In recent years, the public has noted the negative consequences of online social network (OSN) use. Multiple research studies across different disciplines have also noted the considerable negative effects that some patterns of OSN use, most notably OSN addiction, may bring to users. OSN addiction refers to a maladaptive dependency on OSN use to the extent that it negatively affects users’ important life functions (Turel, Serenko, & Giles, 2011). OSN addiction is considered a part of the umbrella term of “IT addiction” that has received attention from the American Psychiatric Association, with Internet gaming disorder being included as a “condition for further studying” in the DSM-5 (APA, 2013). Equally, it has been the topic of discussion among IT scholars given the popularity of applications such as Facebook, Instagram, and Snapchat (Vaghefi & Qahri-Saremi, 2017). Understanding OSN addiction and the factors that can influence its formation are important given the ubiquitous nature and increasing prevalence of OSNs in our society.
Due to the importance and prevalence of addiction to OSN use, it is not surprising that several studies are being conducted to understand the underlying causes and contributors to the development of such addictions. Among these causes are individuals’ personality traits, due to their stability and predictability. Personality traits—neuroticism, conscientiousness, and agreeableness, for example—are rather stable characteristics that are generally independent of the situation (McElroy, Hendrickson, Townsend, & DeMarie, 2007). This independence from situational contingencies renders personality traits as potentially powerful predictors of behaviors, including addiction. The effects of some personality traits on behavioral and substance use issues have already been recognized in prior studies (e.g., Miller, 1991). Further research also hints at the relevance of personality traits vis-à-vis higher Internet use (Young & Rogers, 2009) and/or smartphone addiction (Vaghefi, Lapointe, & Boudreau-Pinsonneault, 2017). Nonetheless, despite evidence suggesting the importance of personality traits in addiction to OSNs, more research is needed to better understand the effects of personality traits on OSN addiction and the possible interaction effects among personality traits (e.g., one trait exacerbating or weakening the effect of another trait on OSN addiction). To this aim, we designed a study to better understand key personality traits influencing addiction to OSNs and show the complexity of interactions among these factors. We drew on three pertinent factors from the five-factor model (FFM; Carver & Connor-Smith, 2010)—neuroticism, conscientiousness, and agreeableness—to examine their direct effects on, along with their interactions with, addiction to OSNs. We tested our propositions about these potential effects using two rounds of data collection via surveys from OSN users.
In common terms, addiction is “a repetitive habit pattern that increases the risk of disease and/or associated with personal and social problems . . . [which] are often experienced subjectively as ‘loss of control’ [and] continues despite volitional attempts to abstain or moderate use” (Marlatt, Baer, Donovan, & Kivlahan, 1988, p. 224). These days, the term is mostly referring to the maladaptive dependence on a substance or a behavior. For instance, we know that some may develop addictions to eating, shopping, gambling, drinking, smoking, working, and sex (Turel et al., 2011; Young, 1998). OSN addiction—the focus of this study—is considered a behavioral addiction with similar characteristics to other known behavioral addictions, namely salience, withdrawal, conflict, relapse, tolerance, and mood modification. For instance, we have all noticed people around us who spend a considerable amount of time on Facebook. This is in line with recent statistics pointing to an alarming rate of technology and media use, including OSNs, for some up to twelve hours a day (“Time flies,” 2018).
OSN addiction—the focus of this study—is considered a behavioral addiction with similar characteristics to other known behavioral addictions, namely salience, withdrawal, conflict, relapse, tolerance, and mood modification.
As a result, there have been scholarly efforts to identify the underlying causes of OSN addiction to help reduce or control this problematic behavior and its negative consequences. Some of these findings highlight, for instance, the role of motivations and needs, habits, self-regulation, perceived enjoyment, prior use, and/or technology features (Turel & Qahri-Saremi, 2016). Despite their contribution, little effort has been devoted to understanding the personality traits as possible antecedents of OSN addiction. One exception is Vaghefi et al. (2017), which identified (but did not test) situational factors that trigger smartphone addiction in occasions when individuals are bored, anxious, and emotional. Yet, rather than considering stable personality traits, these results point to individuals’ temporary and situational feelings or reactions to their surrounding environments. This is in contrast with the ample evidence in psychology research about the role of personality on general addiction (see Evren, Evren, Yancar, & Erkiran, 2007; Le Bon et al., 2004). To address this gap in our study, we drew on the FFM (Carver & Connor-Smith, 2010) to “theoretically capture[s] the essence of one’s personality” (McElroy et al., 2007, p. 810) and its effect on OSN addiction.
FFM comprises of five personality traits: neuroticism, conscientiousness, agreeableness, openness to experience, and extraversion. Neuroticism is the extent to which individuals experience negative emotions such as stress, anxiety, or depression. Conscientiousness is the ability of individuals to direct and persist at goals and to stop impulses. Agreeableness refers to qualities related to empathy, helpfulness, and cooperativeness. Openness to experience reflects the ability to be creative, unconventional, and artistically sensitive. Finally, extraversion is the degree to which individuals are social, warm, and active (Carver & Connor-Smith, 2010; Srivastava, Chandra, & Shirish, 2015; Terracciano, Löckenhoff, Crum, Bienvenu, & Costa, 2008).
Prior research on FFM suggests that the model can predict which individuals show higher levels of substance addiction, including alcohol, smoking, and drug addiction (Lackner, Unterrainer, & Neubauer, 2013). One meta-analysis showed that high neuroticism and low conscientiousness can be correlated with substance abuse (Kotov, Gamez, Schmidt, & Watson, 2010). Other research also suggested that the FFM model can partly explain excessive and compulsive use of the Internet and other communication tools. For instance, Ehrenberg and colleagues (2008) found that neurotic individuals have high tendencies to use instant messaging (IM) for communication and show addictive smartphone use behaviors. Also, more disagreeable individuals, who also have lower self-esteem, tend to use IM more excessively. In another example, Kuss and Griffiths (2011) found that conscientiousness and agreeableness are associated with problematic OSN use. Nevertheless, the evidence on the role of openness to experience and extraversion toward addiction appears inconclusive (Kayiş et al., 2016; Lapointe & Beaudry, 2014). As a result, although we measured the full FFM model, we focus on neuroticism, conscientiousness, and agreeableness in this article.
Drawing on a review of extant literature in psychology and information systems, we developed five propositions about the plausible effects of personality traits on OSN addiction. Our first proposition (P1) argues that neuroticism has a positive direct effect on OSN addiction. Neurotic individuals lack the necessary skills to manage stressful and difficult situations in life and exhibit poor psychological adjustments in the face of stressful situations (Bolger & Schilling, 1991; McElroy et al., 2007; Suls & Martin, 2005). These individuals are more likely to interpret ordinary situations as threatening and perceive minor frustrations as hopelessly difficult (Bolger & Schilling, 1991; Suls & Martin, 2005). As a result, they seek to avoid or evade stressful situations (Bolger & Schilling, 1991; Suls & Martin, 2005). For neurotic users, OSNs can be a place to immerse themselves in order to evade and relieve the negative feelings frequently experienced in real life (McElroy et al., 2007), particularly because OSNs facilitate online interactions without the stress of face-to-face communications (Amiel & Sargent, 2004).
Our second proposition (P2) argues that conscientiousness has a negative effect on OSN addiction, given its affiliation with personal self-regulatory attributes such as impulse control, self-discipline, and reliability (Carver & Connor-Smith, 2010; McElroy et al., 2007). Conscientious people are generally resistant against urges to engage in and can avoid the excessive use of OSNs and spending too much time on them. There is similar evidence regarding the regulating effect of conscientiousness on text and IM (Ehrenberg et al., 2008).
Our third proposition (P3) contends that agreeableness has direct negative effect on OSN addiction. Low agreeableness is exhibited through use of power and aggression when dealing with a conflict or difficult situation (Moore & McElroy, 2012), which could be associated with drug, alcohol, and cocaine use (Lackner et al., 2013). People with low agreeableness turn to OSNs to establish connections since there is less demand for face-to-face, potentially conflicting interactions, or friendly and cooperative behaviors (Wilson, Fornasier, & White, 2010). OSNs can be suitable environments for people with lower levels of agreeableness since they allow invisible, anonymous, and unstructured interactions (e.g., discovering what others are up to without having to directly talk to them).
In addition to the aforementioned propositions regarding the direct effects of neuroticism, conscientiousness, and agreeableness on OSN addiction, we also contend that these traits have complex interactions with each other that result in strengthening or weakening their direct effects (P4). Specifically, we argue that neuroticism negatively moderates the effect of conscientiousness on OSN addiction—that is, when neuroticism is higher, conscientiousness has a weaker regulating effect on OSN addiction. Neurotic individuals who tend to experience negative feelings may turn to OSNs for relief and overrule the self-regulatory forces that stop such urges. This can happen through the reactive and impulsive systems of the brain (Turel & Qahri-Saremi, 2016, 2018).
Furthermore, we also propose that conscientiousness can control the effect of low agreeableness (P5). High conscientiousness is a quality of people who pay close attention to details. Then, once a problem occurs due to OSN use (e.g., spending too much time on Facebook), people with high conscientiousness will engage in direct efforts to correct that behavior. Even when people have low agreeableness with others, which increases the likelihood of OSN addiction, being highly conscientious will help revert it back to normal use. In summary:
Data was collected in two rounds (t1: week one, t4: week four) from OSN users. The respondents included 275 users (who used mostly Facebook and Instagram), 51 percent women and 49 percent men between the ages of eighteen and thirty-nine (mean age = twenty-one, SD = 2.64). On average, they had more than nine years of experience using OSNs and spent more than ninety minutes per day on it. This sample was selected from North American undergraduate university students, which represents one of the most susceptible groups to developing OSN addiction due to their available free time, flexible schedules, and low parental supervision (Turel & Qahri-Saremi, 2016, 2018; Vaghefi & Qahri-Saremi, 2017).
Measurements were selected from previously well-established scales. At week one, personality traits were measured using Srivastava et al.’s (2015) scale. At week four, OSN addiction was measured using a ten-item scale focusing on addiction symptoms (Turel, 2015). Once data was collected, a series of preliminary tests confirmed the reliability and validity of our measurements and data for analysis (Meyers, Gamst, & Guarino, 2006).
To test our propositions, we used the structural equation modeling (SEM) technique (Kline, 2010). Overall, we found that all propositions together as a model show a relatively good fit to our data (RMSEA = 0.059, with a 95 percent confidence interval of [0.050 – 0.068]; SRMR = 0.055; CFI = 0.91; IFI = 0.91; and TLI = 0.90), and all propositions but P3 were supported. First, neuroticism has a significant positive association with OSN addiction (P1: 0.15, p-value = 0.044). Second, conscientiousness is significantly and negatively associated with OSN addiction (P2: -0.19, p-value = 0.021). Third, our proposition regarding the relationship between agreeableness and OSN addiction was not supported (P3: -0.03, p-value = 0.556). Fourth, our expectation about the moderating effect of neuroticism on the negative effect of conscientiousness on OSN addiction was confirmed (P4: 0.15, p-value = 0.018). Lastly, we surprisingly found that conscientiousness negatively (rather than positively) moderated the relationship between agreeableness and OSN addiction (P5: 0.14, p-value = 0.04).
We further conducted two post hoc analyses to better understand the two moderation effects using an interaction plot tool (Soper, 2011). P4 showed that, as expected, neuroticism negatively moderates the negative effect of conscientiousness on OSN addiction. This means that at higher levels of neuroticism, the controlling effect of conscientiousness on OSN addiction will be less effective. Based on the interaction plot for this proposition, we further found that when the neuroticism level goes one standard deviation above the mean, the negative effect of conscientiousness on OSN addiction becomes nonsignificant. More surprisingly, when the neuroticism level is really high (i.e., at least three standard deviations above the mean) the effect of conscientiousness on OSN addiction turns positive and marginally significant. This suggests that for individuals with a low to medium level of neuroticism, conscientiousness can have a controlling effect on the development of OSN addiction. Yet, when individuals are highly neurotic, high conscientiousness may in turn promote addiction, which was unexpected.
Regarding the second moderation effect (i.e., P5), we unexpectedly found that conscientiousness has a significantly negative moderation effect on the relation between agreeableness and OSN addiction. In this case, the interaction plot revealed that the negative effect of agreeableness on OSN addiction is stronger when individuals’ conscientiousness is low (i.e., two standard deviations below the mean). Put simply, individuals with low conscientiousness and agreeableness have a higher likelihood of developing OSN addiction. Nonetheless, when conscientiousness is relatively high, agreeableness has a marginally significant and positive relation with OSN addiction.
While there are ongoing debates in the academic field whether to consider pathological technology use as an addiction, the public is becoming increasingly aware and wary of the negative consequences of these technologies. As a case in point, there are reports about parental neglect leading to infant death due to excessive technology use (Salmon, 2010). In some countries, such as China and Korea, there are now designated camps to help young people withdraw from Internet and social network use. While these do not apply to most individuals in the population, lower instances of technology addiction are more common and hence worthy of attention from researchers and practitioners.
Our study showed that personality traits may directly and indirectly influence developing and controlling addiction to OSNs. While neuroticism has a positive effect on OSN addiction, we found that conscientiousness holds a negative effect.
There are existing studies looking at the role of personality traits vis-à-vis other technology addictions such as smartphones and the Internet (e.g., Kayiş et al., 2016), yet our study was unique to consider the moderation effects (i.e., complex interaction effects). In addition to its direct negative effect, conscientiousness has an indirect effect on OSN addiction by moderating the relation between agreeableness and OSN addiction. Although, in contrary to our expectation, agreeableness was not significantly associated with IT addiction, our findings indicate that this effect changes based on users’ level of conscientiousness. Additional analyses showed that for individuals with low levels of conscientiousness, higher levels of agreeableness are significantly associated with lower levels of IT addiction. This finding explains that while disagreeable people are more likely to develop addiction to OSNs, their lower levels of conscientiousness can further increase this likelihood. However, this association is faded for the users with higher levels of conscientiousness. Indeed, when conscientiousness is relatively high (i.e., at least two standard deviations above the mean), agreeableness has a marginally significant and positive relation with OSN addiction, meaning that agreeableness can, at least marginally, contribute to IT addiction.
To explain the rather unexpected findings, we use the notion of “rational addiction,” denoting that some addicts “anticipate the future consequences of their current behaviors and attempt to maximize utility from their intertemporal consumption choices” (Kwon, So, Han, & Oh, 2016, p. 924). Regarding OSN addiction, this implies that individuals who have high agreeableness levels are likely to value relationships with friends on OSNs and would like to show warmness to others via interaction with friends such as leaving likes, comments, and messages. Having high conscientiousness may mean that such individuals pursue these empathetic goals deliberately, rather than aimlessly. Yet, excessive interaction with these technologies over time develops an OSN addiction with adverse consequences such as lack of time for other productive activities like schoolwork, for example. This is in contrast with the common characterization of addiction as an impulsive, irrational, and unregulated behavior. Given the many positive benefits of technology use, these cautionary views should be considered when examining technology related addictions. For example, many people need their smartphones for their daily activities, yet some may develop a fear of losing their phone or leaving it at home, even for a single day—this is called “nomophobia.” This can cause serious psychological distress, and at some point, can become counterproductive.
Additionally, we hope our findings inform practitioners about the role of personality traits, reminding them that, similar to substance addictions, there are meaningful differences among individuals that contributes to addictive tendencies of using OSNs or other technologies. We would like to emphasize that it is important to look at the profile of personality traits together, rather than cherry picking and testing them separately. For instance, there are recent studies that show the positive impact of narcissism on OSN addiction, saying that narcissistic individuals are more active on Facebook (or other platforms) since it empowers them to create an image of their ideal self through interaction with others (Kuss & Griffiths, 2011). However, it is likely that individuals with high conscientiousness levels can control their behaviors and reduce their risk of developing OSN addiction.
Our findings can inform the way intervention, treatment, and recovery programs can be designed. For instance, for highly neurotic people a treatment that reduces their stress levels—such as psychotherapy or relaxation exercises—could be effective in reducing technology addiction. For highly conscientiousness people, on the other hand, highlighting the adverse outcomes of use might be sufficient to trigger control and reducing OSN use. For instance, there are now applications that track the amount of use in terms of frequency and time spent on OSN applications (e.g., the “Moment” application: https://inthemoment.io) that could raise awareness about use and activate self-regulatory actions by individuals. Via this study, we invite further discussions on the role of personality traits regarding the development of new technological innovations (e.g., OSNs and smartphones), which could impose potentially detrimental effects on individuals, organizations, and society.
Isaac Vaghefi, PhD, is an assistant professor of information systems at Seidenberg School of Computer Science and information systems at Pace University in New York. His research interests pertain to the dark side of technology use and health care IT. In particular, he has done studies on the addictive use of IT and the use of electronic systems in health care organizations. His work has been published in academic journals such as the Information Systems Journal, the Journal of the American Medical Informatics Association, and the Journal of Medical Internet Research as well as appeared in popular media outlets including The Washington Post and The Daily Mail.
Hamed Qahri-Saremi, PhD, is an assistant professor of information systems at the College of Computing and Digital Media at DePaul University in Chicago, Illinois. His research is at the intersection of social and technological systems and particularly addresses the behavioral implications of using information systems and computer-mediated communications. His research papers have appeared in various outlets such as the Journal of Management Information Systems, the Journal of Strategic Information Systems, Information & Management, New Media & Society, and European Financial Review, among others.