ABSTRACT

Considering negative psychological and other consequences of trolling on social media, the present study addressed the link between narcissism and social media trolling and its potential mediators, including malicious and benign envy and exposure to antisocial media content. Participants were recruited via an online crowdsourcing platform, Prolific, and 326 (aged 39.4 [SD = 12.4]; 48% male, 50% female, 2% non-binary or non-disclosed gender) completed our online questionnaire without attention-check failures. Our mediation analyses (controlling for covariates) indicated that malicious envy mediated the links of narcissism to social media trolling and to exposure to antisocial media content. In addition, exposure to antisocial media content also served as an additional mediator, as narcissism was linked to social media trolling indirectly through malicious envy and then exposure to antisocial media content. Narcissism predicted social media trolling directly, even after additionally controlling for the two mediators that were also directly linked to the trolling outcome. These findings provide practical implications, highlighting the importance of addressing the identified antecedents of social media trolling, and indicate future research directions to investigate additional associated factors and explore and develop approaches and interventions to reduce trolling on social media.

Consent to participate

Informed consent was obtained from all individual participants included in the study. They consented at the beginning of our online questionnaire before answering questions for this study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Ethics approval

The institutional review board at Marshall University reviewed the study protocol and granted Exempted approval with Marshall University IRB# 2113720.

Data availability

The data used in the present study will be available upon request.

Notes

1 The narcissism construct is so nuanced that existing multidimensional theories or models may not fully capture it. For example, when proposing their narcissistic admiration and rivalry concept, Back et al. (Citation2013) emphasized their focus on grandiose or ‘normal’ narcissism (observed in the non-clinical population) without addressing vulnerable or pathological narcissism. The present study did not aim to capture the complexity of narcissism or the full range of its consequences, which we acknowledge could be an important area of research.

2 The minimum sample size was determined based on our prior research. More specifically, assuming a standardized effect size of .20 (estimated based on our previous analyses on relevant psychological factors) for each association among our main variables (i.e. narcissism, malicious and benign envy, social media trolling, and antisocial media content), we referred to results from the ‘Monte Carlo Power Analysis for Indirect Effects’ developed by Schoemann, Boulton, and Short (https://schoemanna.shinyapps.io/mc_power_med/). The analyses suggested that the minimum sample size required to achieve our target power of .80 would be 270 for the Two Parallel Mediators models (for the indirect effect through each mediator as depicted in and ) and 310 for the Two Serial Mediators model (for the indirect effect through two mediators as depicted in ). Our sample size of 326 exceeded the minimum. Based on the obtained effect estimates with this sample size, the actual power for the significant associations (as shown in ) was .86 and .96 for the indirect effects of narcissism on social media trolling and antisocial media content, respectively, through malicious envy, and .84 for the indirect effect on social media trolling through malicious envy and antisocial media content. More detailed results of our power analyses will be available upon request.