Relationship between Honesty-Humility and Cyberbullying Perpetration: A Moderated Mediation Model

HH and CP

HH is defined as “the tendency to be fair and genuine in dealing with others, in the sense of cooperating with others, even when one might exploit them without suffering retaliation” (Ashton & Lee, 2007, p. 156). HH is particularly crucial for predicting social behavior (Ashton & Lee, 2008; Hilbig et al., 2014). Specifically, it includes four facets, namely sincerity, fairness, greed avoidance, and modesty (De Vries, 2013). Moreover, HH has been considered to represent an absence of dark traits that can increase the risk of CP (Hodson et al., 2018). Accordingly, some researchers consider HH as a crucial factor of antisocial behavior (e. g., Allgaier et al., 2015; Thielmann & Hilbig, 2018). That is, individuals low in HH are more likely to conduct deviant behaviors to achieve their aims. Moreover, many studies have found the significant associations of HEXACO personality traits with traditional bullying (Farrell et al., 2014) and internet deviant behaviors (e.g., such as sexting) (Morelli et al., 2020). More importantly, HH has been verified as a primary predictor of bullying perpetration over and above all the other HEXACO factors (Book et al., 2012; Volk et al., 2018), and bullying perpetration is the strongest predictor of CP (Antoniadou et al., 2016). Therefore, HH may be negatively associated with CP.

Materialism as a Mediator

Materialism refers to “individual differences in people’s long-term endorsement of values, goals, and associated beliefs that centre on the importance of acquiring money and possessions that convey status” (Dittmar et al., 2014, p. 880). According to McCrae and Costa’s (2008) model, personality relates to basic regulatory mechanisms underlying individual values (Górnik-Durose & Pilch, 2016). More importantly, empirical evidence has indicated that personalities create a foundation for materialistic value (e.g., Hong et al., 2012; Watson, 2014). In relation to materialism, HEXACO has stronger predictive validity than the Big Five model, mainly due to HH, which is more strongly correlated with materialism than other personality traits (Górnik-Durose & Pilch, 2016). That is, HH can be a crucial and negative predictor of materialism.

The relationship between materialism and CP has been theoretically and empirically supported. Based on values conflict theory (Burroughs & Rindfleisch, 2002), materialistic values conflict with self-transcendence values (e.g., prosocial values) that are negatively associated with CP. Accordingly, materialistic adolescents pursue sensual pleasure and acquire possessions, rather than consider the welfare of others beyond their own interests (Kasser, 2016). Thus, materialistic adolescents may be more likely to conduct antisocial behaviors (e.g., cyberbullying). Previous empirical studies have supported this view. For example, materialism is related to poor interpersonal relationship, anxiety, low life satisfaction, and even depression (Dittmar et al., 2014), which are risk factors of CP (Kowalski et al., 2014). More importantly, materialism has been directly verified to be positively associated with CP from an evolutionary perspective (Wang, Wang, et al., 2021).

Based on the relationship between HH and materialism and the relationship between materialism and CP, we concentrated on the mediating effect of materialism between HH and CP. Some researchers have focused on the explanation mechanism of materialism in the effects of personality traits on social behaviors, rather than the simple relationships between personality traits and materialism (Kasser, 2016). Previous evidence has supported this view. For example, some researchers view materialism as a mediator in the models that explain the associations of personalities with particular types of social behavior, such as excessive buying (Otero-López & Villardefrancos, 2013) and compulsive buying (Harnish & Bridges, 2014). Accordingly, materialism may mediate the relationship between HH and adolescents’ CP.

PPC as a Moderator

PPC involves “attempts to control the child through psychological tactics that encourage dependency, inhibit individuation, and invalidate the child’s sense of self” (Scharf & Goldner, 2018, p. 358). The SECF-C considers CP as an outcome of the interactions of individual personalities with environmental factors (especially parental factors) (Cross et al., 2015). Accordingly, PPC may interact with HH on CP. More importantly, PPC frustrates children’s basic needs (Soenens & Vansteenkiste, 2010), which further induces children’s aggressive behaviors, including CP (He et al., 2019). According to the Situation-Trait-Outcome-Activation (STOA) Model, the situation activation mechanism of personality suggests that situation activation factors would increase the behavior difference of individuals with different levels of HH (De Vries et al., 2016). In another word, behavior differences of individuals with different levels of HH are most likely to be activated in some specific situations (Barends et al., 2019). Therefore, we expected that PPC would moderate the relationship between HH and CP.

Some studies have directly supported the interaction of HH with a situational factor on CP. For instance, situational characteristics have greater influences on antisocial behaviors among students low (versus high) in HH (Allgaier et al., 2015). More importantly, poor parenting behaviors, such as poorer mother knowledge, increase the adverse effect of low HH on bullying perpetration (Farrell et al., 2017). Specifically, adolescents low in HH show more bullying perpetration in families with poorer mother knowledge than in families with high mother knowledge; however, those high in HH show less bullying perpetration in general. Accordingly, stronger PPC creates an environment inducing a sense of insecurity (Soenens & Vansteenkiste, 2010), which increases CP among adolescents low in HH than those high in HH. That is, stronger PPC would exacerbate the adverse effect of low HH on CP.

According to the STOA Model, some stressful situations can activate individual differences in HH (Barends et al., 2019; De Vries et al., 2016). In family environments with stronger PPC, children’s relatedness/closeness and secure needs are frustrated (Soenens & Vansteenkiste, 2010), thereby children would pursue materialistic goals to obtain approval and closeness (Ching & Wu, 2018). However, people tend to use different self-regulatory strategies to attain desired states and avoid undesired states (Ozimek et al., 2017). Moreover, some researchers have concentrated on the interaction of stressful environmental factors (e.g., student-student relationships) with personalities (e.g., narcissism) in shaping materialistic values (Ouyang et al., 2020). Based on the description of HH, adolescents high in HH place less value on power, money, and wealth (Lee et al., 2013), and would adopt other effective coping attempts to cope with stress and insecurity (Chirumbolo, 2014). By contrast, adolescents low in HH are described as boastful, deceitful, or greedy (Ashton & Lee, 2008), and adopt self-serving coping attempts to cope with insecurity (Chirumbolo, 2014). Accordingly, compared to adolescents high in HH, those low in HH are more likely to pursue materialistic goals and further conduct more CP to cope with insecurity induced by stronger PPC. That is, increasing PPC would exacerbate the adverse effect of low HH on materialism, which in turn would trigger CP.

Participants

The first author contacted the school administrators of two junior high schools of Zibo city and Jinan city in Shandong province. Then, the current study was approved by the local ethics committee of schools and got informed consent. With their approval, the first author issued questionnaires to 1,010 students in these schools, and it was ensured that each of them had owned and used a mobile phone or a computer to connect to the internet in the past year. They completed the questionnaires using their time for individual study. Finally, excluding the students who have not completed questionnaires or answered the questions regularly, 1,002 Chinese adolescents (479 boys and 523 girls) aged 10–16 years (M = 12.95 years, SD = 1.12 years, 17 adolescents did not report their age) volunteered to complete an anonymous self-report questionnaire including their demographic information and all questionnaire items.

Measures

Honesty-humility. HH was measured by the HH subscale of the 24-item Brief HEXACO Inventory (BHI) developed by De Vries (2013). In the BHI, new, short, and easy to comprehend items were written for all facets to make the inventory suitable for children and young adolescents (De Vries, 2013). HH subscale includes 4 items, namely, “I find it difficult to lie”, “I would like to know how to make lots of money in a dishonest manner”, “I want to be famous”, and “I am entitled to special treatment”. All items were rated on a 4-point Likert scale ranging from 1 (strongly disagree) to 4 (strongly agree). High average scores of the 4 items indicated high levels of HH. In a previous study, the Cronbach’s α for this subscale was 0.57 (De Vries, 2013). In this study, the Cronbach’s α for this subscale was 0.58.

Cyberbullying perpetration. CP was evaluated using the cyberbullying subscale from the Revised Cyber Bullying Inventory (RCBI), revised to be specific to Chinese adolescents by Chu and Fan (2017). Consistent with previous studies (Brack & Caltabiano, 2014; Brewer & Kerslake, 2015; Chu & Fan, 2017), we used a 4-point Likert scale (1 = It has never happened to me; 2 = It happened once; 3 = It happened 2–3 times; 4 = It happened more than three times). Some sample items are “Stealing of personal information from computer” and “Stealing of computer nicknames or screen names”. Participants assessed how frequently they performed the corresponding behavior in the last year on each question. High average scores of the 14 items indicated more CP. In a previous study, the Cronbach’s α for this subscale was 0.79 (Brack & Caltabiano, 2014). In this study, the Cronbach’s α for this subscale was 0.77.

Materialism. Material Values Scale for Children (MVS-c), revised to Chinese version for adolescents (Jiang et al., 2015), was used to evaluate adolescents’ levels of materialism. It is an 18-item questionnaire, using a 4-point Likert scale ranging from 1 (no, not at all) to 4 (yes, very much). Some sample items are “Do you think it’s important to own expensive things”, “ Do you think it’s important to get a lot of presents for your birthday”, “Do you feel unhappy if you don’t get the things you really want to have”, and “Do you think other children like you more if you have expensive brands”. High average scores of 18 items indicated stronger materialism. In a previous study, the Cronbach’s α for MVS-c was 0.93 (Opree et al., 2011). In this study, the Cronbach’s α for MVS-c was 0.94.

Parental psychological control. PPC sub-questionnaire from the Chinese version of the Parental Control Questionnaire (Wang et al., 2007) was used to measure adolescents’ PPC. It includes 18 items about fathers’ behaviors and 18 items about mothers’ behaviors. They were rated using a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Some sample items are “My father/mother tell me that I should feel guilty when I do not meet their expectations”, “My father/mother act cold and unfriendly if I do something they do not like”, and “My father/mother tell me that what they want me to do is the best for me and I should not question it”. High average scores of 36 items indicated high levels of adolescents’ PPC. In a previous study, the Cronbach’s α for this sub-questionnaire was 0.94 (Liu & Chen, 2019). In this study, the Cronbach’s α for this sub-questionnaire was 0.94.

Statistical Analyses

First, all data collected was recorded and processed preliminarily using SPSS 18.0. The outliers and missing data (< .1%) were handled with the mean imputation, which replaced the missing data with the mean of the score of the item to which the missing data belonged. Second, referring to previous studies (e.g., Geng et al., 2018), factor analysis was used to test common method biases. After principal component analysis, 21 eigenvalues > 1 were extracted, and the first factor explained 18.26% of the variance, demonstrating this study did not have significant issues with common method biases (Podsakoff et al., 2003). Third, demographic variables and research variables were analyzed by descriptive statistics, t-tests, and correlation analysis. The fourth step used Model 4 of the PROCESS macro v3.0 for SPSS (Hayes, 2013) to test the total effect of HH on CP and the mediation effect of materialism. Finally, Model 59 of PROCESS macro was used to evaluate the moderating effect of PPC and the hypothetical moderated mediation model.

Descriptive Statistics and Correlation Analysis

Descriptive statistics and correlation matrix of age, HH, CP, materialism, and PPC are provided in Table 1. T-tests indicated that girls, compared to boys, had high levels of HH, t = –2.03, p < .05. Compared to girls, boys perceived more PPC, t = 2.72, p < .01. Boys and girls did not differ on materialism (t = –1.79, p > .05) and CP (t = 1.56, p > .05). Age was significantly and positively associated with CP and PPC (p < .01). Therefore, we included gender and age as covariates in subsequent analyses.

OPEN IN VIEWER

Table 1.

Descriptive Statistics and Correlation Matrix of All Variables.

Variables	M	SD	1	2	3	4
1 Age	12.95	1.12	1
2 HH	2.27	0.55	–0.021	1
3 CP	1.07	0.17	0.096**	–0.298***	1
4 Materialism	1.78	0.60	0.056	–0.454***	0.311***	1
5 PPC	2.54	0.78	0.098**	–0.077*	0.145***	0.208***

Note. N = 1,002. HH = honesty-humility, CP= cyberbullying perpetration, PPC = parental psychological control. ^* p < .05, ^** p < .01, ^*** p < .001.

The results of bivariate correlation analyses showed that CP, materialism, and PPC were significantly and positively correlated with each other (p < .001), and they were significantly and negatively correlated with HH (p < .05).

Testing for the Mediating Effect of Materialism

Model 4 of PROCESS macro was used to test the total effect of HH on CP and the mediating effect of materialism. The results showed that the total effect of HH on CP (total effect = –0.29, p < .001, 95% CI = between –0.35 and –0.23) was significant. Furthermore, the mediation model showed that HH was negatively associated with materialism (b = –0.46, p < .001), which in turn was positively related to CP (b = 0.22, p < .001) (see Table 2). The residual direct effect was also significant (b = –.19, p < .001). Thus, materialism partially mediated the relationship between HH and CP (indirect effect = –.10, 95% CI = between –0.16 and –0.05). This model accounted for 13.61% of the variance in CP among Chinese adolescents.

OPEN IN VIEWER

Table 2.

Testing the Mediation Effect in the Relation between HH and CP.

Predictors	Model 1 (Materialism)			Model 2 (CP)
	b	t	95% CI	b	t	95% CI
Gender	0.17	3.11**	[0.06, 0.29]	–0.10	–1.61	[–0.21, –.02]
Age	0.04	1.73	[–0.01, 0.09]	0.07	2.66**	[0.02, 0.12]
HH	–0.46	–16.33***	[–0.51, –0.40]	–0.19	–5.79***	[–0.26, –0.13]
Materialism				0.22	6.67***	[0.16, 0.29]
R2	0.22			0.14
F	91.63***			39.27***

Note. N = 1,002. HH = honesty-humility, CP = cyberbullying perpetration, PPC = parental psychological control. ^* p < .05, ^** p < .01, ^*** p < .001.

Testing for the Moderating Effect of PPC

Model 59 was used to test the moderating effect of PPC on the direct and indirect associations of HH with CP via materialism, namely the moderated mediation model. As shown in Table 3, two regression models were established and estimated.

OPEN IN VIEWER

Table 3.

Testing the Moderated Mediation Effect in the Relation between HH and CP.

Predictors	Model 1 (Materialism)			Model 2 (CP)
	b	t	95% CI	b	t	95% CI
Gender	0.19	3.39***	[0.08, 0.29]	–0.09	–1.55	[–0.21, 0.02]
Age	0.03	1.14	[–0.02, 0.08]	0.06	2.46*	[0.01, 0.12]
HH	–0.42	–14.85***	[–0.47, –0.36]	–0.18	–5.32***	[–0.24, –0.11]
PPC	0.18	6.64***	[0.13, 0.24]	0.08	2.78**	[0.02, 0.14]
HH×PPC	–0.12	–4.84***	[–0.17, –0.07]	–0.11	–3.41***	[–0.18, –0.05]
Materialism				0.18	5.39***	[0.12, 0.25]
Materialism × PPC				–0.004	–0.13	[–0.07, 0.06]
R2	0.26			0.16
F	71.69***			26.09***

Note. N = 1,002. HH = honesty-humility, CP = cyberbullying perpetration, PPC = parental psychological control. ^* p < .05, ^** p < .01, ^*** p < .001.

As Table 3 illustrates, HH significantly interacted with PPC on materialism (b = –0.12, p < .001) and CP (b = –0.11, p < .001). Besides, materialism non-significantly interacted with PPC on CP (b = –0.004, p > .05). Thus, the moderated mediation model was supported. Considering that some participants have never cyberbullied others in the last year, we tested the moderated mediation model using the participants who have ever cyberbullied others within the last year. The results indicated that the moderated mediation model was also supported.

The results of simple slope analyses are shown in Figure 2 and Figure 3. The negative and direct associations of HH with materialism and CP were stronger with the increase of PPC. Specifically, for the former, if PPC = M – 1SD, then b = –0.29, p < .001; if PPC = M, then b = –0.40, p < .001; if PPC = M + 1SD, then b = –0.55, p < .001; and for the later, if PPC = M – 1SD, then b = –0.06, p > .05; if PPC = M, then b = –0.17, p < .001; if PPC = M + 1SD, then b = –0.30, p < .001.

OPEN IN VIEWER

Figure 2.

Note. PPC = parental psychological control. ^*** p < .001.

OPEN IN VIEWER

Figure 3.

Note. PPC = parental psychological control. ^*** p < .001.

The results showed the indirect relationship between HH and CP became stronger, with the level of PPC increasing. Specifically, when PPC = M – 1SD, the indirect effect was –0.055 (95% CI = between –0.117 and –0.011), when PPC = M, the indirect effect was –0.075 (95% CI = between –0.138 and –0.025), and when PPC = M + 1SD, the indirect effect was –0.099 (95% CI = between –0.175 and –0.025). Thus, PPC exacerbated the direct and indirect associations between HH and CP.

HH and CP

We first verified the negative association of HH with CP. That is, adolescents low in HH were more likely to bully others online, which supported Hypothesis 1. This result was consistent with the conclusions of some studies exploring the association of HH with traditional bullying (Book et al., 2012; Farrell et al., 2014; Volk et al., 2018) and internet deviant behavior (e.g., sexting) (Morelli et al., 2020). Low HH has been verified to be positively related to unethical tendencies, including cheating, sexual harassment, unethical business decision making, and deviant behaviors at work (Hershfield et al., 2012; Lee et al., 2003, 2008). Moreover, people low in HH are more likely to engage in criminal activities (Gelder & Vries, 2012). Thereby, individuals low in HH tend to utilize moral disengagement tactics to minimize feelings of guilt and avoid self-sanctions (Ogunfowora & Bourdage, 2014), which in turn increases the risk of CP (Chen et al., 2016). Consequently, adolescents low in HH are more likely to engage in CP.

Mediating Role of Materialism

We expected that materialism would mediate the relationship between HH and CP, which was supported by the results. The results also supported McCrae and Costa’s (2008) model and values conflict theory (Burroughs & Rindfleisch, 2002), inferring that personalities create a foundation for materialistic values, and materialistic values conflict with prosocial values, which may reduce the likelihood of CP. The current study innovatively integrated these theories to explain why adolescents low in HH are more likely to engage in CP. Moreover, these results also supported the explanation mechanism of materialism in the association of a personality trait (HH) with online antisocial behavior (CP), consistent with the models that explain particular types of social behaviors (Harnish & Bridges, 2014; Otero-López & Villardefrancos, 2013). It could be explained that individuals with low levels of HH are more likely to be attracted by luxury goods, prestige, or high social status, and further tend to conduct deviant behaviors (e.g., CP) to achieve their aims (Allgaier et al., 2015).

In addition to the overall mediation results, results of the individual relationships analyzed in our study are also noteworthy. The mediation model reflected that HH was negatively associated with materialism, which was in turn positively related to CP. For the first part of the mediating process, adolescents low in HH tended to endorse materialistic values. This result supported the model of McCrae and Costa (2008) and was consistent with conclusions of previous studies exploring the relationship between HH and materialism (Górnik-Durose & Pilch, 2016; Pilch & Górnik-Durose, 2016). Individuals high in HH tend to “be fair and genuine in dealing with others, in the sense of cooperating with others” (Ashton & Lee, 2007, p. 156), and place less value on power, money, and wealth (Lee et al., 2013). Thus, adolescents low in HH are more likely to endorse materialistic values and strive for money.

For the latter part of the mediation process, adolescents pursuing materialistic goals were more likely to bully others online. This result supported values conflict theory (Burroughs & Rindfleisch, 2002). It was also consistent with the conclusion of a study that explained the positive relationship between materialism and CP from an evolutionary perspective (Wang, Wang, et al., 2021), and the results of studies on antisocial behaviors (Kasser, 2016). Materialistic adolescents tend to make up excuses, remove responsibility, and self-handicap to protect their threatened self-image (Christopher & Schlenker, 2005), thus, they often experience feelings of social anxiety (Christopher et al., 2007), which has been verified to be positively related to CP among Chinese adolescents (Wang, Xie, et al., 2019). Considering the risk effect of materialism on CP and the growing materialistic values among adolescents (Chaplin et al., 2018), it is pressingly needed to make efforts to intervene the materialistic values among adolescents low in HH.

Moderating Role of PPC

To know when HH is related to CP, the current study explored the moderating effect of PPC. The results supported the hypothetical moderated mediation model and indicated that PPC moderated the direct associations of HH with materialism and CP, and the indirect association of HH with CP, supporting Hypotheses 3 and 4.

First, the moderated mediation model indicated that increased PPC exacerbated the direct effect of HH on CP. This study was the first to explore the interaction of HH with a parental factor on CP. Specifically, stronger PPC increased CP among adolescents low in HH, rather than those high in HH. This result supported the STOA Model (De Vries et al., 2016), inferring that some situational factors (e.g., PPC) would increase the differences in behavioral outcomes among adolescents with different levels of HH (Barends et al., 2019). This study expanded the STOA Model to explain the formation of online antisocial behaviors (i.e., CP). This result was also consistent with the studies that found the interaction of HH with situational characteristics on prosocial and antisocial behaviors (Allgaier et al., 2015; Farrell et al., 2017). Based on the description of HH, people high in HH would behave socially desirable irrespective of the situational circumstances, in contrast, those low in HH might behave depending on situational characteristics (Allgaier et al., 2015). PPC frustrates people’s basic needs (Soenens & Vansteenkiste, 2010) which further induces people’s aggression (He et al., 2019). More importantly, even though all individuals were exposed to the same stressor (stronger PPC), individuals low (versus high) in HH were more affected by it (Allgaier et al., 2015). Thus, the adverse effect of PPC on cyberbullying behavior only occurred among adolescents low in HH. It may be explained that adolescents high in HH may think or use other kinds of responses to cope with stressors rather than engage in antisocial behaviors (Allgaier et al., 2015).

Second, the moderated mediation model indicated that increased PPC exacerbated the direct effect of HH on materialism, which further exacerbated the indirect effect of HH on CP. Specifically, adolescents low in HH engaged in stronger materialistic pursuit when they coped with stronger PPC, but those high in HH showed low levels of materialism in general. This result supported the STOA Model (De Vries et al., 2016), inferring that stronger PPC would induce the individual differences of adolescents with different levels of HH. The result was also in line with a study regarding the interaction between student-student relationship and narcissism in shaping materialistic values (Ouyang et al., 2020). The current study extended previous knowledge by testing the interaction of a new trait with a new stressful environmental factor. Adolescents react differently to the same stressor due to different personalities (Bolger & Zuckerman, 1995). It could be plausible that individuals low in HH originally pursue materialistic goals (Górnik-Durose & Pilch, 2016). When exposed to stronger PPC, individuals low in HH tend to pursue materialistic goals to cope with stress and insecurity usefully and effectively before thinking or using other kinds of responses. By contrast, individuals high in HH may give priority to other coping attempts to copy with stressors.