Abstract :
[en] Just like many school aggression researchers (see the review of studies in Olivier et al., 2021; Smith & Steffgen, 2013), many scholars investigating workplace bullying assume the existence of four classes of employees: Uninvolved employees, victims, perpetrator, and provocative victims (also called ‘bully-victims’; e.g., Zapf & Einarsen, 2020). Yet, empirical evidence supporting these four classes is lacking so far. Instead, studies on workplace bullying mostly focused solely on the victim perspective (Einarsen et al., 2020; see also the review of studies in Gupta et al., 2020), while far less studies investigate the perpetrator perspective (see the enumerated studies in Nielsen & Einarsen, 2018). Only some studies include both perspectives (e.g., Baillien et al., 2016; Fernández-del-Río et al., 2021; Sischka et al., 2021). However, these studies applied a variable-centered approach and treated workplace bullying exposure and perpetration as separate variables. Thus, they tell us little about the interplay of workplace bullying exposure and perpetration. The current study employs a person-centered approach (i.e., latent class analysis) to identify groups of employees that show different patterns of workplace bullying involvement. Moreover, the current study aims to identify predictors of latent class membership.
We investigate the individual disposition hypothesis (Nielsen & Einarsen, 2018), assuming that individual characteristics such as personality traits may be related to being involved in bullying and could therefore be typifying features of workplace bullying groups. Moreover, according to some researchers in workplace bullying, individual characteristics such as certain personality traits (big five dimensions, trait aggression, dominance) are particularly appropriate to identify and distinguish workplace bullying subgroups (Matthiesen & Einarsen, 2007; see also Nielsen et al., 2017). On the other hand, power and social status represent enabling structures (Salin, 2003) of workplace bullying, and power imbalance is often seen as a defining feature of workplace bullying (Einarsen et al., 2020).
To test our hypotheses, we recruited participants via Amazon’s Mechanical Turk (MTurk; Buhrmester et al., 2011; Crump et al., 2013) to participate in an online survey. The final sample contained 1,492 respondents (53.8% females, n = 802), with ages ranging from 19 to 77 (M = 40.2, SD = 10.8). Most respondents worked full time (91.2%, n = 1,360). The survey contained the Short-Negative Acts Questionnaire (S-NAQ; Notelaers et al., 2019) that was applied from the victim as well as from the perpetrator perspective. Furthermore, we assessed the big five with the Mini-IPIP (Donnellan et al., 2006), trait aggression with the brief aggression questionnaire (Webster et al., 2014), and dominance with the domineering subscale of the Inventory of Interpersonal Problems (Soldz et al., 1995). Moreover, power and social status were assessed with a newly developed bipartite measure of social hierarchy (Yu et al, 2019). Latent class analysis was employed. Statistical fit indices, substantive interpretability and utility, as well as classification diagnostics were jointly considered to determine the number of workplace bullying classes. To explore the relationships between the latent categorical variable and the other variables (i.e., personality traits, power and social status), we followed the three-step procedure (e.g., Asparouhov & Muthén, 2014).
Latent class analysis revealed four classes, i.e., (1) mild bullying involvement, (2) bully-victims (3) non-involved and (4) victims. These classes were meaningfully linked with the big five dimensions, trait aggression, dominance, power and social status. Specifically, bully-victims scored high on extraversion, neuroticism, trait aggression, dominance, power and low on status, agreeableness, conscientiousness and openness. Victims on the other hand scored also high on neuroticism and – compared with non-involved – lower on agreeableness, conscientiousness and openness and higher on trait aggression and dominance. Victims showed the lowest levels of extraversion and status.
The present study implements a cross-sectional design precluding a causal interpretation. Moreover, only self-reported measures were employed. Thus, especially indicators of workplace bullying perpetration might be influence by social desirability.
Scholars and policy makers need to be aware of different patterns of being involved in workplace bullying in order to create effective interventions. Personality factors as well as power and social status are important correlates of workplace bullying involvement.
The current study is the first that investigates workplace bullying from a target and a perpetrator perspective by employing a person-centered approach (i.e., latent class analysis).
