Relationships with internet misuse, gaming abuse and emotional intelligence: study

Problem gambling in adolescence: Relationships with internet misuse, gaming abuse and emotional intelligence

Examined the relationship between emotional intelligence (EI) and several addiction-related behaviors (gambling, internet use, and video game playing) in two community-based samples of adolescents: 13–15 year olds (N = 209) and 16–18 year olds (N = 458). EI was measured using the youth version of the Emotional Quotient Inventory (EQ-i:YV; Bar-On & Parker, 2000), while the addiction-related behaviors were assessed using the Internet Addiction Questionnaire (IADQ; Young, 1998), the Problem Video Game Playing Scale (PVGS; Salguero & Moran, 2002), and the South Oaks Gambling Screen-Revised for Adolescents (SOGS-RA; Winters, Stinchfield, & Fulkerson, 1993). EI was found to be a moderate to strong predictor of addiction-related behaviors in both groups (parameter estimates were -.76 for the younger adolescents and -.56 for the older adolescents).

Keywords: Adolescence; Gambling; Emotional intelligence

Despite the fact that many people perceive problem gambling to be an issue prevalent only in adults, recent research indicates that problem and pathological gambling pose serious concerns among adolescents (Dickson et al., 2004 L. Dickson, J.L. Derevensky and R. Gupta, Youth gambling problems: Reduction prevention model, Addiction Research and Theory 12 (2004), pp. 305–316. View Record in Scopus | Cited By in Scopus (3)Dickson, Derevensky, & Gupta, 2004). Like adult gambling, adolescent gambling has been linked with a number of negative outcomes including criminal behavior, poor academic achievement, school truancy (Stitt, Giacopassi, & Vandiver, 2000), financial problems, depression, suicide, deterioration of social relationships (Messerlian, Derevensky, & Gupta, 2005), and substance abuse (Griffiths & Wood, 2000). The prevalence of problem gambling among adolescents has been shown to be 2–4 times that of adults. Recent Canadian studies conducted with large community samples have estimated that 4–8% of adolescents currently have a severe gambling problem (Dickson et al., 2004). In addition, 10–15% of adolescents gamble excessively, making them vulnerable to the development of a more serious gambling problem (Dickson et al., 2004). While some researchers have argued that these prevalence rates may be overestimated ([Derevensky et al., 2003] and [Ladouceur et al., 2005]), there is ample evidence that gambling among adolescents has increased over the past two decades ([Langhinrichsen-Rohling et al., 2004] and [Messerlian et al., 2005]).

One of the reasons that it is difficult to get a clear understanding of the true nature of gambling problems in adolescents is the unique clinical picture. In adults, one of the central problems associated with increased gambling behaviors is financial: the individual is preoccupied with gambling, they gamble for increasing amounts of money, they may attempt to hide the financial implications of their behavior, and they may begin to gamble with both legal and illegal financial sources. In adolescent populations, however, the financial implications may be less important. With more limited financial resources and gambling opportunities, the central problem may relate to the large amounts of time being devoted to gambling-related behaviors. An adolescent playing on-line poker for free, for example, generates limited financial risk. On the other hand, an escalation in the time commitment involved in this gambling behavior may have a considerable negative impact on factors as diverse as the quality of social relationships and school performance.

Contributing to the clinical picture in adolescence is a blurring of gambling-related behaviors and other recreational activities like internet use, computer games, and video gaming. For example, a group of adolescents play a new video game all day, with the loser buying the rest of the group pizza. Is this gambling or gaming behavior? There is empirical evidence that addiction-related problem behaviors including video/computer game and internet use are an increasing problem (Mitchell, Becker-Blease, & Finkelhor, 2005). Together with problem gambling, these types of dysfunctional preoccupations appear to be on the increase among adolescents ([Griffiths and Wood, 2000] and [Pratarelli et al., 1999]). In a study conducted by Mitchell et al. (2005), for example, 15% of individuals classified as having an internet addiction were involved with online gambling and game playing.

While the study of gambling has rapidly expanded over the past three decades there have been relatively few investigations of gambling determinants (Shaffer, LaBrie, LaPlante, Nelson, & Stanton, 2004). Studies have identified the lack of affect regulation abilities as an important risk factor for the development of problem gambling and other addiction-related problems among adults ([Coman et al., 1997], [Lumley and Roby, 1995] and [Taylor et al., 1997]). Lumley and Roby (1995) have suggested that individuals with poor affect regulation abilities may struggle with regulating distressing emotions due to difficulty identifying subjective emotional states, and a limited ability to communicate these feelings to others. As a result, these individuals are unable to obtain the help or comfort needed from others. In addition, poor affect regulation abilities may prevent the individual from using fantasy or other mental activities to modify their own distressing emotional states. Lumley and Roby (1995) hypothesized that these individuals use compulsive behaviors to regulate their emotions. In order to test this hypothesis, Lumley and Roby (1995) looked at the association between cognitive and affective aspects of alexithymia and pathological gambling in a large sample of undergraduate students. Alexithymia is a personality trait associated with a wide range of affect regulation deficits including difficulty identifying feelings, difficulty describing feelings, and a concrete, externally oriented thinking style (Taylor et al., 1997). Lumley and Roby (1995) found that the prevalence of alexithymia was significantly higher in pathological gamblers than in controls, and concluded that there was a significant association between alexithymia and affect regulation. These results were supported by a more recent study using a similar sample of young adults (Parker, Wood, Bond, & Shaughnessy, 2005).

In the past, researchers have suggested that individuals with alexithymia use compulsive behaviors to regulate their emotions, a hypothesis supported by studies linking alexithymia to other addiction-related problems including drug and alcohol abuse ([Ladouceur et al., 1994], [Pinard et al., 1996] and [Uzun, 2003]). It is unclear what role alexithymia may play in adolescent gambling behaviors, since it is unclear whether alexithymia can reliably be measured in this group. Nevertheless, emotional intelligence (a construct that conceptually overlaps with alexithymia; Parker, Taylor, & Bagby, 2001) has been found to be negatively associated with adolescent tobacco and alcohol use (Trinidad & Johnson, 2002), as well as behaviors linked with gambling – deviant school behavior (Petrides, Fredrickson, & Furnham, 2004) and poor academic achievement (Parker et al., 2004).

Although a number of distinct and overlapping conceptual models have been proposed for the emotional intelligence construct (see Matthews, Zeidner, & Roberts, 2007), most models include skills like the ability to accurately appraise emotion in self or other, the ability to express emotion, the ability to effectively regulate emotion, and the ability to use feelings to guide future behaviors (Parker et al., 2001). Petrides and Furnham (2000) have noted that while various models for emotional intelligence have been proposed, most assessment tools fall into one of two broad types. The first type is ability measures of emotional intelligence, like the Mayer–Salovey–Caruso Emotional Intelligence Test; (MSCEIT; Mayer, Salovey, & Caruso, 2002), that assess relevant abilities using a performance based methodology. The second type is trait emotional intelligence measures, like the EQ-i (Bar-On, 1997), that assess core competencies via a self-report format. Regardless of format, however, assessment tools have been adapted for use with children and adolescents (e.g., Bar-On & Parker, 2000), and the emotional intelligence construct has been found to be inversely related to alexithymia ([Mayer et al., 2002] and [Parker et al., 2001]).

The purpose of the present study was to provide further evidence of a relationship between problem gambling and affect regulation abilities (assessed using a measure of emotional intelligence) in a large adolescent sample. In addition, the present study also examined the relationship between gambling behavior and several other addiction-related behaviors that appear to be on the increase among adolescents ([Griffiths and Wood, 2000] and [Mitchell et al., 2005]): internet use and computer/video gaming behavior.


Older adolescents (16–18 years of age) scored higher than younger adolescents (13–15 years of age) on most of the EI dimensions. Females were also found to score higher than males on the intrapersonal and interpersonal scales, while males scored higher than females on the adaptability scale. The intrapersonal scale assesses abilities like recognizing and understanding one’s feelings, while the interpersonal scale taps abilities like empathy and recognizing emotions in others. The adaptability scale assesses abilities like being able to adjust one’s emotions and behaviors to changing situations and conditions. All of the gender differences are consistent with previous research using the EQ-i:YV (Bar-On & Parker, 2000); the results are also consistent with results from previous research using the adult version of the scale ([Bar-On, 1997] and [Bar-On, 2002]).

Males were found to score higher than females on the measures of problem gaming (PVGS) and gambling behaviors (SOGS-RA). These results are not surprising, since most video games are designed for male users (Yang, 2001) and gender differences among gamblers is a well documented finding, with males beginning to gamble at an earlier age, gambling more often, and spending more time and money ([Griffiths and Wood, 2000], [Hardoon and Derevensky, 2002] and [Jacobs, 2000]).

More surprising is the lack of a gender difference for internet use (IADQ). This finding contradicts the results of Ko, Yen, Chen, Chen, and Yen (2005), who found that males use the internet more often, and for longer periods of time than do females. It is possible that because the present study did not distinguish between different internet activities (e.g., chatting, gaming, shopping, viewing pornography, information seeking, etc.), gender differences for varying types of activities cancelled each other out. Yang (2001), for example, reports that adolescent males are more likely to use the internet to play games, while adolescent females are more likely to use the internet to chat with friends.

The present study found empirical evidence that several different variables with strong “addiction” potential (i.e. video/computer game use, internet use, and gambling) can be accounted for by a single latent variable that we have labelled “dysfunctional preoccupation”. This latent variable emerged in both the younger and older adolescent samples, accounting for the greatest amount of variability in addiction-related behaviors in the older adolescent sample. The presence of a “dysfunctional preoccupation” dimension has important implications for intervention and prevention strategies directed at youth. Gambling problems, excessive internet use, and gaming addiction are typically treated as separate mental health issues, with unique etiologies. Intervention and prevention strategies for youth might be more effective if they simultaneously targeted a broad range of addiction-related behaviors.

One of the central goals of the present study was to examine the relationship between emotional intelligence and a constellation of addiction-related behaviors (gambling, gaming, and internet use) in an adolescent community sample. The path analysis revealed that EI is a moderate to strong predictor of addiction-related behaviors. EI was found to account for 58% and 31% of the variance in the addiction-related behaviors for the younger and older groups, respectively. This finding is consistent with previous work that has found affect regulation abilities to be important risk factors in the development of problem gambling and other addiction-related problems ([Coman et al., 1997], [Griffiths and Wood, 2000], [Jacobs, 2000], [Kim et al., 2006], [Lumley and Roby, 1995] and [Mitchell et al., 2005]).

Interpersonal abilities would appear to be the most important EI dimension for addiction-related behaviors, as it produced the highest correlation with the younger adolescents’ scores on the PVGS (-.32) and SOGS-RA (-.26), as well as the older adolescents’ scores on the PVGS (-.36), IADQ (-.29), and SOGS-RA (-.32). It may be that individuals deficient in interpersonal abilities engage in more of these addiction-related behaviors; it may also be the case that adolescents who spend considerable amounts of time engaging in these behaviors do not develop good interpersonal abilities. Further research is needed to determine the direction of this possible relationship.

One limitation of this study is that it did not distinguish between gambling types or internet activities. As mentioned earlier, the lack of distinction between internet activities may have caused results to cancel each other out. As suggested by Morahan-Martin (2005), focus on the internet may be misleading, as the problem is not the internet so much as the activities pursued while on the internet. The same is true for gambling. Many forms of gambling exist, from bets made among friends, to lottery tickets, to the more formal types of games found in casinos. It may be that different results would be found if gambling was broken down into its many forms. Furthermore, since males were found to score higher on all of the addiction-related behavior scales, distinguishing between activities may also reveal more information about the behaviors of females as previous research has found that males and females prefer different types of gambling (Jacobs, 2000) and internet (Yang, 2001) activities. Future research should be focused on distinguishing between these activities and replicating this study on a finer scale.

There are some other noteworthy limitations to mention about the present study. First, the present study used a rather homogeneous sample of adolescents (the majority of adolescents were white), which therefore limits the generalizability of the study. The findings are also limited by the use of a single measure for each key variable. The study needs to be replicated using other measures of key variables, using a broader range of assessment strategies (e.g., self-report, performance based, or observer ratings). As Petrides and Furnham, 2000 K.V. Petrides and A. Furnham, On the dimensional structure of emotional intelligence, Personality and Individual Differences 29 (2000), pp. 313–320. Abstract | PDF (102 K) | View Record in Scopus | Cited By in Scopus (93)Petrides and Furnham (2000) have noted, emotional intelligence measures using either an ability or self-report approach may predict outcome variables quite differently.


This study was supported by research grants to the first author from the Ontario Problem Gambling Research Centre, as well as SSHRC Graduate Scholarships to the second and third authors.


American Psychiatric Association, 1994 American Psychiatric Association, Diagnostic and statistical manual of mental disorders (4th ed.), American Psychiatric Press, Washington, DC (1994).

Bar-On, 1997 R. Bar-On, BarOn Emotional Quotient Inventory: Technical manual, Multi-Health Systems, Toronto, Canada (1997).

Bar-On, 2002 R. Bar-On, BarOn Emotional Quotient Short form (EQ-i:Short): Technical manual, Multi-Health Systems, Toronto (2002).

Bar-On and Parker, 2000 R. Bar-On and J.D.A. Parker, The Bar-On EQ-i:YV: Technical manual, Multi-Health Systems, Toronto, Canada (2000).

Cole, 1987 D.A. Cole, Utility of confirmatory factor analysis in test validation research, Journal of Consulting and Clinical Psychology 55 (1987), pp. 584–594.

Coman et al., 1997 G.J. Coman, G.D. Burrows and B.J. Evans, Stress and anxiety as factors in the onset of problem gambling: Implications for treatment, Stress Medicine 13 (1997), pp. 235–244.

Costa and McCrae, 1992 P.T. Costa and R.R. McCrae, Revised NEO Personality Inventory (NEO-PI-R) and NEO Five-Factor Inventory (NEO-FFI) professional manual, Psychological Assessment Resources, Odessa, FL (1992).

Derevensky et al., 2003 J.L. Derevensky, R. Gupta and K. Winters, Prevalence rates of youth gambling problems: Are the current rates inflated?, Journal of Gambling Studies 19 (2003), pp. 405–425.

Dickson et al., 2004 L. Dickson, J.L. Derevensky and R. Gupta, Youth gambling problems: Reduction prevention model, Addiction Research and Theory 12 (2004), pp. 305–316.

Griffiths and Wood, 2000 M. Griffiths and R.T.A. Wood, Risk factors in adolescence: The case of gambling, videogame playing, and the internet, Journal of Gambling Studies 16 (2000), pp. 199–225.

Hardoon and Derevensky, 2002 K.K. Hardoon and J.L. Derevensky, Child and adolescent gambling behavior: Current knowledge, Clinical Child Psychology and Psychiatry 7 (2002), pp. 263–281.

Personality and Individual Differences
Article in Press, Corrected Proof

James D.A. Parker, Robyn N. Taylor, Jennifer M. Eastabrook, Stacey L. Schell and Laura M. Wood
Department of Psychology, Trent University, Peterborough, Ontario, Canada K9J 7B8

Provided by ArmMed Media