Extraversion as a Moderator of the Efficacy of Self-Esteem Maintenance Strategies

Abstract

Four experiments explored how extraversion’s connection with self-esteem may depend on specific self-enhancement strategies. Participants’ self-esteem threatening feedback indicating that they had performed poorly on a vocabulary or emotional intelligence test. In Experiment 1, participants (N = 80) were randomly assigned to either a control condition (no self-enhancement) or a downward social comparison condition. The procedures for Experiments 2 (N = 470) and 3 (N = 514) were similar, adding a self-serving attribution condition (Experiments 2 and 3) and Basking-in-Reflected-Glory (BIRG) condition (Experiment 3). Across the experiments, extraversion was more related to self-esteem under downward social comparison versus other conditions. BIRGing produced higher self-esteem in Experiment 3 across extraversion levels. Experiment 4 (N = 355) focused on downward social comparison versus control, and provided evidence that an increased perception of being similar to the comparison targets may partially explain extraversion’s self-esteem link. Theoretical implications concerning both extraversion and self-enhancement are discussed.

Keywords

self-esteem extraversion self-enhancement social comparison

Since its original conceptualization in the 1920s (Jung, 1923/1971), extraversion–introversion has been an important construct. The terms introvert and extravert are popularly used among laypeople and definitions of the terms are in any English-language dictionary. Likewise, extraversion is an important dimension of many trait taxonomies (e.g., Eysenck, 1967, 1990; McCrae & Costa, 1987) and assessment instruments (e.g., neuroticism, extraversion, openness [NEO]-Personality Inventory, Costa, & McCrae, 1992). Thus, there is substantial consensus that extraversion is a fundamental personality dimension.

Extraversion is defined by increased sociability, the degree to which an individual actively engages in the social world (e.g., Ashton et al., 2002; Costa & McCrae, 1992; Lucas et al., 2008). Relative to introverts, extraverts tend to both enjoy more, and be more competent in social situations (Ashton et al., 2002). Extraversion has been associated with various behavioral and emotional outcomes in interpersonal contexts. For instance, relationship, job, and overall life satisfaction are higher among extraverts than introverts (Malouff et al., 2010; Scollon & Diener, 2006; Lee et al., 2008), and extraverts tend to be more popular and likable than introverts (Van der Linden et al., 2010). Furthermore, compared with introverts, extraverts perceive and experience more social support (Marigold et al., 2014; Swickert et al., 2002, 2004), and benefit more from social facilitation effects (Uziel, 2007). Therefore, in many interpersonal contexts extraverts seem more content and more successful than introverts.

Although the psychological and behavioral correlates of extraversion are well-known, one important but understudied domain is the degree to which extraversion is associated with differences in how people regulate their self-evaluations. This is important because the ways people react to and understand the world around them are influenced by the processes by which they evaluate and understand themselves (Markus, 1977; Murray et al., 2002; Tesser, 1988). Several motivations influence these self-evaluative processes, including self-enhancement, self-verification, self-assessment, and self-improvement (Sedikides & Strube, 1997). Perhaps the most widely researched of these motives is self-enhancement. Indeed, Sedikides and Strube (1997) identified self-enhancement as the most fundamental of the self-evaluative motives. Self-enhancement is defined as a drive to elevate the positivity of one’s self-views by seeking positive and/or avoiding negative self-relevant information (e.g., Sedikides, 1993; Sedikides et al., 2003; Taylor & Brown, 1988).

A positive association between extraversion and self-esteem has been clearly established (e.g., Francis, 1997; Kawash, 1982; Schmitt & Allik, 2005; Swickert et al., 2004). However, the psychological processes linking extraversion and self-esteem regulation are less understood. Self-esteem regulation processes may provide at least three explanations. First, high self-esteem individuals may be more active in social behaviors, for example, because they are more self-assured that they will be accepted by others. Second, increased extraversion may be associated with stronger self-enhancement motivation. Third, extraverts might be more motivated or proficient in implementing some self-esteem enhancing strategies. We suspect productive research may validate each of these links between extraversion and self-esteem, but in the present experiments we focus on the third possibility.

Psychologists have identified numerous self-enhancement strategies. For example, people can self-enhance through self-serving attributions (i.e., taking personal credit for successes and blaming failures on external factors; Campbell & Sedikides, 1999; Larson, 1977). Likewise, people Basking-in-Reflected-Glory (BIRG) is the tendency to associate with successful others in order that their successes will be linked to oneself (e.g., Cialdini et al., 1976; Miller, 2009). People also self-enhance via downward social comparison: the tendency to evaluate one’s abilities, performance, or circumstances in contrast to those of individuals less successful than oneself (e.g., Wills, 1981; Wood et al., 1985).

Extraverts might be more likely to generate self-esteem when they engage in some of these strategies over others. More specifically, some self-enhancement strategies are clearly more socially oriented than others. For example, downward social comparison necessarily involves contrasting one’s own outcomes with those of a less successful person or group, making this strategy inherently social. In contrast, other strategies including self-serving attribution do not require a social reference point. Some self-serving attributions may have a social focus (e.g., blaming other people for a poor performance), but others do not (e.g., attributing failure to fatigue), so self-serving attributions are not inherently social. Extraversion might be more related to self-esteem when people engage in social self-enhancement strategies, and less strongly related in the context of engaging in nonsocial self-enhancement strategies (or no self-enhancement at all). In four experiments, following a self-esteem threat, participants were given a chance to bolster their self-esteem using inherently social (vs. nonsocial) forms of self-enhancement. We expected that extraversion would be more related to self-esteem during socially oriented self-enhancement.

Experiment 1

Experiment 1 provided an initial test of our hypothesis that extraverts benefit more from social forms of self-enhancement.

Method

Participants

A sample of 80 participants (62 females, 17 males, one unspecified) participated in this experiment in return for partial course credit or a CAN$5 payment.¹ These participants were selected from a larger sample of introductory psychology students who completed the NEO Five-Factor Inventory (NEO-FFI; Costa & McCrae, 1992) as part of a test battery administered several weeks before completing the laboratory portion of the experiment. Only those who scored in the top (M = 37.43, SD = 3.37) or bottom third (M = 24.21, SD = 4.00) of the prescreening sample on the NEO-FFI extraversion subscale (conceptual range: 0–48) were recruited.

Procedure

This experiment was a 2 (Extraversion: High vs. Low) × 2 (Self-Esteem Maintenance Condition: Control vs. Downward Comparison) between-participants design. Participants sat at individual computers, and were told that they would complete a multiple-choice vocabulary test assessing verbal intelligence. After being informed of the predictive value of verbal intelligence for academic, relationship, and occupational success (i.e., issues of likely importance to the self-concepts of university students), participants answered a 20-item vocabulary test composed of difficult words (e.g., alacrity, lutulent, and abstemious). These words were pretested to ensure that participants were familiar with them, but unlikely to know their definitions. Thus, participants might expect to correctly answer some words, but also to believe negative performance feedback. Afterwards, all participants received negative false feedback: 30%. Our primary goal was to threaten/challenge self-esteem to make self-enhancement behaviors more likely (Nussbaum & Dweck, 2008; Tesser, 2000; Wills, 1981).

Participants were randomly assigned to receive one of two types of “additional information” about the test: downward social comparison versus control. Those assigned to the downward comparison condition were informed that the average score of university students was 20%. Those assigned to the control condition read innocuous historical information about the test, providing minimal opportunity for self-enhancement. Participants then completed a State Self-Esteem Scale (SSES; Heatherton & Polivy, 1991).² Participants were fully debriefed and dismissed.

Measures

Verbatim materials appear in Supplemental Materials A and B (see https://osf.io/zs2h9/).

Extraversion

Extraversion was assessed using the extraversion subscale of the NEO-FFI (Costa & McCrae, 1992). This subscale consists of 12 statements rated on 5-point scales, with response options ranging from Strongly disagree (0) to Strongly agree (4). Negatively keyed items were reverse-scored and item responses were summed (α = .85). Possible scores ranged from 0 to 48, higher numbers reflecting increased extraversion.

State self-esteem

We used the SSES (Heatherton & Polivy, 1991) to detect situational variations in self-esteem. This scale consists of a total of 20 items rated on 5-point scales from Not at all (1) to Extremely (5). Although the SSES is three-dimensional, our interest was participants’ overall sense of self-worth so we averaged all items to assess global state self-esteem. Negatively keyed items were reverse scored (α = .90). This resulted in a possible score ranging from 1 to 5, higher numbers reflecting higher self-esteem.

Results and Discussion

Participants’ average self-esteem scores were submitted to a 2 (Extraversion: High vs. Low) × 2 (Self-Esteem Maintenance Condition: Control vs. Downward Comparison) analysis of variance (ANOVA). We found an expected main effect of extraversion, with extraverts (M = 3.60, SE = .09) reporting higher state self-esteem than introverts (M = 3.22, SE = .10), F(1, 76) = 8.21, p = .005, η² = .10, CI₉₀ = [.02, .21]. There was no main effect of downward social comparison (M = 3.42, SE = .10) compared with control (M = 3.40, SE = .08), F(1, 76) = .03, p = .874, η² = .0004, CI₉₀ = [.00, .02].

Importantly, we found an interaction between extraversion level and self-enhancement condition, F(1, 76) = 5.84, p = .018, η² = .07, CI₉₀ = [.01, .18]. In the control condition, extraverts (M = 3.43, SE = .12) did not differ in self-esteem compared with introverts (M = 3.37, SE = .12), t(76) = .35, p = .729, r _contrast = .04,³ shown by the leftmost two bars in Figure 1. As expected, in the downward comparison condition, extraverts (M = 3.76, SE = .13) reported higher self-esteem than did introverts (M = 3.08, SE = .15), t(76) = 3.46, p = .001, r _contrast = .37, as is clear from comparing the rightmost bars of Figure 1. Thus, extraversion only related to self-esteem when participants compared their performance favorably against others.⁴

Figure 1.

Global state self-esteem scores (mean) for introverts and extraverts after receiving control or downward social comparison information (Experiment 1).

Figure 2.

Global state self-esteem scores for introverts and extraverts after receiving control, downward social comparison, or self-serving attribution information (Experiment 2).

Experiment 2

Although Experiment 1 generally supported our key predictions, it did not rule out whether extraverts might extract larger self-esteem benefits from any self-enhancement strategy, social or otherwise. Experiment 2 therefore added a nonsocial self-enhancement condition: self-serving attribution. We expected that introverts and extraverts would react similarly to this strategy.

A second issue with Experiment 1 was its modest sample size. Given the effect size obtained in Experiment 1 for our key contrast, 470 participants gave us 99.8% power to replicate. The larger sample size also facilitated adding another experimental condition.

Method

Participants

A sample of 470 psychology undergraduates (121 males, 342 females, seven prefer not to answer; M _age = 18.2, SD _age = 1.3) participated for partial course credit or CAN$5. Again, we recruited participants from a prescreening sample several weeks earlier, who scored in the top (M = 37.69, SD = 2.96) or bottom third (M = 24.87, SD = 4.26) on extraversion (α = .80).

Design and procedure

This experiment was a 2 (Extraversion: High vs. Low) × 3 (Self-Esteem Maintenance Condition: Control vs. Downward Comparison vs. Self-Serving Attribution) between-participants design. The test and feedback were identical to Experiment 1. After receiving their score, participants were randomly assigned to one of three conditions: control, downward social comparison (both as per Experiment 1), or self-serving attribution. Individuals assigned to the self-serving attribution condition were informed that the test was potentially contaminated by variables like fatigue and stress (factors which we reasoned participants could infer might have undermined their performance). Participants completed the SSES (α = .92) and were debriefed.

Results and Discussion

Participants’ self-esteem scores were submitted to a 2 (Extraversion: Introverts vs. Extraverts) × 3 (Self-Esteem Maintenance Condition: Control vs. Downward Social Comparison vs. Self-Serving Attribution) ANOVA. A main effect of self-esteem maintenance condition revealed that downward social comparison (M = 3.60, SE = .05) resulted in higher state self-esteem scores than control information (M = 3.40, SE = .05) or self-serving attribution (M = 3.46, SE = .05), F(2, 464) = 4.06, p = .018, η² = .01, CI₉₀ = [.0001, .028]. We performed planned contrasts to compare downward social comparison against control, and self-serving attribution versus control. Downward social comparison produced higher state self-esteem than control, contrast = .19, SE = .07, t(464) = 2.74, p = .006, r _contrast = .13, whereas self-serving attribution did not produce higher self-esteem than control, contrast = .05, SE = .07, t(464) = .74, p = .461, r _contrast = .03. Replicating Experiment 1, we found a main effect of extraversion: extraverts (M = 3.56, SE = .04) reported higher self-esteem than introverts (M = 3.43, SE = .04), F(1, 464) = 4.77, p = .030, η² = .01, CI₉₀ = [.001, .031].

These main effects were qualified by a marginal interaction between extraversion and feedback type, F(2, 464) = 2.92, p = .055, η² = .01, CI₉₀ = [.00, .032]. In the control condition (left two bars), extraverts (M = 3.47, SE = .07) had similar self-esteem to introverts (M = 3.35, SE = .07), t(464) = 1.20, p = .232, r _contrast = .06. In the downward social comparison condition (middle two bars), extraverts (M = 3.75, SE = .07) had higher self-esteem than introverts (M = 3.45, SE = .07), t(464) = 3.08, p = .002, r _contrast = .14. Importantly, we also noted that extraverts (M = 3.44, SE = .08) had similar self-esteem to introverts (M = 3.48, SE = .07) in the self-serving attribution condition (right most bars), t(464) = −.40, p = .687, r _contrast = −.02.

Experiment 2 generally replicated Experiment 1: extraversion was positively related to self-esteem when people engaged in downward social comparison, but was unrelated to self-esteem when no specific form of self-esteem enhancement was facilitated. Interestingly, extraversion was also unrelated to self-esteem in the self-serving attribution condition. Importantly, this lack of an association in the self-serving attribution condition cannot be explained by both extraverts and introverts benefiting from this strategy because there was no evidence that self-serving attributions produced greater self-esteem for either group compared with the control condition. Although some caution regarding these findings is advisable given our omnibus interaction fell just short of statistical significance (p = .055), a critical contrast corresponding to our replication of Experiment 1 (comparing introverts and extraverts within downward social comparison) was significant (p = .002). That is, introverts and extraverts only differed in their self-esteem within the downward social comparison condition.

Experiment 3

Experiment 3 had several goals. First, we wanted to rule out the alternative interpretation that the increased relationship between extraversion and self-esteem seen under downward social comparison is merely a function of self-enhancement strategies in general. Self-serving attributions in Experiment 2 do not rule this possibility out because self-serving attributions failed to produce any self-esteem effects. Thus, we attempted to bolster our self-serving attribution manipulation such that it should increase self-esteem. In Experiment 2, we had assumed that informing participants about the many common life factors that could undermine their performance would give them sufficient information to perform self-serving attribution. However, a more directive and personalized approach was reasoned to be more powerful, and used in Experiment 3.

Second, although we predicted that extraversion should be more related to self-esteem under any socially based self-esteem strategy, the prior experiments only explored this hypothesis using downward social comparison. We therefore targeted another socially based self-esteem strategy: Basking-in-Reflected-Glory (BIRGing; Cialdini et al., 1976), representing people’s ability to self-enhance via perceiving their association with successful groups.

Third, we decided to consider another alternative explanation: if extraverts expected a favorable score more than did introverts, extraverts might have gained more self-esteem from downward social comparison simply because they felt more surprised by the feedback. We therefore assessed participants’ test performance expectations before receiving feedback.

Fourth, we added suspicion probes at the end of Experiment 3. Suspicion is a potentially important aspect in false feedback studies, because if false feedback is not believed then participants may be less likely to engage in self-esteem repair.

Finally, we sought to generalize our findings. For example, we shifted from false feedback concerning verbal intelligence to false feedback concerning emotional intelligence. An advantage of this alternative manipulation is that undergraduate samples are likely to have received much genuine feedback about their verbal abilities, possibly making them resistant to false feedback in this domain. In contrast, it is unlikely that students will have received direct evaluation of their emotional intelligence. Furthermore, to generalize to new stimuli, we substituted the Big Five Inventory (BFI-44; John et al., 1991) to assess extraversion. Furthermore, Experiments 1 and 2 utilized an extreme groups paradigm in which we preselected clear introverts and extraverts to highlight potential their differences. Here, we sampled across the full, continuous range of extraversion scores among our undergraduates. Jointly, these changes aimed to show that our phenomenon was robust across procedural, measurement, and sampling variations.

Method

Participants and procedure

Participants were 514 undergraduate students participating for partial course credit.⁵ Given the average effect size of our contrast effect across Studies 1 and 2, 514 participants (based on time-based stopping) gave us 97.3% power to find our key effect.

All participants were told that they would be complete an emotional intelligence quotient (EQ) test and would receive a score. They were told that EQ has major implications for people’s relationship, career, and health outcomes. Participants completed four EQ subtests, adapted from the Mayer–Salovey–Caruso Emotional Intelligence Test (Mayer et al., 2003): understanding emotions (guessing how a target would feel in various circumstances); using emotions (selecting “helpful” emotions for a situation); managing emotions (estimating how a scenario character would go about “preserving” their mood after an event); and identifying emotions (rating the emotions displayed by a face presented for 500 ms). Items were “rigged” to have ambiguous answers, so that negative feedback would be believable. Afterwards, participants were asked to estimate their score. After a brief “calculating” screen all participants were given a 23% score. To make this value meaningful, participants were provided with a table that provided the likely life outcomes of various EQ levels. A 23% score ostensibly placed participants in a “mildly maladaptive” category characterized as having negative implications for their professional development, stress coping, and life satisfaction. Next, participants were randomly assigned to one of four different conditions.

Control condition participants learned generic information about EQ test’s historical development.

BIRG condition participants listed the geographical region that they identified with most (e.g., Canadian provinces like Alberta or Ontario). Next, they were asked to generate several positive thoughts that they had about their region.⁶ Participants were exposed to a series of headlines, supposedly from magazines and research organizations, about their region (e.g., “Canadians Hailing from [region name] More Likely to Donate to Health Drives,” credited to a well-known news magazine). For each, participants were to rate the headline’s valence from Strongly negative (1) to Strongly positive (5). Participants rated these headlines as highly positive (M = 4.35, SD = .65).

Downward social comparison participants were told that they would next view the scores of several previous test-takers, which were fixed to be 15% and 18%.

Finally, the self-serving attribution condition began as in Experiment 2. However, participants answered a questionnaire about their recent habits, and the computer ostensibly re-calculated their EQ score accordingly. They answered a series of ten questions with yes/no response options. The goal of these questions was to present participants with “risk factors” that would be applicable to most participants. All participants were told that their “true score” would have been between 44% and 63%, correcting for these external factors.

All participants next completed the SSES (α = .92), intermixed with 15 filler items to reduce suspicion, and then the BFI-44, before completing the suspicion items, and being thoroughly debriefed.

Measures

Extraversion was measured using the eight-item extraversion subscale of the BFI-44 (John et al., 1991). Participants completed all five subscales but we only analyzed extraversion (α = .86). Items were rated from Disagree strongly (1) to Agree strongly (5). Scores were averaged, with higher scores indicating more extraversion.

Suspicion was assessed using a four-item funnel interview. Questions were ordered along a gradient from innocuous (i.e., “What do you think the purposes of the study might have been?” Q1) to blatant (i.e., “Did you at any time doubt that the testing you performed was not real, or that the results were suspicious?…” Q4). Thus, only Q1 and Q2 were considered likely to be diagnostic of participants’ pre-existing suspicion, whereas Q3 and Q4 determined the extent to which the experimental manipulations were resistant to overt questioning. Participants were coded as suspicious if they expressed awareness that the false feedback was untrue or that we had manipulated their self-esteem.

Results and Discussion

Performance expectancy

Participants expected to obtain scores on the EQ task (M = 64.1%, SD = 14.8) that were much higher than the false feedback value of 23%, as confirmed by a one-sample t test, t(501) = 62.4, p < .001, d = 2.78. These expectations did not vary as a function of extraversion, r(500) = .06, p = .159, making differential expectations an unlikely explanation of our findings.

Suspicion probes

Scores on the four suspicion items were all moderately related, .21 > rs(512) > .66, all ps < .001. Only 9% of participants expressed suspicious responses on the first probe, with a cumulative increase to 11% by the second, 22% by the third, and 43% by the fourth. Thus, the vast majority of participants expressed no suspicion on initial suspicion probes, and a majority mentioned no suspicion despite repeated, pointed questioning. Main analyses represent the sample size with participants who expressed suspicion on either/both of the first two items removed (n = 457); analyses with all participants included (n = 514) and analyses with participants who were suspicious on just the first suspicion item (n = 467) are footnoted.

Cumulative suspicion by the second probe did not differ by experimental condition, F(3, 510) = 1.22, p = .303, and correlated weakly with extraversion, r(512) = .09, p = .039.

Main analyses

State self-esteem was regressed on extraversion (a centered, continuous variable), dummy variables for the three noncontrol conditions (BIRG, downward social comparison, self-serving attribution; each dummy variable was centered), and two-way combinations of Extraversion × Condition Variables. Because dummy variables for each noncontrol condition were entered simultaneously, each dummy variable represents the unique effect of a condition relative to control. The level of multicollinearity was acceptable (tolerance > .660, variance inflation factor [VIF] < 1.52).

Figure 3 demonstrates the results. Lines trace self-esteem (on the y-axis) as predicted by extraversion, ranging from low (−1 SD) on the left to high (+1 SD) on the right, across the four experimental conditions.⁷ First, we once again found a main effect of extraversion, β = .36 [.28, .45], t(448) = 8.30, p < .001, where the positive coefficient indicates that extraverted individuals had higher self-esteem than introverts. This is clearly indicated by the generally positive slopes of all lines running from low to high extraversion.

Figure 3.
Global state self-esteem scores across levels of extraversion after receiving control, downward social comparison, self-serving attribution, or BIRGing information (Experiment 3).

We then track the effect of extraversion on self-esteem across the four experimental conditions. To provide a baseline of comparison, the light gray, dotted line at the bottom of the figure represents the self-esteem of individuals in the control group (across levels of extraversion). We would anticipate that these individuals should have the lowest self-esteem because they have undergone a self-esteem threat without any self-enhancing information; indeed, this line consistently indicates the lowest self-esteem across conditions.

We next examine condition main effects. We note that the BIRG (gray/dashed-dots) line is consistently above the control line, indicating slightly greater self-esteem compared with control, β = .11 [.01, .22], t(448) = 2.08, p = .038. Downward social comparison (solid/black line), β = .08 [−.02, .19], t(448) = 1.58, p = .114, and self-serving attributions (dark gray/dashed line), β = .04 [−.06, .15], t(448) = .78, p = .438, both showed no overall self-esteem advantage versus control. Thus, even our revised self-serving attribution condition failed to bolster participants’ self-esteem.

Next, we examine our interaction effects. Critically, as hypothesized, extraversion interacted significantly with downward social comparison, β = .11 [.004, .21], t(448) = 2.05, p = .041. This is the third demonstration of this effect, with a sample of individuals who were not preselected as extreme introverts/extraverts, and with an alternative extraversion measure. Compared with the relatively weak control slope of extraversion on self-esteem, β = .29 [.12, .45], t(452) = 3.37, p < .001, the downward social comparison condition slope was more pronounced, β = .53 [.37, .69], t(452) = 6.41, p < .001.

Finally, we examine whether extraversion was more highly linked with self-esteem in any other self-esteem maintenance condition

BIRGing did not interact with extraversion, β = .03 [−.08, .13], t(448) = .51, p = .609, contrary to the idea that any social self-esteem maintenance strategy would increase the extraversion/self-esteem link. Furthermore, there was no interaction with self-serving attribution, β = −.002 [−.10, .10], t(448) = −.03, p = .973, matching Experiment 2. These nonsignificant interactions make sense when one visually examines the simple extraversion slopes in Figure 3: the BIRG condition, β = .35 [.18, .52], t(452) = 3.94, p < .001, and the self-serving attribution condition, β = .28 [.10, .47], t(452) = 3.02, p = .003, each had simple slopes roughly parallel to control’s extraversion slope-line. In summary, when participants engaged in BIRGing or self-serving attribution, their extraversion was not especially related to self-esteem (with slopes highly similar to, and not significantly different from, the control group). In conjunction with BIRGing’s absence of an interaction with extraversion, its significant main effect on self-esteem suggests that introverts did obtain self-esteem benefits from BIRGing, revealing that introverts can be shown to engage in self-enhancement.⁸

Summary

This experiment establishes several important developments. First, we once again replicated the Extraversion × Downward Social Comparison Effect, and demonstrated its robustness with our procedural variations (different measure, no extreme group selection, alternative self-esteem threat). Second, by showing a main effect of BIRGing without moderation by extraversion, we identified a condition wherein introverts performed self-enhancement. In addition, although BIRGing is understood to be a self-esteem-related behavior (Cialdini et al., 1976), we are unaware of any prior direct demonstration that BIRGing increases self-esteem. Furthermore, this pattern of effects suggests that extraversion’s link with self-esteem is magnified under downward social comparison specifically, but not increased when engaging in a second self-esteem strategy that is also social: BIRGing. We also found that suspicion was quite low for the nonleading suspicion probes, and that inclusion/exclusion of suspicious participants did not appreciably change our primary results.

Experiment 4

Experiments 1–3 revealed a robust tendency for extraversion to be more related to self-esteem during downward social comparison. However, we have not yet identified what psychological mechanism(s) drive this effect. That is, what qualities of extraverts lead them to obtain larger self-esteem boosts from downward social comparison? We therefore designed Experiment 4 to focus on this question, consequently including only the control and downward social comparison conditions to maximize statistical power on the critical contrast and its examined mechanism(s).

We considered four mechanisms. First, extraverts may be more inclined to see other people as similar to themselves, with similarity forming a prerequisite for meaningful social comparison (Byrne, 1971; Festinger, 1954; Tesser, 1988). Due to their increased tendency to socialize with other people (Lucas et al., 2008), extraverts might see other people as “like me,” facilitating meaningful social comparisons. Previous scholarship suggests that people can “unmake” social comparisons they perceive as illegitimate (Gilbert et al., 1995), so introverts seeing others as dissimilar (and therefore illegitimate comparison targets) might explain our prior experiments.

Second, extraversion might predict self-esteem more under downward social comparison because extraverts are more competitive than introverts (Ross et al., 2003). Specifically, personality researchers have suggested that extraverts engage in more personal development competitiveness (Ross et al., 2003), defined as discovering one’s abilities and forging connections through competition (Ryckman et al., 1990, 1994).

Third, because of their frequent social activity (Lucas et al., 2008), extraverts may naturally use other people’s performance to evaluate the diagnostic value of tests. In the present context, extraverts might obtain larger self-esteem rewards from learning that other people performed poorly on the test, more readily concluding that a test which fails everybody need not challenge their self-worth.

Fourth, extraverts might have a greater tendency to see social relationships as core to their self-definition/identity. For example, extraverts are more inclined to feel closely connected with others, forming deeper personal bonds (Lee et al., 2008; Wu et al., 2008). If so, favorable or unfavorable comparisons with those other people might result in larger shifts in self-evaluation.

Method

Participants and procedures

Given that Experiments 1–3 provided an average effect size of r = .135 for our critical interaction, our recruitment of 355 participants (based on a time-based stopping rule) provided us 72% power to test our hypotheses. The procedure of Experiment 4 mirrored Experiment 3, but we included only control and downward social comparison conditions. However, we added several mediator measures at the end of the procedure. This timing was intended to provide a clean replication of a subset of Experiment 3 while probing mechanisms. The four mediator measures were presented in randomized order, with items randomized within each scale.

Measures

Peer-group similarity

We included four items to probe whether participants felt their peers were similar to themselves (i.e., appropriate comparison targets). Each item was rated from 1 (Not at all) to 9 (Extremely). Internal consistency was high, α = .92.

Personal development competitiveness

We used Ryckman et al.’s (1990) scales of competitiveness. The personal development motivation represents belief that competition bolsters one’s capabilities. Each item is rated from 1 (Never true of me) to 5 (Always true of me). Internal consistency was high, α = .81.⁹

Perceived test validity

We examined perceptions of the EQ test’s validity with three items. Each item was rated from 1 (Not at all accurate) to 9 (Extremely accurate). Internal consistency was high, α = .89.

Relevance of others to self

We used four items to assess participants’ belief that other people were central to their self-identity, defining who they were. Items were rated on 9-point scales from 1 (Strongly disagree) to 9 (Strongly agree). Internal consistency was high, α = .83.

Results and Discussion

Performance expectancy

As in Experiment 3, participants expected to obtain scores on the EQ task (M = 64.7%, SD = 13.9%) that were much higher than the false feedback value of 23%, t(313) = 53.32, p < .001, d = 3.00. Extraversion was only weakly related to performance expectancy, r(312) = .17, p = .002, again making this an unlikely explanation for our data.

Suspicion probes

Scores on the four suspicion items were interrelated, .25 > rs(353) > .13, all ps < .02; other than Item 1 with 3, r(353) = .05, p = .348. Only 5.4% of participants expressed suspicious responses on the first probe. The second (8.7%), third (19.7%), and fourth (40.6%) items raised cumulative suspicion. Main analyses represent the sample size with participants who expressed suspicion on one/both of the first items removed (N = 324); analyses with all participants included (N = 355) and analyses with participants who were suspicious by the first suspicion item (N = 336) are footnoted. Suspicious responses to the first two probes did not differ by experimental condition, t(353) = .20, p = .843, and did not correlate significantly with extraversion scores, r(353) = −.07, p = .165.

Conditional self-esteem effects

We contrast-coded condition (control = −.5, downward social comparison = +.5) and centered extraversion scores around zero before forming their interaction term. We used moderated regression to determine whether extraversion benefited self-esteem more in the downward social comparison condition (versus control) condition. We once again found a main effect of extraversion, β = .24 [.14, .35], t(320) = 4.60, p < .001, where the positive coefficient indicates that extraverted individuals had higher self-esteem than introverts, collapsing across conditions. We found no significant main effect of downward social comparison, β = .00 [−.11, .11], t(320) = .01, p = .990.

Figure 4 demonstrates the patterns of extraversion and the experimental conditions predicting state self-esteem. The light gray, dotted line at the bottom represents the self-esteem of individuals in the control group (across levels of extraversion). The black, solid line represents the self-esteem of people in the downward social comparison group. Note that the slope lines do not appear to run parallel, as will be examined in our interaction effect test. Once again, the Extraversion × Downward Social Comparison Interaction effect emerged significantly, β = .11 [.01, .22], t(320) = 2.13, p = .034. In the control condition, extraversion is not significantly related to higher self-esteem, meaning that extraverts felt no better about themselves than did introverts, β = .13 [−.02, .28], t(320) = 1.74, p = .082. In the downward social comparison condition, however, extraversion was powerfully linked with higher self-esteem, β = .36 [.21, .51], t(320) = 4.77, p < .001. Thus, downward social comparison magnified extraversion’s self-esteem association, replicating Experiments 1–3.

Figure 4.
Global state self-esteem scores across levels of extraversion after receiving control or downward social comparison information (Experiment 4).

Psychological mechanism(s)

We next evaluated our four proposed mechanisms of the Extraversion × Downward Social Comparison Interaction. We conducted a series of empirical tests for these mediators of our moderation effect. (a) To serve as a mechanism for extraversion, the variable should correlate with extraversion. (b) The mechanism should show a similar pattern of moderation when substituted in extraversion’s place. (c) When the mechanism’s interactive term is combined in the same regression model with extraversion’s interaction term, the latter term should become nonsignificant because both interactions should be at least partially overlapping (i.e., statistically redundant) effects. (d) Finally, a mediation analysis should reveal that the mechanism variable accounts for extraversion’s moderating role. We evaluated our mechanisms using these tests, dropping variables when they failed to meet a given criterion (see Table 1).

Table 1.
Evaluation of Four Possible Mediators of the Conditional Extraversion/Self-Esteem Association (Experiment 4).

Possible Mediator Step 1: Association with Extraversion Step 2: Substitutes as Moderator Step 3: Reduces Extraversion’s Moderating Role Step 4: Mediation Analysis

Similarity Yes
r(322) = .29
p < .001
Yes
β = .10 [−.004, .21]
p = .058 Somewhat: Extraversion interaction weakened to β = .09 [−.02, .20], t(318) = 1.68, p = .094 Somewhat: Pattern is as expected, but formal difference test is χ²−Δ(1) = 2.00, p = .157

Personal development competitiveness Yes
r(322) = .28
p < .001 No
β = .04 [−.07, .14]
p = .478
— —

Perceived test validity Yes
r(322) = −.16
p = .005
No
β = −.006 [−.11, .10]
p = .909
— —

Relevance of other to self/identity No
r(322) = −.01
p = .909 — — —

Step 1: Association with extraversion

We found that extraversion correlated with an increased perception of others’ similarity to oneself, increased personal development competitiveness, and decreased perceptions of the test’s validity. Thus, we dismissed self/other identity as a mechanism.

Step 2: Substitution as moderator

We centered each of the three viable mechanisms, regressing self-esteem on their main and two-way interaction effects (with downward social comparison). Three regression tests were performed; one for each variable that passed Step 1. We found a marginal interaction of Similarity × Downward Social Comparison, plotted as Figure 5. Similarity was unrelated to self-esteem in the control condition, β = .11 [−.04, .25], t(320) = 1.42, p = .156, reflected in the gray/dotted line’s modest slope. In contrast, similarity had a much more pronounced association with self-esteem in the downward social comparison condition, β = .31 [.16, .47], t(320) = 3.91, p = .0001, reflected in the black/solid slope line. Comparing Figures 4 and 5, the interaction is extremely similar to the Extraversion × Downward Social Comparison Interaction. Neither personal development competitiveness nor perceived test validity produced interaction effects predicting self-esteem, although personal development competitiveness increased, β = .30 [.20, .41], t(320) = 5.59, p < .001, and perceived test validity decreased, β = −.16 [−.27, −.05], t(320) = −2.89, p = .004, self-esteem.

Figure 5.
Global state self-esteem scores across levels of perceived similarity after receiving control or downward social comparison information (Experiment 4).

Step 3: Reduction of extraversion’s moderation

Given Steps 1 and 2, only similarity remained. We regressed self-esteem onto the main effects of extraversion, similarity, and downward social comparison, and the interaction terms of both extraversion and similarity with downward social comparison. We examined whether the Extraversion × Downward Social Comparison term was reduced to nonsignificance with Similarity × Downward Social Comparison added to the model. Indeed, Extraversion × Downward Social Comparison was reduced to nonsignificance, as was the Similarity × Downward Social Comparison term, β = .07 [−.04, .18], t(318) = 1.27, p = .204. Although the shift in extraversion’s interactive role was modest, its decrement in magnitude is consistent with the idea that extraversion and similarity’s interactive roles overlap.

Step 4: Mediation analysis

In Step 4 we conducted multi-group structural equation modeling. Statisticians have debated how to implement mediated moderation (Edwards & Lambert, 2007; Muller et al., 2005), with most concluding that the test is mathematically similar to moderated mediation. Therefore, we evaluated a model in which extraversion predicted self-esteem via similarity, with experimental condition moderating the link between similarity and self-esteem. Additional details on these models are noted in Supplemental Materials D.

To do this, we conducted a multi-sample structural equation analysis in which we estimated our mediational model separately in the two social comparison conditions. This mediational model included an extraversion latent variable (eight item indicators), a similarity latent variable (four item indicators), and a state self-esteem latent variable (three subscale score indicators: performance, appearance, and social). Extraversion was specified to have a direct effect on similarity which was in turn specified to have a direct effect on state self-esteem. A direct path from extraversion to self-esteem was also included. After fitting this model in both conditions, we next assessed whether the measurement model was invariant across conditions. We did this by constraining each factor loading to be equivalent across conditions. A chi-square test between models indicated that these constraints did not significantly alter the fit of the model thereby indicating measurement invariance, χ²−Δ(12) = 6.88, p = .866. This invariant model was used for subsequent analyses.

The paths among latent variables across each condition are reported in Table 2. As expected, our analysis indicated that in both conditions extraversion significantly predicted similarity (Row 2). However, as indicated in Row 3, the similarity to self-esteem path did suggest differences across conditions. Specifically, this path was not significant in the control condition, but was significant in the social comparison condition, although a test of the difference between these coefficients was not significant. Most importantly, we found that the indirect mediational path from extraversion to self-esteem via similarity was significant in the downward social comparison condition, but was nonsignificant in the control condition (Row 4). This pattern follows our reasoning: extraversion should be related to similarity in both conditions, but similarity should only produce benefits to self-esteem when engaging in social comparison. It is also interesting to note that the direct path from extraversion to self-esteem was not significant in the control condition, but was in the social comparison condition (Row 1), although a test of the difference between these coefficients produced a marginal result, χ²−Δ(1) = 2.73, p = .098. This suggests that perhaps some other mechanism in addition to similarity also produced self-esteem benefits in this condition.¹⁰

Table 2.
Comparison of Predictors of Self-esteem by Condition, from Structural Equation Modeling in Step 4 (Experiment 4).

Pathway Control Downward Social Comparison

Direct path
(Extraversion → self-esteem) β = .09, Z = .94, p = .349 β = .29, Z = 3.15, p = .002

Indirect a-path
(Extraversion → similarity) β = .30, Z = 3.54, p < .001 β = .36, Z = 4.20, p < .001

Indirect b-path
(Similarity → self-esteem) β = .07, Z = .80, p = .426 β = .23, Z = 2.55, p = .011

Overall indirect path (a × b) β = .02, Z = .780, p = .436 β = .08, Z = 2.23, p = .026

Summary

Experiment 4 again showed extraverts (vs. introverts) having higher self-esteem primarily during downward social comparison. It also evaluated four possible mechanisms through a multi-step empirical assessment. Although there was a conceptual basis to expect any of these mechanisms could have contributed to our self-esteem moderator effect, only perceived similarity was provisionally supported by our data. That is, extraverts were more likely to see their peers (i.e., the targets of downward social comparison) as similar to themselves. Furthermore, similarity played a similar moderating role as extraversion, such that downward social comparison primarily increased self-esteem among participants who saw peers as similar to themselves. Including the similarity interaction term reduced extraversion’s moderating role to nonsignificance. Finally, an indirect path from extraversion to similarity to self-esteem was significant under downward social comparison and nonsignificant under control.

General Discussion

Summary and Implications

These experiments contribute to both the extraversion and social comparison literatures. Concerning extraversion, they highlight the conditional nature of its association with self-esteem. After a self-threat, extraversion was strongly related to self-esteem when participants were explicitly encouraged to engage in social comparisons. However, extraversion was less related or unrelated to self-esteem in other conditions. Our effects were consistent across procedural and sampling variations. In Experiments 3 and 4, we also show that our effects were generally resistant across levels of suspicion, and that effects emerged although extraverts did not have higher performance expectations before the negative task feedback, showing that expectation differences are not necessary for effects to emerge.

Although our experiments produced a fairly consistent pattern of results, one would expect some variation in any set of studies (e.g., see Stanley & Spence, 2014) and the present studies are no exception. We also conducted several experiments not formally included in this paper. As reported in Supplemental Materials E, a meta-analysis including all data sets supported main effects of extraversion, as well as main effects of downward social comparison and BIRGing on self-esteem. In contrast, self-serving attributions did not have a significant main effect on self-esteem. Moreover, the critical interaction of Extraversion × Downward Social Comparison was significant, whereas extraversion did not interact significantly with BIRGing or self-serving attributions.

These findings help to address a gap in the extraversion literature by providing some of the first evidence that extraverts and introverts differ in how they self-enhance. Thus, these findings advance well-documented correlations between extraversion and self-esteem by suggesting a mechanism that might partially account for this relationship: extraversion was especially related to self-esteem when engaging in comparisons with peers who performed worse than themselves. Moreover, Experiment 4 revealed a link with perceived similarity, whereby extraverts saw themselves as more typical of their peer group than did introverts. This increased sense of similarity did not lead to unconditionally higher self-regard, but instead produced higher self-esteem only when participants were engaging in explicit social comparisons.

The present findings contribute to a large literature on moderators of social comparison processes. Thus, these findings reinforce a long-recognized fact that social comparison processes are not universally impactful, or at least do not affect everyone in the same way (Jones & Buckingham, 2005). However, understanding of individual differences relevant to social comparison is quite nascent with existing research pointing either to direct self-reports of people’s propensity to socially compare (Gibbons & Buunk, 1999), or work on dysphoria or self-esteem as moderators (Jones & Buckingham, 2005; Wheeler, 2000). By revealing that extraverts respond more favorably than introverts to downward social comparison, these findings contribute to a more nuanced understanding of when/for whom social comparison may be more influential.

Turning to the other self-esteem strategies, we did not find any indication that self-serving attribution improved self-esteem. Of course, it is possible that we simply did not design effective manipulations that gave participants sufficient opportunity to bolster their self-esteem. For example, a careful reader may have noticed in Experiment 3’s self-serving attribution condition that the full range of the bogus projected score estimate (44%−63%) intended as consoling feedback was still lower than the average participant’s estimation of their performance (63.4%)! Thus, even our attempt at reassuring participants with a relatively enhanced estimate may have still been disappointing to participants, rendering the condition ineffective in boosting self-esteem.

However, an alternative possibility is that self-serving attribution does not restore self-esteem, consistent with a “cold cognitive” interpretation of self-serving attributions (Miller & Ross, 1975). Although self-esteem threatening factors increase people’s tendency to engage in self-serving attributions (Campbell & Sedikides, 1999), we are not aware of any evidence demonstrating that self-serving attributions successfully increase self-esteem. Miller and Ross (1975) have argued that self-serving attribution may be part of the generalized tendency for people to attribute unexpected outcomes to external factors, whether or not they hurt one’s self-esteem. Self-serving attribution also might be driven by motivations other than feeling good about oneself—for instance, maintaining a consistent self-schema to aid in prediction of one’s own behaviors (Markus, 1977). In that case, its primary function may be more cognitive than egoistic, with limited consequences for self-esteem.

Our meta-analysis showed that the BIRG condition led to greater self-esteem than the control condition; so far as we know, a demonstration of BIRGing causally affecting self-esteem is novel. Previous work on BIRGing (Cialdini et al., 1976; Cialdini & Richardson, 1980; Miller, 2009) consistently shows that BIRGing is related to self-esteem and impression management motives, but without measuring self-esteem directly. For example, Cialdini and Richardson (1980) showed that after experiencing a personal failure (likely to affect self-esteem, among other factors including negative affect), people were more likely to BIRG. But such findings could support several interpretations; for instance, BIRGing could merely be generating positive affect, or collective self-esteem (Luhtanen & Crocker, 1992), rather than necessarily enhancing one’s personal self-esteem. Alternatively, personal failure conditions could motivate participants to attempt to restore self-esteem without necessarily succeeding. Thus, the present findings provide informative data strengthening the theoretical suggestions of these researchers, as BIRGing indeed increased self-esteem relative to control feedback.

Because BIRGing seemed to share downward social comparison’s social focus, we originally hypothesized that extraverts would gain more from this strategy, and were surprised that no such interaction emerged. Putting aside the possibility of a beta error, this pattern may indicate that our original hypothesis that extraverts would respond more effectively to all social self-enhancement tactics was too simplistic. How, then, should we interpret introverts’ apparent ability to engage successfully in BIRGing? Experiment 4’s mechanism data suggests a possible reason for the difference. Downward social comparison requires one to see a given target as similar to oneself, or else the comparison is less meaningful (Festinger, 1954) and presumably less beneficial to one’s self-worth. Extraverts see others as more similar to themselves, fueling such comparisons. However, BIRGing does not require that one see one’s successful group members as similar to oneself. Major theories of social identity (social identity theory, Tafjel & Turner, 1979; terror management theory, Solomon et al., 1991) posit identification with a good group as a universal psychological need separate from mere sociability. Introverts may engage less energetically in social activities including parties and dating (Paunonen, 2003), but this does not mean that they are apathetic about their social identity. The fact that introverts are less likely to see themselves as typical group members may not undermine this group-identification process.

Constraints on Generality

We exclusively recruited young adults from a Canadian university population. Research extending our ideas to additional populations would benefit from carefully pilot-testing initial materials. For example, researchers should determine whether false feedback stimuli are perceived as possible but highly challenging (so that the test seems fair but negative feedback seems plausible), and we recommended checking whether suspicion rates remain at a tolerable level within other groups. Indeed, researchers have flagged non-naïveté as problematic in some populations (Chandler et al., 2014), and groups expected to have familiarity with false feedback paradigms could pose issues for this research. Furthermore, we selected vocabulary tests to give ostensibly real feedback about intelligence, and emotional intelligence tests to give feedback about emotional intelligence, precisely because our psychology student population were expected to be invested in these attributes—general intelligence being instrumental to academic success, and emotional intelligence likely seen as instrumental to thriving within psychology. Alternative populations (e.g., MTurk, representative samples) may require an alternative type of feedback to optimally invest participants in the experimental realism (Carlsmith et al., 1976) likely necessary for this area of research.

Limitations and Future Directions

Although our findings interestingly demonstrate one reason why extraversion and self-esteem may relate (i.e., because extraverts gain more from downward comparisons), other explanatory bridges likely exist between these constructs. Other plausible mechanisms include differences in received social support among extraverts/introverts (Swickert et al., 2004), and high self-esteem individuals engaging in more extraverted/social behaviors, for example, because they are more receptive of positive feedback from others (Marigold et al., 2014).

One important limitation of this work is that despite a reasonably well-powered Experiment 4, we found only provisional evidence for a psychological mechanism. Specifically, we found that similarity passed three of our four tests concerning its ability to explain the extraversion social comparison advantage, as it correlated moderately with extraversion, showed a virtually identical (albeit statistically marginal) interaction with social comparison as did extraversion, and reduced extraversion’s interaction effect when included in the same regression model with it. However, our formal mediated moderation test provided somewhat modest support for similarity playing a mediating role. There are several interpretations for these somewhat mixed results. First, mediated moderation may demand larger sample sizes than what we collected to firmly establish. Second, our measure of similarity was brief and impromptu. Short measures constrain validity (Bakker & Lelkes, 2018), so follow-up research could substitute a lengthier and carefully developed similarity measure for our simple four-item scale. Third, it is entirely possible that similarity is only one of potentially several mechanisms that contribute to extraverts’ advantage over introverts in downward social comparison, meaning that much of the variance remains to be explained.

To maximize statistical power for our main research question, we did not include control or positive performance feedback groups. Instead, all participants received negative feedback, to maximize the salience of self-esteem threat among participants (Tesser, 2000). Thus, we cannot be certain that all respondents were engaging in self-esteem “repair” per se; indeed, many participants likely “bolstered” self-esteem in our experiment (i.e., if they did not decrease in self-esteem due to feedback, but increased in self-esteem after the downward social comparison). Originally, we included the feedback only to make self-esteem concerns salient, as past work indicates that this is sometimes necessary to “activate” self-enhancement (Beauregard & Dunning, 1998; Spencer et al., 1993), so this does not threaten our hypotheses—but remains a characteristic of the data that is open-ended.

The present findings also point to interesting future directions. For instance, we conceptualize the pattern of extraverts gaining more self-esteem from downward social comparison but not BIRGing as being due to the psychological differences in these strategies; namely, that downward social comparison may benefit from seeing oneself as similar to other people, whereas BIRGing is about identity benefits of group membership. Our interpretation implies that self-enhancement strategies should work better for extraverts when people focus on their similarity/dissimilarity to others. Future work could revise the BIRGing task to have people list ways in which they are typical or representative of a successful group. Because extraverts should easily identify as representative of group members, whereas introverts should struggle to do so, this emphasis on representativeness might lead introverts to talk themselves out of being legitimate group members. After all, legitimate group membership requires a minimum of perceived similarity to other members (Leonardelli et al., 2010). This revised BIRG approach might favor extraverts over introverts’ self-esteem. Similarly, downward comparison might be disconnected from similarity if individuals are asked to compare their performance against their younger selves (see Wilson & Ross, 2000), rather than against other people, highlighting their improvement across time. In such a task, we suspect that extraverts would no longer show larger self-esteem boosts than would introverts, because comparing oneself against one’s past does not rely on seeing oneself as similar to one’s peers. We are confident that these future studies would be informative in extending the present research in exciting new directions.

Supplemental Material

Supplemental Material, Vaughan-Johnston_Online_Appendix - Extraversion as a Moderator of the Efficacy of Self-Esteem Maintenance Strategies

Supplemental Material, Vaughan-Johnston_Online_Appendix for Extraversion as a Moderator of the Efficacy of Self-Esteem Maintenance Strategies by Thomas I. Vaughan-Johnston, Karen E. MacGregor, Leandre R. Fabrigar, Lyndsay E. Evraire and Louise Wasylkiw in Personality and Social Psychology Bulletin

Possible Mediator	Step 1: Association with Extraversion	Step 2: Substitutes as Moderator	Step 3: Reduces Extraversion’s Moderating Role	Step 4: Mediation Analysis
Similarity	Yes r(322) = .29 p < .001	Yes β = .10 [−.004, .21] p = .058	Somewhat: Extraversion interaction weakened to β = .09 [−.02, .20], t(318) = 1.68, p = .094	Somewhat: Pattern is as expected, but formal difference test is χ²−Δ(1) = 2.00, p = .157
Personal development competitiveness	Yes r(322) = .28 p < .001	No β = .04 [−.07, .14] p = .478	—	—
Perceived test validity	Yes r(322) = −.16 p = .005	No β = −.006 [−.11, .10] p = .909	—	—
Relevance of other to self/identity	No r(322) = −.01 p = .909	—	—	—

Pathway	Control	Downward Social Comparison
Direct path (Extraversion → self-esteem)	β = .09, Z = .94, p = .349	β = .29, Z = 3.15, p = .002
Indirect a-path (Extraversion → similarity)	β = .30, Z = 3.54, p < .001	β = .36, Z = 4.20, p < .001
Indirect b-path (Similarity → self-esteem)	β = .07, Z = .80, p = .426	β = .23, Z = 2.55, p = .011
Overall indirect path (a × b)	β = .02, Z = .780, p = .436	β = .08, Z = 2.23, p = .026

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The present research was supported by a Grant (435-2015-0114) to the third author from the Social Sciences and Humanities Research Council of Canada (SSHRC). A SSHRC Doctoral Award held by the first author (#767-2018-1484) also supported this research.

ORCID iD

Thomas I. Vaughan-Johnston

Supplemental Material

Supplemental material is available online with this article.

Notes

References

Ashton

M. C.

Lee

Paunonen

(2002). What is the central feature of extraversion? Social attention versus reward sensitivity. Journal of Personality and Social Psychology, 83, 245–252.

Bakker

B. N.

Lelkes

(2018). Selling ourselves short? How abbreviated measures of personality change the way we think about personality and politics. The Journal of Politics, 80, 1311–1325.

Beauregard

K. S.

Dunning

(1998). Turning up the contrast: Self-enhancement motives prompt egocentric contrast effects in social judgments. Journal of Personality and Social Psychology, 74(3), 606–621.

Byrne

. (1971). The attraction paradigm. Academic Press.

Campbell

W. K.

Sedikides

(1999). Self-threat magnifies the self-serving bias: A meta-analytic integration. Review of General Psychology, 3, 23–43.

Carlsmith

J. M.

Ellsworth

P. C.

Aronson

. (1976). Methods of Research in Social Psychology. Addison-Wesley.

Chandler

Mueller

Paolacci

(2014). Nonnaïveté among Amazon Mechanical Turk workers: Consequences and solutions for behavioral researchers. Behavior Research Methods, 46(1), 112–130.

Cialdini

R. B.

Borden

R. J.

Thorne

Walker

M. R.

Freeman

Sloan

L. R.

(1976). Basking in reflected glory: Three (football) field studies. Journal of Personality and Social Psychology, 34, 366–375.

Cialdini

R. B.

Richardson

K. D.

(1980). Two indirect tactics of image management: Basking and blasting. Journal of Personality and Social Psychology, 39, 406–415.

10.

Costa

P. T.

Jr & McCrae

R. R.

(1992). Normal personality assessment in clinical practice: The NEO Personality Inventory. Psychological Assessment, 4, 5–13.

11.

Edwards

J. R.

Lambert

L. S.

(2007). Methods for integrating moderation and mediation: A general analytical framework using moderated path analysis. Psychological Methods, 12, 1–22.

12.

Eysenck

H. J

. (1967). The biological basis of personality. Thomas.

13.

Eysenck

H. J

. (1990). Biological dimensions of personality. In Pervin

L. A.

(Ed.), Handbook of personality: Theory and research (pp. 244–276). Guilford.

14.

Festinger

(1954). A theory of social comparison processes. Human Relations, 7, 117–140.

15.

Francis

L. J.

(1997). The relationship between Rosenberg’s construct of self-esteem and Eysenck’s two-dimensional model of personality. Personality and Individual Differences, 21, 483–488.

16.

Furr

R. M.

Rosenthal

(2003). Evaluating theories efficiently: The nuts and bolts of contrast analysis. Understanding Statistics: Statistical Issues in Psychology, Education, and the Social Sciences, 2(1), 33–36.

17.

Gibbons

F. X.

Buunk

B. P.

(1999). Individual differences in social comparison: Development of a scale of social comparison orientation. Journal of Personality and Social Psychology, 76(1), 129–142.

18.

Gilbert

D. T.

Giesler

R. B.

Morris

K. A.

(1995). When comparisons arise. Journal of Personality and Social Psychology, 69(2), 227–236.

19.

Heatherton

T. F.

Polivy

(1991). Development and validation of a scale for measuring state self-esteem. Journal of Personality and Social Psychology, 60(6), 895–910.

20.

John

O. P.

Donahue

E. M.

Kentle

R. L.

(1991). The Big Five Inventory—versions 4a and 54. University of California, Berkeley, Institute of Personality and Social Research.

21.

Jones

A. M.

Buckingham

J. T.

(2005). Self–esteem as a moderator of the effect of social comparison on women’s body image. Journal of Social and Clinical Psychology, 24(8), 1164–1187.

22.

Jung

C. G

. (1971). Psychological types ( Hull

R. F. C.

Baynes

H. G.

, Trans.). Princeton University Press. (Original work published 1923)

23.

Kawash

G. F.

(1982). A structural analysis of self-esteem from pre-adolescence through young adulthood: Anxiety and extraversion as agents in the development of self-esteem. Journal of Clinical Psychology, 38, 301–311.

24.

Larson

J. R.

(1977). Evidence for a self-serving bias in the attribution of causality. Journal of Personality, 45, 430–441.

25.

Lee

R. M.

Dean

Jung

(2008). Social connectedness, extraversion, and subjective well-being: Testing a mediation model. Personality and Individual Differences, 44, 656–667.

26.

Leonardelli

G. J.

Pickett

C. L.

Brewer

M. B

. (2010). Optimal distinctiveness theory: A framework for social identity, social cognition, and intergroup relations. In Zanna

M. P.

Olson

J. M.

(Eds.), Advances in experimental social psychology (Vol. 43, pp. 63–113). Academic Press.

27.

Lucas

R. E.

Dyrenforth

P. S.

(2008). Explaining the extraversion/positive affect relation: Sociability cannot account for extraverts’ greater happiness. Journal of Personality, 76, 385–414.

28.

Luhtanen

Crocker

(1992). A collective self-esteem scale: Self-evaluation of one’s social identity. Personality and Social Psychology Bulletin, 18(3), 302–318.

29.

Malouff

J. M.

Thorsteinsson

E. B.

Schutte

N. S.

Bhullar

Rooke

S. E.

(2010). The five-factor model of personality and relationship satisfaction of intimate partners: A meta-analysis. Journal of Research in Personality, 44, 124–127.

30.

Marigold

D. C.

Cavallo

J. V.

Holmes

J. G.

Wood

J. V.

(2014). You can’t always give what you want: The challenge of providing social support to low self-esteem individuals. Journal of Personality and Social Psychology, 107(1), 56–80.

31.

Markus

(1977). Self-schemata and processing information about the self. Journal of Personality and Social Psychology, 35(2), 63–78.

32.

Mayer

J. D.

Salovey

Caruso

D. R.

Sitarenios

(2003). Measuring emotional intelligence with the MSCEIT V2.0. Emotion, 3, 97–105.

33.

McCrae

R. R.

Costa

P. T.

Jr (1987). Validation of the five-factor model of personality across instruments and observers. Journal of Personality and Social Psychology, 52(1), 81–90.

34.

Miller

C. B.

(2009). Yes we did! Basking in reflected glory and cutting off reflected failure in the 2008 presidential election. Analyses of Social Issues and Public Policy, 9, 283–296.

35.

Miller

D. T.

Ross

(1975). Self-serving biases in the attribution of causality: Fact or fiction. Psychological Bulletin, 82(2), 213–225.

36.

Muller

Judd

C. M.

Yzerbyt

V. Y.

(2005). When moderation is mediated and mediation is moderated. Journal of Personality and Social Psychology, 89, 852–863.

37.

Murray

S. L.

Rose

Bellavia

Holmes

J. G.

Kusche

(2002). When rejection stings: How self-esteem constrains relationship enhancement processes. Journal of Personality and Social Psychology, 83, 556–573.

38.

Nussbaum

A. D.

Dweck

C. S.

(2008). Defensiveness versus remediation: Self-theories and modes of self-esteem maintenance. Personality and Social Psychology Bulletin, 34(5), 599–612.

39.

Paunonen

S. V.

(2003). Big Five factors of personality and replicated predictions of behavior. Journal of Personality and Social Psychology, 84(2), 411–424.

40.

Preacher

K. J.

Rucker

D. D.

MacCallum

R. C.

Nicewander

W. A.

(2005). Use of the extreme groups approach: A critical reexamination and new recommendations. Psychological Methods, 10(2), 178–192.

41.

Rosenberg

(1965). Society and the adolescent self-image. University Press.

42.

Ross

S. R.

Rausch

M. K.

Canada

K. E.

(2003). Competition and cooperation in the five-factor model: Individual differences in achievement orientation. The Journal of Psychology, 137(4), 323–337.

43.

Ryckman

R. M.

Hammer

Kaczor

L. M.

Gold

J. A.

(1990). Construction of a hypercompetitive attitude scale. Journal of Personality Assessment, 55(3–4), 630–639.

44.

Ryckman

R. M.

Thornton

Butler

J. C.

(1994). Personality correlates of the hypercompetitive attitude scale: Validity tests of Horney’s theory of neurosis. Journal of Personality Assessment, 62(1), 84–94.

45.

Schmitt

D. P.

Allik

(2005). Simultaneous administration of the Rosenberg self-esteem scale in 53 nations: Exploring the universal and culture-specific features of global self-esteem. Journal of Personality and Social Psychology, 89(4), 623–642.

46.

Scollon

C. N.

Diener

(2006). Love, work, and changes in extraversion and neuroticism over time. Journal of Personality and Social Psychology, 91(6), 1152–1165.

47.

Sedikides

(1993). Assessment, enhancement, and verification determinants of the self-evaluation process. Journal of Personality and Social Psychology, 65(2), 317–338.

48.

Sedikides

Gaertner

Toguchi

(2003). Pancultural self-enhancement. Journal of Personality and Social Psychology, 84(1), 60–79.

49.

Sedikides

Strube

M. J.

(1997). Self-evaluation: To thine own self be good, to thine own self be sure, to thine own self be true, and to thine own self be better. In Zanna

M. P.

(Ed.), Advances in experimental social psychology (Vol. 29, pp. 209–269). Academic Press.

50.

Solomon

Greenberg

Pyszczynski

(1991). A terror management theory of social behavior: The psychological functions of self-esteem and cultural worldviews. In Zanna

M. P.

(Ed.), Advances in experimental social psychology (Vol. 24, pp. 93–159). Academic Press.

51.

Spencer

S. J.

Josephs

R. A.

Steele

. (1993). Low self-esteem: The uphill struggle for self-integrity. In Baumeister

(Ed.), Self-esteem: The puzzle of low self-regard (pp. 21–36). Plenum.

52.

Stanley

D. J.

Spence

J. R.

(2014). Expectations for replications: Are yours realistic? Perspectives on Psychological Science, 9(3), 305–318.

53.

Swickert

Hittner

J. B.

Kitos

Cox-Fuenzalida

(2004). Direct or indirect, that is the question. A re-evaluation of extraversion’s influence on self-esteem. Personality and Individual Differences, 36, 207–217.

54.

Swickert

Rosentretes

C. J.

Hittner

J. B.

Mushrush

J. E.

(2002). Extraversion, social support processes, and stress. Personality and Individual Differences, 32, 877–891.

55.

Tafjel

Turner

J. C

. (1979). An integrative theory of intergroup conflict. In Austin

W. G.

Worchel

(Eds.), The social psychology of intergroup relations (pp. 33–47). Brooks-Cole.

56.

Taylor

S. E.

Brown

J. D.

(1988). Illusion and well-being: A social psychological perspective on mental health. Psychological Bulletin, 103, 193–210.

57.

Tesser

. (1988). Toward a self-evaluation maintenance model of social behavior. In Berkowitz

(Ed.), Advances in experimental social psychology (Vol. 21, pp. 181–227). Academic Press.

58.

Tesser

(2000). On the confluence of self-esteem maintenance mechanisms. Personality and Social Psychology Review, 4(4), 290–299.

59.

Uziel

(2007). Individual differences in the social facilitation effect: A review and meta-analysis. Journal of Research in Personality, 41, 579–601.

60.

Van der Linden

Scholte

R. H.

Cillessen

A. H.

te Nijenhuis

Segers

(2010). Classroom ratings of likeability and popularity are related to the Big Five and the general factor of personality. Journal of Research in Personality, 44(5), 669–672.

61.

Watson

Clark

L. A.

Tellegen

(1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54(6), 1063–1070.

62.

Wheeler

(2000). Individual differences in social comparison. In Suls

J. M.

Wheeler

(Eds.), Handbook of social comparison (pp. 141–158). Springer.

63.

Wills

T. A.

(1981). Downward comparison principles in social psychology. Psychological Bulletin, 90, 245–271.

64.

Wilson

A. E.

Ross

(2000). The frequency of temporal-self and social comparisons in people’s personal appraisals. Journal of Personality and Social Psychology, 78(5), 928–942.

65.

Wood

J. V.

Taylor

S. E.

Lictman

R. R.

(1985). Social comparison in adjustment to breast cancer. Journal of Personality and Social Psychology, 49(5), 1169–1183.

66.

P. C.

Foo

M. D.

Turban

D. B.

(2008). The role of personality in relationship closeness, developer assistance, and career success. Journal of Vocational Behavior, 73(3), 440–448.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

2.29 MB