Predicting SNS Addiction with the Big Five and the Dark Triad

Introduction

Social network sites (SNS) are web-based applications that enable users to construct personal profiles with the purpose of regulating and maintaining contacts online (Boyd & Ellison, 2007). One of the common issues of SNS is described as the failure to regulate SNS usages that lead to negative outcomes (Andreassen & Pallesen, 2014; LaRose, Kim, & Peng, 2010), commonly known as SNS addiction. Existing research revealed the detrimental effects of SNS addiction on several domains of functioning that include well-being (Koc & Gulyagci, 2013), interpersonal relationships (Elphinston & Noller, 2011) as well as self-esteem and life satisfaction (Błachnio, Przepiorka, & Pantic, 2016). 

According to Andreassen and Pallesen (2014), personality is an important factor in the development of SNS addiction. In the literature, research that intends to identify personality traits that are indicative of the risk of SNS addiction often operationalized personality with the Five-Factor model. The Five-Factor model consists of five broad domains known as Extraversion that refers to the tendency to seek for companionship from others, Neuroticism that refers to the tendency to experience negative emotions, Conscientiousness that refers to the tendency to be organized, Openness to Experience that refers to the tendency to expose oneself to new stimulus or experience, and Agreeableness that refers to the tendency to be cooperative with others (Benet-Martinez & John, 1998; John et al., 1991). According to existing research, there are several indicators of the risk of SNS addiction, namely high Extraversion (Andreassen, Torsheim, Brunborg, & Pallesen, 2012; Wang, Ho, Chan, & Tse, 2015; Wilson et al., 2010), high Neuroticism (Caci, Cardaci, Scrima, & Tabacchi, 2017; Wang et al., 2015), low Conscientiousness (Andreassen et al., 2013; Błachnio & Przepiorka, 2016a; Wilson et al., 2010), and low Openness to Experience (Andreassen et al., 2013; Błachnio & Przepiorka, 2016a). However, these traits accounted for a small amount of variance even though the findings supported the significance of these traits in predicting SNS addiction. For instance, Wilson et al. (2010) found that the broad five domains significantly predicted addictive tendencies toward SNS with 8.5% of the variance while Błachnio and Przepiorka (2016a) found that these domains accounted for 6% of the variance. Hence, this illustrates the limited contribution of these traits in predicting SNS related behaviours (Amichai-Hamburger & Vinitzky, 2010; Ross et al., 2009). Alternatively, the small amount of variance accounted by the Five-Factor model traits implies that these traits have insufficient specificity to predict specific behaviours (Hough, 1992; McAdams, 1992). As a result, it is necessary for researchers to consider other theories of personality to predict the usage of SNS (Ross et al., 2009; Skues, Williams, & Wise, 2012).

It was demonstrated that subclinical personality traits can be indicative of behavioural addiction. Research suggests that users that are addicted to the internet are more likely to exhibit the characteristics of borderline personality disorder (Dalbudak, Evren, Aldemir, & Evren, 2014). Similarly, users with high degree of callousness or lack of empathy are prone to internet addiction (Trumello, Babore, Candelori, Morelli, & Bianchi, 2018). In relation to SNS usages, the facilitating role of narcissism has been supported by research findings (e.g. Buffardi & Campbell, 2008). The inclusion of narcissism contributed significantly to the predictive model of SNS usages after accounting for Extraversion (Ong et al., 2011), indicating that the measured Extraversion did not account entirely for the measured usages of SNS. In addition, recent findings extend that individuals with elevated self-centeredness are prone to SNS addiction due to the urge to gratify their needs for affiliation and self-assurance (Andreassen, Pallesen, & Griffiths, 2017; Casale, Fioravanti, & Rugai, 2016). However, the effect size of .06 indicates that narcissism is weakly correlated with SNS addiction (Andreassen et al., 2017). Additionally, the small effect size accounted by narcissism after controlling for Extraversion (Ong et al., 2011) suggests the need to further explore the role of other related traits in predicting SNS usages. In this case, the present research aims to explore the relationship of SNS addiction with the three aversive traits of the Dark Triad, namely narcissism, psychopathy, and Machiavellianism (Paulhus & Williams, 2002). Narcissism is marked by a sense of entitlement, dominance and a grandiose self-view (Raskin & Terry, 1988). Machiavellianism refers to the tendency to manipulate others for own gain (Christie & Geis, 1970). Psychopathy consists of the lack of empathy, callousness, and erratic behaviours (Hare, 2003). These three aversive traits form the Dark Triad that has been linked to several negative outcomes such as interpersonal difficulties (Stead, Fekken, Kay, & McDermott, 2012), impulsivity (Crysel et al., 2013; Jones & Paulhus, 2011a; Jonason & Tost, 2010; Malesza & Ostaszewski, 2016), clinical psychopathy (Jones & Figueredo, 2013) and sadism (Chabrol et al., 2009). Apart from that, recent research on the Dark Triad has associated these traits with online behaviours such as cyber-aggression (Pabian, De Backer, & Vandebosch, 2015), problematic online gaming (Kircaburun, Jonason, & Griffiths, 2018a), cyberstalking (Kircaburun, Jonason, & Griffiths, 2018b), cyberbullying (Goodboy & Martin, 2015; Kircaburun et al., 2018b), and online deception or cybertrolling (Buckels, Trapnell, & Paulhus, 2014; Kircaburun et al., 2018b).  

Research extends that certain features on SNS, such as status updates (Garcia & Sikstrom, 2014) and selfies (Fox & Rooney, 2015; McCain et al., 2016), appeal to users with salient features of the Dark Triad. Hence, they tend to exploit these features, which could be a possible cause to their frequent (Fox & Rooney, 2015) and intense use of SNS (Pabian et al., 2015). It was indicated that frequent use of SNS may escalate into addiction (Hong et al., 2014) when users are unable to restraint the use of SNS (Błachnio & Przepiorka, 2016b). Recent findings support the link of the Dark Triad and problematic use of SNS (Kircaburun, Demetrovics, & Tosuntaş, 2018; Kircaburun et al., 2018b), suggesting that the escalation of prolonged (Fox & Rooney, 2015) and intense use of SNS (Pabian et al., 2015) into SNS addiction is possible through poor self-control and high degree of impulsivity that are evident amongst those with salient features of the Dark Triad (Crysel et al., 2013; Jones & Paulhus, 2011a; Jonason & Tost, 2010; Malesza & Ostaszewski, 2016). Each of the Dark Triad traits significantly correlated with problematic use of SNS, with psychopathy formed the weakest relationship (Kircaburun, Demetrovics, & Tosuntas, 2018; Kircaburun et al., 2018b). However, psychopathy has been consistently linked to poor self-control and impulsivity (Jones & Paulhus, 2011a; Jonason & Tost, 2010) that are central to the development of SNS addiction (Błachnio & Przepiorka, 2016b). Reflecting from this, the present research anticipates the significance of psychopathy in predicting SNS addiction.

In addition, the purpose of adopting the Dark Triad traits is to address the limited influence of the Five-Factor model in predicting SNS addiction. It was previously indicated that the Five-Factor model does not reflect adequately on human personality (Veselka, Schermer, & Vernon, 2011; Veselka, Schermer, & Vernon, 2012). Consistent with this argument, a research found that the employment of the Dark Triad traits accounted for additional variance in predicting social values after controlling the Five-Factor model (Kajonius, Persson, & Jonason, 2015), which suggests that the influence exerted by the Dark Triad traits are independent of the Five-Factor model (Carter, Campbell, & Muncer, 2014). Therefore, investigation performed using these traits may provide novel insights to the present understanding on SNS addiction.  

One of the objectives of the present research is to examine the significance of the Dark Triad traits in predicting SNS addiction. The present research hypothesizes the significance of psychopathy in predicting SNS addiction considering that psychopathy is associated with poor self-control and impulsivity (Jones & Paulhus, 2011a; Jonason & Tost, 2010). The next objective of this research is to compare the variances accounted by the Dark Triad and the Five-Factor model in the prediction of SNS addiction. Hence, the present research hypothesizes that the Dark Triad traits will account for additional variance in predicting SNS addiction after controlling the Five-Factor model. 

Method

Participants

The sample of the present research consisted of 204 undergraduate students at Malaysia. They were all Malaysians, aged between 18 and 27 (M_age = 22.94, SD_age = 3.43). In the demographic aspect, majority of the participants were female (Male = 77, 38%; Female = 124, 60%; Not mentioned = 3, 2%), and majority of participants were identified as Chinese  (Malay = 79, 39%; Chinese = 99, 49%; Indian = 8; 4%; Others = 18, 8%).  

Procedure

The present research was advertised at a university’s lecture halls. Those who were interested contacted the researcher for further arrangement to complete of the survey questionnaire. They were invited to complete the survey questionnaire in a classroom specifically booked for data collection. Informed consents from the participants were not obtained to ensure anonymity, and no monetary reward was offered to the participants. In this research, data analysis was conducted using SPSS 19. Ethical clearance was granted by the Monash University Human Research Ethics Committee.

Measures

The demographic details of the participants that include age, gender, and ethnicity were obtained at the beginning of the questionnaire. The traits of the Five-Factor model were measured using the Big Five Inventory (BFI; Benet-Martinez & John, 1998; John et al., 1991). It consists of 44 items that reflects on five dimensions, namely Openness to Experience (e.g., “I am someone who is curious about many different things”), Conscientiousness (e.g., “I am someone who is dependable, steady”), Extraversion (e.g., “I am someone who is outgoing, sociable”), Agreeableness (e.g., “I am someone who is compassionate, has a soft heart”), and Neuroticism (e.g., “I am someone who is moody, has up and down mood swings”). In the present research, these items were rated on a five-point Likert scale (1 = Very inaccurate, 5 = Very accurate). Each factor was formed by summing the corresponding items. The measured five factors exhibited acceptable to high internal consistencies (Extraversion’s α = .71, Agreeableness’ α = .72, Conscientiousness’ α = .72, Neuroticism’s α = .79, Openness to Experience’s α = .67).

In the present research, the Dark Triad traits were measured using the Short Dark Triad (SD3; Jones & Paulhus, 2014). It consists of 27 items that reflect on the three aversive traits, namely psychopathy (e.g., “I like to get revenge on authorities”), Machiavellianism (e.g., “It’s not wise to tell your secrets”), and narcissism (e.g., “People see me as a natural leader”). These items were rated on a five-point Likert scale (1 = Strongly disagree, 5 = Strongly agree). Each subscale was formed by summing the corresponding items. The measured traits exhibited acceptable to high internal consistencies (psychopathy’s α = .75; Machiavellianism’s α = .65; narcissism’s α = .66). 

SNS addiction was measured using the Bergen Facebook Addiction Scale (Andreassen et al., 2012). In the present research, the items were rephrased to ensure that they are suitable for the SNS contexts. This measure consists of six items that were rated on a five-point Likert scale (1 = Very rarely, 5 = Very often). A sample item of this measure is “How often during the last year have you spent more time on SNS than initially intended?”. The items were summed to form the score for SNS addiction. This measure exhibited high internal consistency (SNS addiction’s α = .85). 

Results

Table 1 summarizes the descriptive statistics and the correlation coefficients. The results indicated that the correlations ranged from small to moderate, indicating that multicollinearity was not a concern. The VIF values ranged from 1 – 2.15, with the highest value exhibited by the measured psychopathy. The maximum VIF value of 2.15 does not exceed 10, which further support that multicollinearity was not an issue (Kennedy, 1992). From Table 1, SNS addiction was significantly correlated with psychopathy. The positive correlation suggests that higher degree of psychopathy increases the risk of SNS addiction. In addition, Neuroticism was positively correlated with SNS addiction, which indicates that individuals with poor emotional stability are susceptible to SNS addiction. Agreeableness, Conscientiousness, and Openness to Experience were negatively correlated with SNS addiction. 

Table 1. Descriptive Statistics and Correlation Coefficients.

Hierarchical regression with three blocks was conducted. The results of this regression are summarized in Table 2. The first block consisted of the age and gender of the participants, followed by the Five-Factor model, and finally, the Dark Triad. From Table 2, the initial block accounted for 5% (R²) of the variance. The entry of the Five-Factor model contributed 8% (R²) of the variance, and the entry of the Dark Triad contributed an additional 3% (R²) of the variance. The inclusion of the Dark Triad traits significantly contributed to the predictive model, suggesting that the Five-Factor model does not exhaust the variance in the prediction of SNS addiction. SNS addiction was significantly predicted by psychopathy, suggesting that greater degree of psychopathy increases vulnerability towards SNS addiction. Additionally, the Openness to Experience negatively predicted SNS addiction, which implies that individuals with high openness are less likely to develop SNS addiction. Overall, the predictors involved in this study accounted for 16% (R²) of the variance.

Table 2. Hierarchical Regression Predicting SNS Addiction with Demographics and Personality Factors.

Discussion

The present research intended to examine the significance and the contribution of the Dark Triad in predicting SNS addiction. The hypothesized significance of the measured psychopathy as a predictor of SNS addiction was supported. The secondary psychopathy measured in the present research (see Jones & Paulhus, 2014) has been linked to risky decision-making (e.g. Dean et al., 2013), poor inhibition and high impulsivity (Jonason & Tost, 2010; Jones & Paulhus, 2011a; Jones & Paulhus, 2011b; Malesza & Ostaszewski, 2016) that will induce excessive use of SNS or SNS addiction (Błachnio & Przepiorka, 2016b). Therefore, the findings of the present research suggested that high degree of secondary psychopathy elevate the risk of SNS addiction. However, the findings of the present research indicated that the traits of narcissism and Machiavellianism were not significant predictors of SNS addiction even though previous research found that frequent (Fox & Rooney, 2015) and intense use of SNS (Pabian et al., 2015) were significantly correlated to these traits. Hence, this highlighted the importance of psychopathy as the only significant predictor of SNS addiction due to its conceptual link with impulsivity (Jones & Paulhus, 2011b; Malesza & Ostaszewski, 2016). The significance of psychopathy contradicted previous findings on the non-significant relationship of psychopathy and problematic use of SNS (Kircaburun, et al., 2018). The inconsistency can be explained by the measures used. The Dirty Dozen (Jonason & Webster, 2010) used by previous researchers to measure the traits of the Dark Triad (Kircaburun et al., 2018) has been criticized for its brevity that leads to omission of important content (Miller et al., 2012; see also Jones & Paulhus, 2014). Research implicates for caution when utilizing the Dirty Dozen’s (Jonason & Webster, 2010) psychopathy subscale since it might attenuate correlations with psychopathic related constructs (Miller et al., 2012). Therefore, it is likely that the relationship of psychopathy and problematic SNS usage found previously (Kircaburun et al., 2018; Kircaburun et al., 2018b) was attenuated by the measure used. From a direct comparison, the SD3 (Jones & Paulhus, 2014) used in the present research demonstrated better validity than the Dirty Dozen (Maples, Lamkin, & Miller, 2014). This suggests that future research should consider the psychometrics of the measures used to assess the Dark Triad traits. Narcissism is often linked to self-promotional features on SNS (Fox & Rooney, 2015; Garcia & Sikstrom, 2014) that are vaguely represented by SNS addiction (Andreassen & Pallesen, 2014; LaRose et al., 2010). Hence, the link between narcissism and SNS addiction has been portrayed as weak despite of its significance (Andreassen et al., 2017). The present research found the correlation of .07 that was declared as not significant for narcissism and SNS addiction, which is slightly larger than the effect size of .06 found by Andreassen et al. (2017). Hence, the large sample size involved in the respective research may have inflated the link of SNS addiction and narcissism (Andreassen et al., 2017). Alternatively, the weak correlation of narcissism and SNS addiction can be accounted by the facets of narcissism. The narcissism subscale of the SD3 (Jones & Paulhus, 2014) used in the present research reflects on the grandiose sense of self (Maples et al., 2014), which is similar to the Narcissistic Personality Inventory-16 (NPI-16; Ames, Rose, & Anderson, 2006) used in Andreassen et al.’s (2017) research. Therefore, the link of grandiose narcissism and SNS addiction has been portrayed as weak. The significant correlation found previously (Kircaburun et al., 2018; Kircaburun et al., 2018b) was derived from the narcissism subscale of the Dirty Dozen (Jonason & Webster, 2010) that reflects on both grandiose and vulnerable narcissism (Maples et al., 2014). Hence, it is likely that problematic SNS usage such as SNS addiction relates better to the vulnerable aspect of narcissism. Individuals with high Machiavellianism trait may prefer other forms of social network or interaction to achieve their goals (Fox & Rooney, 2015). Moreover, individuals with salient features of Machiavellianism possess greater self-monitoring as an effort to disguise their manipulative tendencies (Abell & Brewer, 2014). In this light, these individuals are less likely to be impulsive (Jones & Paulhus, 2011a). With the absence of poor self-control that is central to the development of SNS addiction (Błachnio & Przepiorka, 2016b), the trait of Machiavellianism did not correlate significantly with SNS addiction.  

The present research found that high level of openness predicted lesser risk of SNS addiction, which is akin to previous findings (e.g. Andreassen et al., 2013; Błachnio & Przepiorka, 2016a). It was also revealed that the trait of openness to new experience is protective against negative outcomes. Specifically, high degree of openness is known to reduce the risk of mortality (Ferguson & Bibby, 2012; Jonassaint et al., 2007; Turiano, Sprio, & Mroczek, 2012) as well as the likelihood to develop physical ailment (Israel et al., 2014). Similarly, individuals with high degree of openness tend to possess higher stress tolerance (Oswald et al., 2006; Williams, Rau, Cribbet, & Gunn, 2009). In light of these findings, the measured trait of openness may be protective against SNS addiction. Of the five broad domains, low level of Agreeableness indicates for saliency of the Dark Triad traits (Muris, Merckelbach, Otgaar, & Meijer, 2017; O'Boyle, Forsyth, Banks, Story, & White, 2015). The present research replicated similar findings, except for narcissism that was not significantly correlated with Agreeableness. This can be accounted by the weak link of Agreeableness and narcissism (Muris et al., 2017; O’Boyle et al., 2015). Additionally, the measure utilized to assess the dark traits might further weakened the relationship between Agreeableness and narcissism. It was indicated that the effect sizes that reflect on the strength of the relationships of the Five-Factor model and the Dark Triad traits were larger when the dark traits were measured by the original measures (e.g. Mach-IV, Christie & Geis, 1970; NPI-16, Ames et al., 2006; Psychopathy Checklist-Revised, Hare, 2003; see Muris et al., 2017). Therefore, the SD3 (Jones & Paulhus, 2014) utilized in the present research may have attenuated the correlation coefficients, and thus, further reduce the weak link of narcissism and Agreeableness.            

The present results supported that the inclusion of the Dark Triad traits contributed additional variance in the prediction of SNS addiction after controlling the Five-Factor model. This suggests that the Five-Factor model did not exhaust the variance in the measured SNS addiction entirely, leaving some additional variance to the Dark Triad. The present results are consistent with previous research that supported the significance of the Dark Triad after controlling the broad five domains in predicting perceived male attractiveness (Carter et al., 2014) and social values (Kajonius et al., 2015). This is an indication that the Five-Factor model does not reflect entirely on human personality, and the inclusion of the Dark Triad can extend the inadequacy (Veselka et al., 2011; Veselka et al., 2012). Therefore, the effects exerted by the Dark Triad traits on SNS addiction are independent of the Five-Factor model. The present findings also supported that the broad five domains of the Five-Factor model are lacking the required specificity to predict specific behaviours (Hough, 1992; McAdams, 1992) such as SNS addiction.

The findings obtained in the present research further expand the existing literature by focusing on the conceptual links of the traits measured. In relation to this, the present research revealed that other traits such as psychopathy of the Dark Triad are also influential even though some of the traits covered by the Five-Factor model have been significantly associated with SNS addiction. The measured psychopathy was found as the most influential predictor of SNS addiction due to its conceptual link with impulsivity. Additionally, the importance of the conceptual links of the measured traits and SNS addiction is further illuminated by the illustrated relationships. Specifically, some traits may increase vulnerability towards SNS addiction such as psychopathy that predicted SNS addiction positively, whereas some traits can be protective against SNS addiction as illustrated by the Openness to Experience that predicted SNS addiction negatively. In this light, it is recommended for future research to consider the conceptual link and to be more selective with the traits that will be used to predict SNS addiction. Moreover, the present results have supported that the Five-Factor model has limited influence in predicting SNS addiction (Amichai-Hamburger & Vinitzky, 2010; Ross et al., 2009). Apart from that, the small effect sizes accounted by the Five-Factor model supports the recommendation to explore the significance of other personality traits (Ross et al., 2009; Skues et al., 2012). Therefore, the present findings suggest that further exploration with the Five-Factor model will only fixate the literature of SNS addiction. 

One of the limitations of the present research is its inability to infer for causality due to the correlational design employed. Moreover, it is important to note that the generalizability of the results is limited due to the diverse and unequal demographics as well as the small number of student sample. In addition, the small effect sizes from the hierarchical regression show the need to expand the investigation on the conceptual link of SNS addiction with other personality traits. The use of concise measures of the Dark Triad, such as the SD3 (Jones & Paulhus, 2014), affect the results by attenuating the effect sizes. Hence, future research should measure the traits of the Dark Triad with the original measures. Due to the conciseness of the SD3 (Jones & Paulhus, 2014), the present research could only measure a fragment of the Dark Triad traits while these traits may be multifaceted (see Furnham, Richards, & Paulhus, 2013). Thus, it is also imperative for future research to assess the role of different facets of the Dark Triad traits (e.g. primary psychopathy, secondary psychopathy) in shaping SNS addiction.  

In summary, the present research concludes that the measured psychopathy of the Dark Triad is the most influential predictor of SNS addiction. It is also concluded that the Dark Triad traits accounted for more specificity compared to the Five-Factor model in the prediction of SNS addiction. Therefore, this further supports the recommendation of exploring traits other than the Five-Factor model.