, L. (2003). The social determinants of and data collection. youth gambling in South Australian adolescents. Journal of Adolescence, 26(3), 313–330. doi:10.1016/S0140-1971(03) 00013-7 REFERENCES Dodig, D. (2013). Assessment challenges and determinants of adolescents’ adverse psychosocial consequences of gambling. Anderson, J. C., & Gerbing, D. W. (1988). Structural equation Kriminologija i Socijalna Integracija, 21, 1–29. modeling in practice: A review and recommended two-step Endler, N. S., & Parker, J. D. A. (1990). Coping Inventory for approach. Psychological Bulletin, 103(3), 411–423. doi:10. Stressful Situations (CISS): Manual. Toronto, Canada: Multi- 1037/0033-2909.103.3.411 Health Systems. Journal of Behavioral Addictions 6(4), pp. 648–657 (2017) | 655 Calado et al. Fisher, S. (2000). Developing the DSM-IV-MR-J criteria to iden- Kuntsche, E. N., & Kuendig, H. (2006). What is worse? A tify adolescent problem gambling in nonclinical populations. hierarchy of family-related risk factors predicting alcohol use Journal of Gambling Studies, 16(2–3), 253–273. doi:10.1023/ in adolescence. Substance Use and Misuse, 41(1), 71–86. A:1009437115789 doi:10.1080/10826080500368694 Flay, B. R., & Petraitis, J. (1994). The theory of triadic influence: A Langhinrichsen-Rohling, J., Rohde, P., Seeley, J. R., & Rohling, new theory of health behavior with implications for preventive M. L. (2004). Individual, family and peer correlates of adoles- interventions. Advances in Medical Sociology, 4, 19–44. cent gambling. Journal of Gambling Studies, 20(1), 23–46. Folkman, S., & Lazarus, R. A. (1985). If it changes it must be doi:10.1023/B:JOGS.0000016702.69068.53 a process: A study of emotion and coping during three MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., stages of a college examination. Journal of Personality and & Sheets, V. (2002). A comparison of methods to test media- Social Psychology, 48(1), 150–170. doi:10.1037/0022- tion and other intervening variable effects. Psychological 3514.48.1.150 Methods, 7(1), 83–104. doi:10.1037/1082-989X.7.1.83 Fritz, M. S., Taylor, A. B., & MacKinnon, D. P. (2012). Explana- Magoon, M. E., & Ingersoll, G. M. (2006). Parental modeling, tion of two anomalous results in statistical mediation analysis. attachment, and supervision as moderators of adolescent gam- Multivariate Behavioral Research, 47(1), 61–87. doi:10.1080/ bling. Journal of Gambling Studies, 22(1), 1–22. doi:10.1007/ 00273171.2012.640596 s10899-005-9000-6 Froberg, F. (2006). Gambling among young people. A knowledge McComb, J. L., & Sabiston, C. M. (2010). Family influences on review. Stockholm, Sweden: Swedish National Institute of adolescent gambling behavior: A review of the literature. Public Health. Journal of Gambling Studies, 26(4), 503–520. doi:10.1007/ Griffiths, M. D. (1994). The role of cognitive bias and skill in fruit s10899-010-9181-5 machine gambling. British Journal of Psychology, 85(3), 351– McNally, A. M., Palfai, T. P., Levine, R. V., & Moore, B. M. 369. doi:10.1111/j.2044-8295.1994.tb02529.x (2003). Attachment dimensions and drinking-related problems Griffiths, M. D. (1995). Adolescent gambling. London, UK: among young adults. The mediational role of coping motives. Routledge. Addictive Behaviors, 28(6), 1115–1127. doi:10.1016/S0306- Griffiths, M. D. (2011). Adolescent gambling. In Bradford Brown, 4603(02)00224-1 B. & Prinstein, M. J. (Eds.), Encyclopedia of adolescence (Vol. Miller, H. V., Jennings, W. G., Alvarez-Rivera, L. L., & Lanza- 3, pp. 11–20). San Diego, CA: Academic Press. Kaduce, L. (2009). Self-control, attachment, and deviance Gupta, R., Derevensky, J., & Marget, N. (2004). Coping strategies among Hispanic adolescents. Journal of Criminal Justice, employed by adolescents with gambling problems. Child and 37(1), 77–84. doi:10.1016/j.jcrimjus.2008.12.003 Adolescent Mental Health, 9(3), 115–120. doi:10.1111/j.1475- Moore, S. M., & Ohtsuka, K. (1999). Beliefs about control over 3588.2004.00092.x gambling among young people, and their relation to problem Gupta, R., & Derevensky, J. L. (2000). Adolescents with gambling gambling. Psychology of Addictive Behaviors, 13(4), 339–347. problems: From research to treatment. Journal of Gambling doi:10.1037/0893-164X.13.4.339 Studies, 16(2–3), 315–342. doi:10.1023/A:1009493200768 Muthén, L. K., & Muthén, B. O. (2015). Mplus user’s guide (7th Hayer, T., & Griffiths, M. D. (2015). The prevention and treatment ed.). Los Angeles, CA: Muthén & Muthén. of problem gambling in adolescence. In Gullotta, T. P. & Olason, D. T., Kristjansdottir, E., Einarsdottir, H., Haraldsson, H., Adams, G. (Eds.), Handbook of adolescent behavioural pro- Bjarnason, G., & Derevensky, J. L. (2011). Internet gambling blems: Evidence-based approaches to prevention and treat- and problem gambling among 13 to 18 year old adolescents in ment (2nd ed., pp. 539–558). New York, NY: Springer. Iceland. International Journal of Mental Health and Addiction, Howard, M. S., & Medway, F. J. (2004). Adolescents’ attachment 9(3), 257–263. doi:10.1007/s11469-010-9280-7 and coping with stress. Psychology in the Schools, 41(3), 391– Potenza, M. N., Wareham, J. D., Steinberg, M. A., Rugle, L., 402. doi:10.1002/pits.10167 Cavallo, D. A., Krishnan-Sarin, S., & Desai, R. A. (2011). Hu, L., & Bentler, P. M. (1999). Cutoff criteria in fix indexes in Correlates of at-risk/problem internet gambling in adolescents. covariance structure analysis: Conventional criteria versus new Journal of American Academy of Child and Adolescent Psy- alternatives. Structural Equation Modeling, 6(1), 1–55. doi:10. chiatry, 50(2), 150–159.e3. doi:10.1016/j.jaac.2010.11.006 1080/10705519909540118 Preacher, K., & Hayes, A. (2008). Asymptotic and resampling Jackson, A. C., Dowling, N., Thomas, S. A., Bond, L., & Patton, G. strategies for assessing and comparing indirect effects in (2008). Adolescent gambling behaviour and attitudes: A multiple mediator models. Behavior Research Methods, prevalence study and correlates in an Australian population. 40(3), 879–891. doi:10.3758/BRM.40.3.879 International Journal of Mental Health and Addiction, 6(3), Raylu, N., & Oei, T. P. (2004). The Gambling Related Cognitions 325–352. doi:10.1007/s11469-008-9149-1 Scale (GRCS): Development, confirmatory factor validation Kopak, A. M., Chen, A. C. C., Haas, S. A., & Gillmore, M. R. and psychometric properties. Addiction, 99(6), 757–769. (2012). The importance of family factors to protect against doi:10.1111/j.1360-0443.2004.00753.x substance use related problems among Mexican heritage and Reinecke, L. (2009). Games and recovery: The use of video and White youth. Drug and Alcohol Dependence, 124(1–2), 34–41. computer games to recuperate from stress and strain. Journal of doi:10.1016/j.drugalcdep.2011.12.004 Media Psychology, 21(3), 126–142. doi:10.1027/1864-1105. Kristiansen, S. G., & Jensen, S. M. (2014). Prevalence and 21.3.126 correlates of problematic gambling among Danish adolescents. Seiffge-Krenke, I. (2011). Coping with relationship stressors: A International Journal of Social Welfare, 23(1), 89–99. decade review. Journal of Research on Adolescence, 21(1), doi:10.1111/ijsw.12021 196–210. doi:10.1111/j.1532-7795.2010.00723.x 656 | Journal of Behavioral Addictions 6(4), pp. 648–657 (2017) A model to predict youth problematic gambling Seiffge-Krenke, I., & Beyers, W. (2005). Coping trajectories from Vandenberg, R. J. (2006). Statistical and methodological myths adolescence to young adulthood: Links to attachment state of and urban legends: Where, pray tell, did they get this idea? mind. Journal of Research on Adolescence, 15(4), 561–582. Organizational Research Methods, 9(2), 194–201. doi:10. doi:10.1111/j.1532-7795.2005.00111.x 1177/1094428105285506 Shead, N. W., Derevensky, J. L., & Gupta, R. (2010). Risk and Wardle, H., Moody, A., Spence, S., Orford, J., Volberg, R., protective factors associated with youth problem gambling. Jotangia, D., Griffiths, M., Hussey, D., & Dobbie, F. International Journal of Adolescent Medicine and Health, 22, (2011). The British gambling prevalence survey 2010. London, 39–58. UK: The Stationery Office. Skokauskas, N., & Satkeviciute, R. (2007). Adolescent pathologi- West,W., Rose,S.M., Spreng,S., Sheldon-Keller,A., &Adam, cal gambling in Kaunas, Lithuania. Nordic Journal of Psychi- K. (1998). Adolescent attachment questionnaire: A brief atry, 61(2), 86–91. doi:10.1080/08039480701226054 assessment of attachment in adolescence. Journal of Youth Sobel, M. (1982). Asymptotic confidence intervals for indirect and Adolescence, 27(5), 661–673. doi:10.1023/A:102289 effects in structural equation models. Sociological Methodology, 1225542 13, 347–355. doi:10.2307/270723 Wood, R. T., & Griffiths, M. D. (2004). Adolescent lottery and Tang, C. S.-K., & Wu, A. M. S. (2012). Gambling-related cognitive scratchcard players: Do their attitudes influence their gambling biases and pathological gambling among youths, young adults, behaviour? Journal of Adolescence, 27(4), 467–475. doi:10. and mature adults, in Chinese societies. Journal of Gambling 1016/j.adolescence.2003.12.003 Studies, 28(1), 139–154. doi:10.1007/s10899-011-9249-x Wood, R. T., & Griffiths, M. D. (2007). A qualitative investigation Turner, N. E., Macdonald, J., Bartoshuk, M., & Zangeneh, M. of problem gambling as an escape-based coping strategy. (2008). Adolescent gambling behaviour, attitudes, and gam- Psychology and Psychotherapy: Theory, Research and Prac- bling problems. International Journal of Mental Health and tice, 80(1), 107–125. doi:10.1348/147608306X107881 Addiction, 6(2), 223–237. doi:10.1007/s11469-007-9117-1 Journal of Behavioral Addictions 6(4), pp. 648–657 (2017) | 657

Journal

Journal of Behavioral Addictions – Pubmed Central

Published: Oct 23, 2017

There are no references for this article.