Correcting for norm misperception of anti-bullying attitudes

Abstract

Research has documented the presence of norm misperception in the context of school bullying, as children and adolescents typically underestimate the degree to which their peers disapprove of bullying behavior. Despite commonly held attitudes in opposition to bullies and in support of helping victims, widespread misperception of the norm makes students vulnerable to acting in a manner that reinforces bullying, as they aim to align themselves with perceived peer beliefs. This study investigated whether personalized normative feedback, a social norms intervention that juxtaposes individuals’ own perceptions of peer norms against their peers’ true normative values, could operate as a mechanism by which to reduce norm misperception of peer attitudes toward bullying. Whereas this type of intervention has shown promising effects in a variety of contexts, no study to date has examined its utility in the specific context of bullying. Baseline participants included 188 seventh grade students, 175 of whom were randomized into four study groups for follow-up data collection. Individuals in the experimental condition received personalized normative feedback on attitudes toward bullying. Control conditions were the following: general normative feedback on anti-bullying attitudes, the absence of normative feedback, and personalized normative feedback on a construct separate from bullying (i.e., antidrug use attitudes). Findings indicated that personalized normative feedback on bullying attitudes led to significant change in perceived peer attitudes in the direction of the group norm, with an effect size in the small-to-medium range. No intervention effects emerged on personal attitude change. Implications highlight strategies for improving the strength of similar interventions in future research as well as the positive clinical outcomes that could result from reduced norm misperception and increased engagement in prosocial bystander behavior.

Keywords

Bullying victimization social norms norm misperception personalized normative feedback bystanders

Introduction

School-age bullying is recognized as a specialized form of aggression comprising three components: (1) repeated physical, verbal, or relational attacks, (2) committed by a person more powerful than the victim, with (3) intent to harm (Olweus, 1993). This behavior is prevalent, as research findings estimate that one third to one half of students are directly involved in bullying incidents as bullies, victims, or bully-victims (Schwartz, 2000). The negative short- and long-term effects associated with school bullying are well-documented, with perpetrators demonstrating delinquency, violence, and criminal behavior (Sourander et al., 2007) and victims vulnerable to anxiety, depression, poor academic performance, and school avoidance (Brendgen & Poulin, 2018). These detrimental outcomes drive home the need for effectiveness research aimed at implementing evidence-based programs capable of fostering meaningful change.

Social Context of Bullying and Bystander Behavior

Bullying typically occurs within a larger social context in which it influences and is influenced by the attitudes and behaviors of others (Simon & Nail, 2013). Present for a vast majority of these bullying incidents (Hawkins, Pepler, & Craig, 2001), peer bystanders serve a critical role, either exacerbating or attenuating the degree to which bullying permeates their school environments (Salmivalli, Voeten, & Poskiparta, 2011). The literature has identified a number of bystander roles: assistants join in on the bullying, reinforcers provide positive feedback to bullies by laughing/cheering (active) or by simply providing an audience (passive), defenders offer support to victims, and outsiders remove themselves from bullying situations (Salmivalli, Lagerspetz, Bjorkqvist, Osterman, & Kaukiainen, 1996). Despite evidence that active intervention on behalf of victims often puts a quick end to bullying episodes, peers only defend one another in approximately one of every five incidents (Hawkins et al., 2001). These findings call into question why more students do not intervene, particularly as research has consistently demonstrated their generally held prosocial attitudes and view that intervention is the right course of action (Hymel, Schonert-Reichl, Bonanno, Vaillancourt, & Henderson, 2010).

Norm Misperception of Attitudes Toward Bullying and Bystander Behavior

Bystanders often cite such factors as concern for their own safety and potential hits to their social status as barriers to standing up for victimized peers (Stevens, DeBourdeaudhuij, & Van Oost, 2000). Importantly, these are valid considerations, as challenging bullying behavior certainly exposes students to physical and emotional harm and carries with it the risk of making oneself an ongoing target (Poüyhoünen, Juvonen, & Salmivalli, 2010). Further, disrupting the status quo can have the consequence of diminishing one’s reputation (Juvonen & Galván, 2008), which has real implications for general well-being and sense of self-worth (Gest, Sesma, Masten, & Tellegen, 2006). While many instances of nonintervention are driven by safety concerns, many others likely reflect awareness of the social cost associated with deviating from perceived peer norms. Social norms theory originated with the work of Perkins and Berkowitz (1986) and describes situations in which individuals misperceive the attitudes of their peers, falsely believing them to be different from their own (Berkowitz, 2003). This phenomenon, also called “pluralistic ignorance” (Miller & McFarland, 1991), can lead to behavior that is intended to better approximate the perceived group norm (Prentice & Miller, 1993) but is inconsistent with the individuals’ true beliefs and values (Miller & McFarland, 1991).

Norm misperception typically reflects exaggerated approval of negative attitudes or behavior, as those problematic or risky in nature are usually overestimated while those healthy or protective in nature are usually underestimated. Although less common, norm misperception can also emerge with regard to exaggerated approval of positive attitudes or behavior (Raggatt et al., 2018). This process results from observation of proximal (e.g., peers) and distal (e.g., media) models, leading to a group consensus of the norm based on which behaviors are rewarded and which behaviors are punished (Lewis, Litt, Cronce, Blayney, & Gilmore, 2014). These norms are then continually reinforced, as individuals often seek to reap the social benefits of conformity rather than assume the risk that comes with deviation (Veenstra, Dijkstra, & Kreager, 2018). This cycle points to the ubiquity of social comparison and highlights the valuable role that accurately perceived peer norms could play in the development of healthy attitudes and behaviors.

Social norms theory can be extended to situations in which individuals refrain from intervening in the problem behavior of others (Berkowitz, 2004). These bystanders, as “carriers of the misperception” (Perkins, 1997), contribute in a meaningful way to the social climate that allows the problem behavior to persist. In the specific context of school bullying, children perceive themselves as holding more prosocial attitudes than their peers (Sandstrom, Makover, & Bartini, 2013). However, it is not possible for members of a group to systematically possess more prosocial attitudes than the group as a whole. This widespread misperception of the norm makes students vulnerable to responding in a manner that reinforces bullying, as they may feel motivated to align themselves with perceived peer beliefs despite the resulting conflict with their personally held values. This concern is further compounded by the perception of students who bully as popular among their classmates (Juvonen & Galván, 2008), which has the potential to negatively impact behavior across the peer group. Not only may popular students model the bullying behavior of their popular peers in order to maintain their popularity (Peets, Pöyhönen, Juvonen, & Salmivalli, 2015), but unpopular students may also be inclined to model this behavior in an effort to elevate their own social status (Dijkstra, Lindenberg, & Veenstra, 2008).

Although few studies have attempted to uncover the mechanism behind this norm misperception as it relates to beliefs about school bullying, pluralistic ignorance has generally been identified as the predominant explanatory factor (Sandstrom & Bartini, 2010; Sandstrom et al., 2013). This phenomenon may go hand in hand with the fundamental attribution error, which describes a tendency to attribute others’ behavior to dispositional traits without considering the role of situational factors (Andrews, 2001). Taken together, it may be that when children observe assistant, reinforcer, and outsider behavior, they assume it reflects their peers’ internal thoughts and feelings (i.e., approval of bullying). Thus, despite personal disapproval of bullying and inclinations toward helping the victim, students match the bystander behavior of their peers, aiming to align themselves with the perceived group norm. Because students fail to recognize that the social pressures motivating their own bystander behavior could also explain the behavior of their peers (Willer, Kuwabara, & Macy, 2009), the cycle of norm misperception and bystander passivity is able to persist. Several studies (Pozzoli & Gini, 2010; Sandstrom & Bartini, 2010; Sandstrom et al., 2013) have indeed shown that certain bystander responses, including assisting the bully, actively or passively reinforcing the bully, and defending the victim, were consistent with perceived peer attitudes toward bullying and helping behaviors rather than personal beliefs.

Correcting for Norm Misperception

Social norms theory predicts that by providing accurate normative information, perceived peer attitudes will shift in the direction of the true norm, leading to similar changes in personal attitudes (Perkins, Craig, & Perkins, 2011). Through the paired phenomena of pluralistic ignorance and the fundamental attribution error, a majority of students come to view themselves as holding more prosocial attitudes than their peers, resulting in the establishment of relative anchor points that mark this discrepancy. Thus, the presentation of corrective normative data serves to disrupt and shift students’ perceived peer attitudes anchor. By making a similar adjustment to their personal attitudes, students are able to maintain their perceptions of themselves as prosocial in comparison to the members of their peer group. For the minority of students who do not characterize themselves as prosocial, but rather, hold the pro-bullying attitudes they believe their peers to share, the process of personal attitude change is rooted in cognitive dissonance (Berkowitz, 2004). In essence, accurate normative feedback reveals to these students that their attitudes are in direct contrast to those of their peers, challenging their previously held notions of fitting in with their peer group. Provided that this new information is perceived as credible, students’ experience of cognitive dissonance is best relieved through an adjustment of their personal attitudes to better match the prosocial attitudes of their peers.

Social norms theory further predicts that correcting for norm misperception by revealing the true, healthier norm will have a positive impact on most individuals, either by reducing their participation in problem behavior or by encouraging them to engage in protective behavior (Berkowitz, 2003). Middle and high school students may be particularly receptive to corrective normative information, given the heightened attention to and desire for peer approval that is characteristic of adolescence (Gremmen et al., 2018; Veenstra & Dijkstra, 2011). Widespread increased understanding of peers’ disapproving attitudes toward bullying could, in effect, minimize students’ perceptions of bullying as an effective means for climbing the social ladder (Oldenburg, Van Duijn, & Veenstra, 2018). Further, popular students who previously had been hesitant to intervene on victims’ behalves (Kaufman, Kretschmer, Huitsing, & Veenstra, 2018) may feel empowered to defend victims in the setting of diminished social risk (Peets et al., 2015). This not only would offer reprieve to victimized students but could also contribute to a more positive social environment in which popular students increasingly display prosocial behavior, perhaps providing motivation for their peers to follow.

The present study aimed to employ a social norms intervention that provided participants with direct feedback about their misperceptions of peers’ attitudes toward bullying behavior. This was accomplished through the use of personalized normative feedback, which juxtaposes individuals’ own perceptions of peer norms against their peers’ true normative values (Reid & Aiken, 2013). This type of intervention has been utilized successfully in a variety of contexts, including drinking behavior (Neighbors, Larimer, & Lewis, 2004), gambling habits (Celio & Lisman, 2014), and sunscreen usage (Reid & Aiken, 2013). Results largely indicated reduced norm misperception and either positive plans toward or actual behavioral change in the direction of the true norm. To date, no study has examined the correction of norm misperception through personalized normative feedback as a mechanism by which to change perceived peer and personal attitudes in the context of bullying.

Purpose and Hypotheses

The first aim of the study was to confirm that norm misperception of bullying attitudes was present in the current sample. Perceived peer attitudes toward bullying were compared to the actual group norm, operationalized as the mean of participants’ baseline personal attitude ratings across all classrooms. It was hypothesized that, at baseline, (1) students would underestimate their peers’ anti-bullying and pro-helping attitudes.

The second aim of the study was to examine whether a targeted intervention could effectively reduce norm misperception. This intervention provided personalized normative feedback to participants, illustrating the discrepancy between their ratings of perceived peer attitudes toward bullying and the actual group norm. Because the social norms intervention model predicts that both perceived peer and personal attitudes will shift in the direction of the true norm, changes in norm misperception are not measured by evaluating the gap between perceived and actual norms. Rather, changes in norm misperception can be evaluated by assessing differences between baseline and post-manipulation ratings for both perceived peer and personal attitudes (Perkins et al., 2011). Post-manipulation perceived peer attitude ratings, controlled for baseline scores, were compared between the experimental and three control conditions. These controls included normative feedback on anti-bullying attitudes without a personalized component, feedback with no meaningful normative data, and personalized normative feedback on a construct different from anti-bullying attitudes (i.e., antidrug use attitudes). It was hypothesized that, following the manipulation, (2a) increases in anti-bullying and pro-helping perceived peer attitudes would be greater in the experimental group than in the control groups. Similarly, post-manipulation personal attitude ratings, controlled for baseline scores, were compared between the experimental and three control conditions. It was hypothesized that, following the manipulation, (2b) increases in anti-bullying and pro-helping personal attitudes would be greater in the experimental group than in the control groups.

Method

Participants and Procedure

Participants were drawn from a metropolitan region middle school located in the southern U.S. Parental informed consent forms were sent home with the full seventh grade population (n = 226), spanning 10 classrooms, and were returned at a rate of 93%. Ultimately, baseline data were collected in a single day at the classroom level from 83% (n = 188) of seventh grade students, with the number of participants in each classroom ranging from 16 (70% of classroom students) to 26 (93% of classroom students). Those who failed to return their consent forms, whose parents did not provide consent, and who were absent during data collection despite having parental consent were not included. Participants were drawn from a sample of pre-adolescents and adolescents (ages 12–14) and were 49.5% male, 51.5% female, 91% African American, and 9% Other. Of the 188 baseline subjects, 175 received parental consent (and provided assent) to participate in the follow-up data collection. These students were randomly assigned to the experimental condition or to one of three control conditions. Over the course of 3 weeks, participants met one-on-one with researchers in private classrooms. They first received the study manipulation per their randomized condition and then repeated the measures initially completed during baseline data collection.

Measures

Personal and Perceived Peer Anti-Bullying Attitudes

At baseline and post-manipulation, participants were presented with 10 statements about bullying attitudes, adapted from a measure of children’s moral disengagement in the context of bullying (Hymel, Rocke-Henderson, & Bonanno, 2005), and utilized in an examination of norm misperception’s role in peer harassment at school (Sandstrom et al., 2013). Consistent with prior research on social norms estimation (Prentice & Miller, 1993; Shelton & Richeson, 2005), participants were asked to indicate (1) the extent to which they agree with each statement and (2) the extent to which they think their peers agree with each statement. For example, one item reads (a) “I think that kids who get picked on a lot usually deserve it” and (b) “Most kids in my grade think that kids who get picked on a lot usually deserve it.” Responses were provided on a 4-point Likert-type scale ranging from 1 (strongly disagree) to 4 (strongly agree). A subset of items was reverse scored, such that higher scores represented more anti-bullying attitudes (e.g., less tolerance for bullies, more empathy for victims, and greater expectation for bystanders to intervene) and lower scores reflected less prosocial attitudes across these domains. Prior research (Sandstrom et al., 2013) has documented an α of .83 for both the Perceived Peer Anti-Bullying Attitudes scale and the Personal Anti-Bullying Attitudes scale. Similarly, strong internal consistency was demonstrated in the current sample, with α values of .86 and .80, respectively.

Perceived Peer Disapproval of Drug Use

For control group purposes, at baseline and post-manipulation, participants completed a 3-item self-report adaptation of the Perceived Peer Disapproval measure, previously utilized in an examination of individual and social factors that influence adolescent drug use (Rice, Donohew, & Clayton, 2003). The items center on the acceptability of smoking cigarettes, drinking alcohol, and smoking marijuana. As with the Personal and Perceived Peer Anti-Bullying Attitudes measure, participants were asked to indicate (1) the extent to which they agree with each statement and (2) the extent to which they think their peers agree with each statement. Ratings were provided on a 4-point Likert-type scale ranging from 1 (strongly disagree) to 4 (strongly agree). On all 3 items, higher scores reflected antidrug use attitudes. Strong internal consistency was demonstrated in the current sample, revealing an α of .83 on the Perceived Peer Anti-Drug Use Attitudes scale and an α of .80 on the Personal Anti-Drug Use Attitudes scale.

Manipulation and Manipulation Check

Approximately 1 month after baseline data were collected and immediately preceding follow-up data collection, participants received a brief manipulation. Consistent with prior research (Reid & Aiken, 2013), this manipulation was administered even in cases of participants overestimating their peers’ anti-bullying and pro-helping attitudes. All subjects were shown one of four bar graphs, the content of which was determined by their study condition, and offered an accompanying verbal explanation of the depicted information. Following this graph presentation, a manipulation check in the form of a simple test question was employed in order to ensure participants’ understanding of the feedback provided.

Participants in the experimental condition (i.e., Bullying Attitudes: Personalized Normative Feedback; n = 44) received personalized normative feedback from the Personal and Perceived Peer Anti-Bullying Attitudes measure, illustrating both their ratings of perceived peer anti-bullying attitudes and the actual attitudinal norm (i.e., mean of baseline personal anti-bullying attitudes). They were encouraged to consider and compare their perceptions with the true normative value.

Participants in the first control condition (i.e., Bullying Attitudes: General Normative Feedback; n = 44) received half of this personalized normative feedback chart, depicting only the actual attitudinal norm, absent of a personalized comparison (i.e., participants’ own ratings of perceived peer anti-bullying attitudes). By providing information about the group norm in a format paralleling a universal intervention, the researchers were able to examine whether a personalized component of normative feedback contributed meaningfully to any significant change in the experimental condition.

Participants in the second control condition (i.e., Bullying Attitudes: No Normative Feedback; n = 43) received the remaining half of the experimental condition’s personalized normative feedback chart, depicting only their ratings of perceived peer anti-bullying attitudes. Failing to provide a comparison to the actual attitudinal norm allowed this control group to act as a placebo, indicating that any significant change in the experimental condition was due to the presented social comparison.

Participants in the third control condition (i.e., Drug Use Attitudes: Personalized Normative Feedback; n = 44) received personalized normative feedback from the Perceived Peer Disapproval of Drug Use measure, illustrating both their ratings of perceived peer antidrug use attitudes and the actual attitudinal norm (i.e., mean of baseline personal antidrug use attitudes). As with the experimental condition, they were encouraged to consider and compare their perceptions with the true normative value. By providing personalized normative feedback on a separate construct, the researchers were able to examine whether any significant change in the experimental condition was driven purely by general social comparison or, rather, by social comparison specific to bullying attitudes.

Results

Preliminary Analyses: Baseline Descriptives and Randomization of Sample

Figure 1 shows the means and standard deviations of the study variables at baseline (n = 188). Initial analyses of skewness and kurtosis revealed primarily normal distributions; however, the Personal Anti-Drug Use Attitudes variable emerged as negatively skewed. Further exploration of the data revealed five outliers, with z-scores falling beyond a value of ±3.29. These scores reflect a small subset of study participants who, in contrast with their peers, strongly endorsed positive attitudes toward drug use. To better normalize the distribution, their scores were capped to match the peer group’s next highest level of drug use endorsement. Table 1 presents study variable correlations at baseline for the full sample.

Figure 1.

Baseline Study Variable Descriptives and Misperception of Peer Norms.

Table 1.

Baseline Study Variable Correlations.

	Personal anti-bullying attitudes	Perceived peer anti-bullying attitudes	Personal antidrug use attitudes	Perceived peer antidrug use attitudes
Personal anti-bullying attitudes	—
Perceived peer anti-bullying attitudes	.15*	—
Personal antidrug use attitudes	.50**	.09	—
Perceived peer antidrug use attitudes	.16*	.52**	.33**	—

Note. n = 188.

*Correlation is significant at the .05 level.

**Correlation is significant at the .01 level.

Prior to collection of post-manipulation data (n = 175), participating baseline subjects were randomized into four conditions: one experimental, “Bullying Attitudes: Personalized Normative Feedback” (n = 44), and three controls, “Bullying Attitudes: General Normative Feedback” (n = 44), “Bullying Attitudes: No Normative Feedback” (n = 43), and “Drug Use Attitudes: Personalized Normative Feedback” (n = 44). To assess for baseline differences between groups, a one-way multivariate analysis of variance (MANOVA) was performed. The following variables were included: perceived peer anti-bullying attitudes, personal anti-bullying attitudes, perceived peer antidrug use attitudes, and personal antidrug use attitudes. The MANOVA yielded no significant differences between groups, F(12, 432) = .67, p = .78.

Primary Analyses: Baseline Norm Misperception

Aim 1

A one-sample t-test was conducted to evaluate whether norm misperception of bullying attitudes was present in the current sample. Perceived peer anti-bullying attitudes (M = 2.59) were significantly discrepant from the actual attitudinal norm (M = 3.50) with a large effect size, t(184) = −20.35, p < .001, d = 1.50 (see Figure 1). Thus, Hypothesis 1 was confirmed, as participants significantly underestimated their peers’ anti-bullying and pro-helping attitudes.

Of the current study’s 188 baseline participants, 94% provided an estimation of perceived peer anti-bullying attitudes falling below the true norm. Eleven subjects overestimated their peers’ anti-bullying and pro-helping attitudes, and only nine participated in the manipulation and post-manipulation data collection. During baseline randomization of the full study sample, these participants landed in the following conditions: Experimental: “Bullying Attitudes: Personalized Normative Feedback” (n = 2), Control 1: “Bullying Attitudes: General Normative Feedback” (n = 3), Control 2: “Bullying Attitudes: No Normative Feedback” (n = 2), and Control 3: “Drug Use Attitudes: Personalized Normative Feedback” (n = 2). Following administration of the manipulation, four participants maintained their overestimation of perceived peer anti-bullying attitudes, one dropped their ratings to match the actual attitudinal norm, and four showed a slight underestimation of their peers’ anti-bullying attitudes, with no clear patterns across conditions.

Intermediate Analyses: Control Condition Design

To mirror the use of participants’ true attitudes toward bullying in the experimental condition, the current study proposed utilizing participants’ true attitudes toward drug use in the “Drug Use Attitudes: Personalized Normative Feedback” control condition. However, the feasibility of this proposal rested on two prerequisites: first, that norm misperception of drug use attitudes be present in the current sample. This stipulation was evaluated via a one-sample t-test. Perceived peer antidrug use attitudes (M = 2.65) were significantly discrepant from the actual attitudinal norm (M = 3.57) with a large effect size, t(186) = −15.94, p < .001, d = 1.17, as participants significantly underestimated their peers’ disapproval of drug use.

Having met this first prerequisite, it was also necessary that the degree of norm misperception with regard to bullying and drug use attitudes be comparable. This requirement was assessed through a paired-sample t-test comparing bullying and drug use difference scores, computed for both constructs by subtracting each participant’s perceived peer ratings from the actual attitudinal norm. No significant differences emerged on the degree of norm misperception between anti-bullying (M = .91) and antidrug use (M = .92) attitudes, t(183) = −.12, p = .91.

Primary Analyses: Group Differences on Perceived Peer and Personal Anti-bullying Attitudes

Aim 2

A one-way multivariate analysis of covariance (MANCOVA) was performed to examine whether the implemented intervention effectively reduced norm misperception with regard to bullying attitudes. Study condition (i.e., experimental and controls) was entered as the independent variable, and post-manipulation (Time 2) scores for perceived peer and personal anti-bullying attitudes were entered as dependent variables. Preliminary analyses revealed moderate correlations between baseline values and Time 2 scores, significant at p < .001: perceived peer anti-bulling attitudes, r = .53; personal anti-bullying attitudes, r = .54. Thus, baseline values for these variables were entered as covariates in order to account for their effect on the degree of change across time (Neighbors et al., 2004; Reid & Aiken, 2013).

The MANCOVA yielded a statistically significant difference between groups across the perceived peer and personal anti-bullying attitudes outcome measures, F(6, 328) = 2.89, p = .009, η² = .05, 90% confidence interval (CI): [.01, .08]. Follow-up univariate analyses of variance revealed a statistically significant difference between groups on perceived peer anti-bullying attitudes Time 2 scores and an effect size within the small-to-medium range, F(3, 165) = 4.95, p = .003, η² = .08, 90% CI [.02, .14] (see Figure 2), using a Bonferroni-adjusted α level of .025. No significant difference between groups on personal anti-bullying attitudes Time 2 scores emerged, F(3, 165) = .22, p = .88, η² = .004, 90% CI [.00, .01] (see Figure 3). In order to compare the control conditions against the experimental condition with regard to perceived peer anti-bullying attitudes, a simple contrast was added to the analysis. Reported mean Time 2 scores for each study condition were adjusted to account for the effect of the covariates. Results indicated that Time 2 scores were significantly greater in the “Bullying Attitudes: Personalized Normative Feedback” (M = 2.93) experimental condition than in the following controls: “Bullying Attitudes: No Normative Feedback” (M = 2.59) at p < .001 and “Drug Use Attitudes: Personalized Normative Feedback” (M = 2.66) at p = .006, using a Bonferroni-adjusted α level of .0125. No significant difference was found between the experimental group and the “Bullying Attitudes: General Normative Feedback” (M = 2.75) control group, p = .11. Control conditions were compared to one another with regard to perceived peer anti-bullying attitudes via simple contrasts, yielding no significant findings.

Figure 2.

Post-manipulation Perceived Peer Anti-bullying Attitudes.

Figure 3.

Post-manipulation Personal Anti-bullying Attitudes.

Secondary Analyses: Restricted Range on Personal Anti-bullying Attitudes Scale

Additional analyses were completed in order to explore whether restricted range on the Personal and Perceived Peer Anti-Bullying Attitudes scale could account for the absence of hypothesized personal attitude change. With a baseline mean of 3.50 and an SD of .40 on a 4-point Likert-type scale, there was little room for upward movement of personal attitude ratings during post-manipulation data collection. Three hierarchical regression analyses were performed, each examining whether baseline personal anti-bullying attitudes moderated differences between the experimental and one of three control groups on post-manipulation personal anti-bullying attitudes. Findings were nonsignificant across all three regressions, suggesting a reduced likelihood that level of baseline personal anti-bullying attitudes moderated the group effect. Specific results were as follows: Regression 1, conditions including “Bullying Attitudes: Personalized Normative Feedback” and “Bullying Attitudes: General Normative Feedback” (B = .263, p = .14); Regression 2, conditions including “Bullying Attitudes: Personalized Normative Feedback” and “Bullying Attitudes: No Normative Feedback” (B = .211, p = .25); Regression 3, conditions including “Bullying Attitudes: Personalized Normative Feedback” and “Drug Use Attitudes: Personalized Normative Feedback” (B = .199, p = .24).

Discussion

Research has documented that bystander intervention has some ability to curb school bullying behavior (Hawkins et al., 2001). However, studies show that despite personal attitudes opposed to bullying and in support of helping, students rarely intervene on victims’ behalves (Hymel et al., 2010). Consistent with social norms theory, this hesitation may reflect individuals’ attempts to behave in line with misperceived group norms, given that students often overestimate their peers’ endorsement of bullying (Pozzoli & Gini, 2010; Sandstrom et al., 2013). In keeping with previous findings, norm misperception of bullying attitudes emerged in the current study sample, as individuals generally viewed themselves as more prosocial than their peers. The present study extended prior research by examining whether a brief manipulation could reduce this norm misperception, thereby bringing students to a more accurate understanding of their peers’ attitudes regarding bullying and helping behavior.

Perceived Peer Attitude Change

To explore whether this norm misperception could be corrected to some degree, subjects were presented with one of four bar graphs, each depicting different pieces of normative information. It was hypothesized that, following the administration of the study manipulation, participants in the experimental condition would significantly alter their perceived peer anti-bullying attitudes in the direction of the group norm. This hypothesis was partially realized, as predicted differences emerged between subjects who received personalized normative feedback on bullying attitudes (comparing their own perceived peer anti-bullying attitudes to the actual attitudinal norm) and those who received no normative feedback (only their own perceived peer anti-bullying attitudes), as well as those who received personalized normative feedback on drug use attitudes (comparing their own perceived peer antidrug use attitudes to the actual attitudinal norm). These findings illustrate that changes in perceived peer attitudes were due not only to a social comparison but to a social comparison specifically on the construct in question.

Results of the current study offer early evidence that personalized normative feedback has the potential to reduce the presence of norm misperception with an adolescent population in the context of school bullying. However, in contrast to the stated hypothesis, no differences emerged on perceived peer attitude change between the personalized and general normative feedback study groups, indicating a possibility that either type of feedback could serve as an effective intervention. One important consideration centers on the issue of power, as observable group differences in Figure 2 offer some indication that significant findings could emerge with a larger study sample. Alternatively, should nonsignificant findings hold in future research, it may be that participants who receive general feedback consider the presented depiction of the true norm in relation to what they know to be their own expectations, essentially adding a personalized component to the feedback provided.

Personal Attitude Change

Social norms theory predicts that as perceived peer attitudes change, so will personal attitudes. However, this hypothesis was not supported in the present study, as no group differences emerged regarding change in personal anti-bullying attitudes. Although the potential impact of low power or restricted range cannot be ruled out, a 90% CI under .01 paired with visual inspection of Figure 3 lends evidence to suggest that no true effect occurred.

One possible explanation for this outcome lies in the stability of personal attitudes. While individuals can only make inferences about the mental states of others based on their external, observable behavior, they have direct access to their own mental states (Andrews, 2001). Because their internal thoughts and feelings are personally knowable, it logically follows that individuals may be more certain of their own beliefs than they are of their peers’. Attitude certainty has been linked to both greater resistance to persuasion (Petrocelli, Tormala, & Rucker, 2007) and greater attitude stability over time (Bassili, 1996). Thus, it may be that while perceived peer attitudes have some flexibility to adjust in response to new information, personal attitudes do not share the same fluidity.

Alternatively, it’s possible that personal attitudes are not entirely unchanging but are simply slow to change. Theoretically, personal attitude change would follow perceived peer attitude change on the basis of social comparison (Mussweiler, 2003) through a process of anchoring and adjustment (Tamir & Mitchell, 2013). However, this comparison relies on the accessibility of the comparative target (i.e., how prosocial others are). The more accessible this piece of knowledge, the more likely it is to be used in a social comparison and the more likely it is to influence characterization of the self (Higgins, 1996). The fact that perceived peer attitude ratings were newly adjusted may have diminished participants’ ability to access them as a reference point for personal attitude ratings. In fact, there is evidence to suggest that adjustment from an anchor point requires both time and cognitive effort (Epley, Keysar, Van Boven, & Gilovich, 2004). Personal attitude change, therefore, may only occur once individuals’ newly acquired perceived peer attitudes have had the opportunity to settle and operate as a more salient comparative target.

Real-World Implications

The current study yielded a relatively small effect size (small-to-medium range) on perceived peer attitude change in comparison to the findings presented in other norm misperception research utilizing a similar personalized normative feedback intervention (Celio & Lisman, 2014; Neighbors et al., 2004; Reid & Aiken, 2013). These studies yielded medium-to-large effect sizes on perceived peer attitude change at immediate, short-term, and long-term follow-up. Further, the investigations that examined real-world behavioral change found effect sizes within the small-to-medium range (Neighbors et al., 2004; Reid & Aiken, 2013). Given that these studies indicated greater change at the attitude level than at the behavioral level, it is reasonable to conclude that for the current study’s intervention to have a meaningful, real-world impact, the strength of the intervention would need to improve. This is particularly true in light of the social stakes (Juvonen & Galván, 2008) associated with bystander helping in the context of school bullying. Whereas the previously noted studies have demonstrated meaningful change with regard to constructs that reflect individual choices and self-selected peer groups (e.g., drinking behavior, gambling habits, sunscreen usage), bullying and bystander interventions seek to produce similar change on a construct that is relational in nature and in a setting that offers individuals little control or flexibility regarding those comprising their larger social environments. Future investigations would likely benefit from further consideration of the ways in which norm misperception interventions could address the unique factors related to bullying and bystander behavior in middle and high school populations.

Limitations and Future Directions

Of particular importance, data were collected from a racially homogenous sample, with participants comprising only seventh grade students from a single school. This lack of diversity limits the generalizability of significant findings. Future investigations should seek to collect data from a more heterogeneous sample, allowing for the examination of intervention effects across race, grade, and school culture.

Second, factors related to methodology and study design warrant consideration. In view of recent findings regarding the average effect size of school bullying intervention programs (Gaffney, Ttofi, & Farrington, 2018), the current investigation is underpowered, likely limiting the degree to which intervention effects could be accurately detected. In social norms research, there is evidence to suggest that change is greatest when normative feedback is tailored to the target group (Berkowitz, 2004). For an adolescent population, it could therefore be appropriate to deliver normative data via social media or apps or by incorporating interactive group discussions around the data’s credibility. Further, research has demonstrated that popularity norms, based upon the normative behavior of popular peers, drive individual behavior more strongly than broader classroom norms (Dijkstra et al., 2008). Thus, to improve relevancy of the feedback provided (Zieve et al., 2017), researchers may wish to examine whether current study findings are affected by the use of normative data comprising only popular students’ attitude ratings. An exploration of moderators, such as friendships, social standing, self-efficacy, and empathy (Poüyhoünen et al., 2010), could provide additionally meaningful information regarding characteristics linked to attitude change in response to corrective normative feedback.

Lastly, a third data collection point would strengthen the current study by allowing for three additional investigations. First, this would enable researchers to assess the stability of perceived peer attitude change. Second, researchers could gain insight into whether personal attitude change follows perceived peer attitude change over time, as newly adjusted perceived peer attitudes have the opportunity to stabilize as more comparative targets for personal beliefs. Third, this time point would make possible an examination of whether bullying and bystander behavioral change might follow any attitude change that occurs. Although the purpose of the present study was first to explore whether a social norms intervention could produce perceived peer and personal attitude change, significant findings regarding the former suggest that extending this investigation to include behavioral change could be a valuable area of future study. Ultimately, it is the degree to which this bullying and bystander behavioral change does or does not occur that will speak to the real-world, clinical utility of this intervention.

Conclusions

The current study offers some insight into a potentially effective bullying and bystander intervention that could feasibly be introduced into the school setting. The ability of normative feedback to move perceived peer attitudes toward bullying in the direction of the group norm has important clinical implications. For instance, increasing students’ awareness of the prosocial values held by their peers could improve feelings of safety and support within the school environment as well as encourage students to positively assert their expectation for the kind treatment of one another. Certainly, interventions that target bystander behavior have the potential to minimize both the occurrence and the negative impact of bullying by working toward a student body increasingly willing to trade in passive observation for the active defense of others.

Footnotes

Funding

The author(s) declared receipt of the following financial support for the research, authorship, and/or publication of this article: This research was partially funded by the University of Alabama Department of Psychology and the University of Alabama Graduate School.

ORCID iD

Casey E. Dillon

References

Andrews

P. W.

(2001). The psychology of social chess and the evolution of attribution mechanisms: Explaining the fundamental attribution error. Evolution and Human Behavior, 22, 11–29.

Bassili

J. N.

(1996). Meta-judgmental versus operative indexes of psychological attributes: The case of measures of attitude strength. Journal of Personality and Social Psychology, 71, 637–653.

Berkowitz

(2003). Applications of social norms theory to other health and social justice issues. In Perkins

H. W.

(Ed.), The social norms approach to preventing school and college age substance abuse: A handbook for educators, counselors, and clinicians (pp. 259–279). San Francisco, CA: Jossey-Bass.

Berkowitz

A. D.

(2004). An overview of the social norms approach. In Lederman

Stewart

(Eds.), Changing the culture of college drinking (pp. 193–214). Cresskill, NJ: Hampton Press.

Brendgen

Poulin

(2018). Continued bullying victimization from childhood to young adulthood: A longitudinal study of mediating and protective factors. Journal of Abnormal Child Psychology, 46, 27–39.

Celio

M. A.

Lisman

S. A.

(2014). Examining the efficacy of a personalized normative feedback intervention to reduce college student gambling. Journal of American College Health, 62, 154–164. doi:10.1080/07448481.2013.865626

Dijkstra

J. K.

Lindenberg

Veenstra

(2008). Beyond the class norm: Bullying behavior of popular adolescents and its relation to peer acceptance and rejection. Journal of Abnormal Child Psychology, 36, 1289–1299.

Epley

Keysar

Van Boven

Gilovich

(2004). Perspective taking as egocentric anchoring and adjustment. Journal of Personality and Social Psychology, 87, 327–339. doi:10.1037/0022-3514.87.3.327

Gaffney

Ttofi

M. M.

Farrington

D. P.

(2018). Evaluating the effectiveness of school-bullying prevention programs: An updated meta-analytical review. Aggression and Violent Behavior, 45, 111–133. doi:10.1016/j.avb.2018.07.001

10.

Gest

S. D.

Sesma

Masten

A. S.

Tellegen

(2006). Childhood peer reputation as a predictor of competence and symptoms 10 years later. Journal of Abnormal Child Psychology, 34, 509–526.

11.

Gremmen

M. C.

Berger

Ryan

A. M.

Steglich

C. E. G.

Veenstra

Dijkstra

J. K.

(2018). Adolescents’ friendships, academic achievement, and risk behaviors: Same-behavior and cross-behavior selection and influence processes. Child Development, 90, 192–211. doi:10.1111/cdev.13045

12.

Hawkins

Pepler

Craig

(2001). Naturalistic observations of peer interventions in bullying. Social Development, 10, 512–527.

13.

Higgins

E. T.

(1996). Knowledge activation: Accessibility, applicability, and salience. In Higgins

E. T.

Kruglanski

A. W.

(Eds.), Social psychology: Handbook of basic principles (pp. 133–168). New York, NY: Guilford Press.

14.

Hymel

Rocke-Henderson

Bonanno

R. A.

(2005). Moral disengagement: A framework for understanding bullying among adolescents. Journal of Social Sciences Special Issue, 8, 1–11.

15.

Hymel

Schonert-Reichl

K. A.

Bonanno

R. A.

Vaillancourt

Rocke Henderson

(2010). Bully and morality: Understanding how good kids can behave badly. In Jimerson

S. R.

Swearer

S. M.

Espelage

D. L.

(Eds.), Handbook of bullying in schools: An international perspective (pp. 101–118). New York, NY: Routledge.

16.

Juvonen

Galván

(2008). Peer contagion in involuntary social groups: Lessons from research on bullying. In Prinstein

M. J.

Dodge

K. A.

(Eds.), Understanding peer influence in children and adolescents (pp. 225–244). New York, NY: Guilford Press.

17.

Kaufman

T. M. L.

Kretschmer

Huitsing

Veenstra

(2018). Why does a universal anti-bullying program not help all children? Explaining persistent victimization during an intervention. Prevention Science, 19, 822–832. doi:10.1007/s11121-018-0906-5

18.

Lewis

M. A.

Litt

D. M.

Cronce

J. M.

Blayney

J. A.

Gilmore

A. K.

(2014). Underestimating protection and overestimating risk: Examining descriptive normative perceptions and their association with drinking and sexual behaviors. Journal of Sex Research, 51, 86–96.

19.

Miller

McFarland

(1991). When social comparison goes awry: The case of pluralistic ignorance. In Suls

Wills

T. A.

(Eds.), Social comparison: Contemporary theory and research (pp. 287–313). Hillsdale, NJ: Lawrence Erlbaum Associates Inc.

20.

Mussweiler

(2003). Comparison processes in social judgment: Mechanisms and consequences. Psychological Review, 110, 472.

21.

Neighbors

Larimer

M. E.

Lewis

M. A.

(2004). Targeting misperceptions of descriptive drinking norms: Efficacy of a computer-delivered personalized normative feedback intervention. Journal of Consulting and Clinical Psychology, 72, 434–447. doi:10.1037/0022-006X.72.3.434

22.

Oldenburg

Van Duijn

Veenstra

(2018). Defending one’s friends, not one’s enemies: A social network analysis of children’s defending, friendship, and dislike relationships using XPNet. PLoS One, 13, e0194323. doi:10.1371/journal.pone.0194323

23.

Olweus

(1993). Bullying at school: What we know and what we can do. Malden, MA: Blackwell Publishing.

24.

Peets

Pöyhönen

Juvonen

Salmivalli

(2015). Classroom norms of bullying alter the degree to which children defend in response to their affective empathy and power. Developmental Psychology, 51, 913–920. doi:10.1037/a0039287.

25.

Perkins

H. W.

(1997). College student misperceptions of alcohol and other drug use norms among peers: Exploring causes, consequences and implications for prevention programs. In Designing alcohol and other drug prevention programs in higher education: Bringing theory into practice (pp. 177–206). Newton, MA: The Higher Education Center for Alcohol and Other Drug Prevention.

26.

Perkins

H. W.

Berkowitz

(1986). Perceiving the community norms of alcohol use among students: Some research implications for campus alcohol education programming. The International Journal of the Addictions, 21, 961–976.

27.

Perkins

H. W.

Craig

D. W.

Perkins

J. M.

(2011). Using social norms to reduce bullying: A research intervention among adolescents in five middle schools. Group Processes & Intergroup Relations, 14, 703–722. doi:10.1177/1368430210398004

28.

Petrocelli

J. V.

Tormala

Z. L.

Rucker

D. D

. (2007). Unpacking attitude certainty: Attitude clarity and attitude correctness. Journal of Personality and Social Psychology, 92, 30–41.

29.

Poüyhoünen

Juvonen

Salmivalli

. (2010). What does it take to stand up for the victim of bullying? The interplay between personal and social factors. Merrill-Palmer Quarterly, 56, 143–163.

30.

Pozzoli

Gini

(2010). Active defending and passive bystanding behavior in bullying: The role of personal characteristics and perceived peer pressure. Journal of Abnormal Child Psychology, 38, 815–827.

31.

Prentice

D. A.

Miller

D. T.

(1993). Pluralistic ignorance and alcohol use on campus: Some consequences of misperceiving the social norm. Journal of Personality and Social Psychology, 64, 243–256.

32.

Raggatt

Wright

C. J. C.

Carrotte

Jenkinson

Mulgrew

Prichard

Lim

M. S. C

. (2018). “I aspire to look and feel healthy like the posts convey”: Engagement with fitness inspiration on social media and perceptions of its influence on health and wellbeing. BMC Public Health, 18, 1002. doi:10.1186/s12889-018-5930-7

33.

Reid

A. E.

Aiken

L. S.

(2013). Correcting injunctive norm misperceptions motivates behavior change: A randomized controlled sun protection intervention. Health Psychology, 32, 551–560. doi:10.1037/a0028140

34.

Rice

R. E.

Donohew

Clayton

(2003). Peer network, sensation seeking, and drug use among junior and senior high school students. Connections, 25, 5–19.

35.

Salmivalli

Lagerspetz

Björkqvist

Österman

Kaukiainen

(1996). Bullying as a group process: Participant roles and their relations to social status within the group. Aggressive Behavior, 22, 1–15.

36.

Salmivalli

Voeten

Poskiparta

(2011). Bystanders matter: Associations between reinforcing, defending, and the frequency of bullying behavior in classrooms. Journal of Clinical Child and Adolescent Psychology, 40, 668–676. doi:10.1080/15374416.2011.597090

37.

Sandstrom

M. J.

Bartini

(2010). Do perceptions of discrepancy between self and group norms contribute to peer harassment at school? Basic and Applied Social Psychology, 32, 217–225. doi:10.1080/01973533.2010.495645

38.

Sandstrom

Makover

Bartini

(2013). Social context of bullying: Do misperceptions of group norms influence children’s responses to witnessed episodes? Social Influence, 8, 196–215. doi:10.1080/15534510.2011.651302

39.

Schwartz

(2000). Subtypes of victims and aggressors in children’s peer groups. Journal of Abnormal Child Psychology, 28, 181–192.

40.

Shelton

J. N.

Richeson

J. A.

(2005). Intergroup contact and pluralistic ignorance. Journal of Personality & Social Psychology, 88, 91–107.

41.

Simon

J. B.

Nail

P. R.

(2013). Introduction to special issue on bullying: A social influence perspective. Social Influence, 8, 81–86.

42.

Sourander

Jensen

Ronning

Elonheimo

Niemela

Helenius

… Almqvist

. (2007). Childhood bullies and victims and their risk of criminality in late adolescence. Archives of Paediatric and Adolescent Medicine, 161, 546–552.

43.

Stevens

DeBourdeaudhuij

Van Oost

(2000). Bullying in Flemish schools: An evaluation of anti-bullying intervention in primary and secondary schools. British Journal of Educational Psychology, 70, 195–210.

44.

Tamir

D. I.

Mitchell

J. P.

(2013). Anchoring and adjustment during social inferences. Journal of Experimental Psychology, 142, 151–162. doi:10.1037/a0028232

45.

Veenstra

Dijkstra

J. K.

(2011). Transformations in adolescent peer networks. In Laursen

Collins

W. A.

(Eds.), Relationship pathways: From adolescence to young adulthood (pp. 135–154). Los Angeles, CA: Sage.

46.

Veenstra

Dijkstra

J. K.

Kreager

D. A.

(2018). Pathways, networks, and norms: A sociological perspective on peer research. In Bukowski

W. M.

Laursen

Rubin

K. H.

(Eds.), Handbook of peer interactions, relationships, and groups (2nd ed., pp. 45–63). New York, NY: Guilford.

47.

Willer

Kuwabara

Macy

(2009). The false enforcement of unpopular norms. American Journal of Sociology, 115, 451–490.

48.

Zieve

G. G.

Richardson

L. P.

Katzman

Spielvogle

Whitehouse

McCarty

C. A.

(2017). Adolescents’ perspectives on personalized E-feedback in the context of health risk behavior screening for primary care: Qualitative study. Journal of Medical Internet Research, 19, e261. doi:10.2196/jmir.7474