- Email: [email protected]
From Honesty–Humility to fair behavior – Benevolence or a (blind) fairness norm?
Personality and Individual Differences 80 (2015) 91–95
Contents lists available at ScienceDirect
Personality and Individual Differences journal homepage: www.elsevier.com/locate/paid
From Honesty–Humility to fair behavior – Benevolence or a (blind) fairness norm? Benjamin E. Hilbig a,⇑, Isabel Thielmann a, Johanna Wührl b, Ingo Zettler c a
University of Koblenz-Landau, Germany University of Jena, Germany c University of Copenhagen, Denmark b
a r t i c l e
i n f o
Article history: Received 4 December 2014 Received in revised form 11 February 2015 Accepted 14 February 2015 Available online 7 March 2015 Keywords: Honesty–Humility Fairness Pro-social Benevolence HEXACO model Social norm
a b s t r a c t Recent research has identified trait Honesty–Humility as a vital determinant of a variety of cooperative, fair, or pro-social behaviors. However, it is less clear what drives the link between Honesty–Humility and the tendency to exhibit fair behavior which we first demonstrate in a meta-analysis of prior work using the dictator game. To close this gap, we test two potential explanations, namely benevolence versus an internalized (blind) fairness norm. In a redistribution paradigm, we implemented an unfair situation in which participants and their hypothetical counterparts received asymmetric rewards for the same performance in a quiz. The asymmetry was either to the participants’ advantage or disadvantage, manipulated experimentally. Finally, participants were empowered to redistribute all rewards at will. In line with the benevolence account, individuals high in Honesty–Humility established a fair distribution if they themselves were advantaged (thus making sure they do not exploit others) whereas they refrained from rectifying the asymmetry (by taking points from the other) if they were disadvantaged. Vice versa, the pattern cannot be accommodated by the idea of a (blind) fairness norm which would have implied that individuals high in Honesty–Humility establish an equal distribution irrespective of who is advantaged. Ó 2015 Elsevier Ltd. All rights reserved.
Cooperation represents a key aspect of social behavior, fostering benefits of groups rather than individual agents (cf. Sullivan, Snyder, & Sullivan, 2008). A basic ingredient of individuals’ inclination to cooperate is fairness (e.g., Arneson, 1982; Fehr & Schmidt, 1999), which – in distributive terms – denotes ‘‘the concern for how money or goods are distributed among individuals’’ (Kamas & Preston, 2012, p. 538). Understanding individual differences in fairness hence provides important insights into interpersonal as well as societal functioning. Correspondingly, many studies have investigated the relation between fairness and basic personality traits (e.g., Ben-Ner, Kong, & Putterman, 2004; Kurzban & Houser, 2001). In recent years, research relying on the HEXACO Model of Personality (comprising the traits Honesty–Humility, Emotionality, Extraversion, Agreeableness, Conscientiousness, and Openness to Experience; e.g., Ashton & Lee, 2007; Lee & Ashton, 2004) has provided a fruitful approach and a more fine-grained understanding of the traits influencing fairness. The HEXACO model has emerged from lexical studies across various languages, ⇑ Corresponding author at: Cognitive Psychology Lab, Department of Psychology, University of Koblenz-Landau, Fortstraße 7, 76829 Landau, Germany. Tel.: +49 (0)6341 280 34234. E-mail address: [email protected] (B.E. Hilbig). http://dx.doi.org/10.1016/j.paid.2015.02.017 0191-8869/Ó 2015 Elsevier Ltd. All rights reserved.
essentially representing a variant and extension of the classical Big Five framework (e.g., McCrae & Costa, 1985): Beyond factors that are equivalent across the two frameworks (Extraversion, Conscientiousness, and Openness), the HEXACO model encompasses variants of Emotionality (i.e., Neuroticism) and Agreeableness (see Ashton, Lee, & De Vries, 2014, for a recent review). In addition – and most prominently – the model proposes a sixth basic factor termed Honesty–Humility, specifically capturing individual differences in fairness. Essentially, Honesty–Humility represents ‘‘the tendency to be fair and genuine in dealing with others, in the sense of cooperating with others even when one might exploit them without suffering retaliation’’ (Ashton & Lee, 2007, p. 156). As such, high levels of Honesty–Humility stand for sincerity, fairness, greed-avoidance, and modesty whereas low levels imply dishonesty, unfairness, greed, and pretentiousness. In line with this theoretical conceptualization, Honesty–Humility relates negatively to crime, delinquency (e.g., Dunlop, Morrison, Koenig, & Silcox, 2012; Van Gelder & De Vries, 2014), and counterproductive work behavior (e.g., Marcus, Lee, & Ashton, 2007) but positively to integrity (Lee, Ashton, Morrison, Cordery, & Dunlop, 2008) and fairness-related behaviors such as pro-social distribution decisions (Ackermann, Fleiß, & Murphy, in press; Hilbig, Glöckner, & Zettler, 2014).
92
B.E. Hilbig et al. / Personality and Individual Differences 80 (2015) 91–95
In particular, several studies (Hilbig, Thielmann, Hepp, Klein, & Zettler, 2015; Hilbig & Zettler, 2009; Hilbig, Zettler, Leist, & Heydasch, 2013; Thielmann & Hilbig, 2014; Thielmann, Hilbig, & Niedtfeld, 2014) have linked Honesty–Humility to fair behavior in the dictator game (for a recent review of personality influences on various games, see Zhao & Smillie, in press). In this paradigm, an individual is asked to divide an endowment at will between herself and an unknown other. Across studies, individuals high in Honesty– Humility gave more, thus refraining from abusing their power to exploit the other. Summarizing these findings in a meta-analysis (N = 1231), the sample-size weighted average correlation (cf. Field, 2001) between Honesty–Humility and dictator game giving was r = .29 and thus a medium-sized effect (Cohen, 1988). Although these findings clearly demonstrate that individuals high in Honesty–Humility behave more pro-socially, the conclusions are merely relative. That is, such a positive association does not necessarily imply that individuals high in Honesty–Humility are also more inclined towards fair behavior in absolute terms – in the sense of sharing equally which most people (i.e., more than 80%) perceive to be the fair solution in the dictator game (Mellers, Haselhuhn, Tetlock, Silva, & Isen, 2010). To investigate this more directly, we re-analyzed the data referred to above, testing whether individuals high in Honesty–Humility are actually more likely to share equally in this game. Indeed, amongst individuals high in Honesty–Humility (i.e., those with values larger or equal to the median), 64.4% chose the equal split, whereas only 34.6% of their counterparts low in Honesty–Humility did so, OR = 3.42, 95% CI [2.71, 4.33], v2(1) = 109.48, p < .001.1 This finding clearly indicates that high levels of Honesty–Humility are not only related to more pro-sociality or non-exploitation in relative terms, but also to more fairness in absolute terms. However, although extant evidence linking Honesty–Humility to fair behavior is consistent, it remains unclear why individuals with high levels of Honesty–Humility are more inclined towards fair behavior. Specifically, the empirical picture allows for two alternative explanations. First, the HEXACO theory suggests that this pattern is essentially due to benevolence. That is, Honesty– Humility is conceptualized in terms of non-exploitation, implying that individuals high in Honesty–Humility should be motivated to avoid having an edge over others (i.e., receiving more for similar input compared to others), but not mind the opposite. In simple terms, they may share equally in the dictator game due to their benevolent nature – placing high importance on justice for others, without insisting on justice for themselves. Alternatively, extant evidence could also be accounted for by an internalized (‘‘blind’’) fairness norm (i.e., a general preference for equality) in individuals high in Honesty–Humility. In terms of equity sensitivity (Huseman, Hatfield, & Miles, 1985, 1987), individuals high in Honesty–Humility might generally prefer equivalent input–outcome-ratios both for others and themselves and thus insist that no-one is under- or over-rewarded. In simple terms, they may share equally in the dictator game because this is the social fairness norm (Mellers et al., 2010), thus placing high importance on justice for both others and themselves. The purpose of the present study was to disentangle these two explanations. The benevolence account implies a mechanism that may be summarized as ‘‘I need to ensure I do not get more than others; I should not take advantage of others’’ in individuals high in Honesty–Humility. The fairness norm account, by contrast, implies ‘‘I need to ensure we all get the same (for similar input)’’. So, to tease the two apart, we used a paradigm in which participants were asked to re-distribute points that they – and a 1 Regressing the dictator game allocation (equal split: yes vs. no) on the (continuous) Honesty–Humility scores in a logistic regression analysis yielded similar results, OR = 2.90, 95% CI [2.36, 3.55], p < .001.
hypothetical other – earned through exerting (cognitive) effort in a quiz. Establishing an unfair situation, there were asymmetric rewards in the form of different gains (points earned) for the same number of correct solutions. Thus, the same effort and success in the quiz came with different outcomes. This asymmetry was either in favor of the participant (advantage-self) or in favor of the hypothetical other (advantage-other). Finally, participants could redistribute all points at will between themselves and the other. That is, participants were empowered to give points to or take points from the other and thus to increase or reduce the asymmetry (or leave it unaltered). This redistribution paradigm allows for disentangling benevolence from an internalized fairness norm as the two make incompatible predictions: According to benevolence – and the HEXACO theory – individuals high in Honesty–Humility should use their redistribution-power to reduce asymmetry only if they themselves are in the advantageous position. That is, if participants get more points than the other for the same performance in the quiz, those high in Honesty–Humility should give points to the other to reduce this unfairness. However, if participants actually get fewer points in the quiz than the other, those high in Honesty–Humility should refrain from taking points from the other due to their unwillingness to abuse their power over the final distribution of points. Participants’ final share should therefore be lower for individuals with high levels of Honesty–Humility, irrespective of the direction of asymmetry. In statistical terms, the benevolence account would thus essentially imply a main effect of Honesty–Humility on the final share. By contrast, according to an internalized fairness norm, individuals high in Honesty–Humility should be motivated to establish fairness in general, that is, an equal split (e.g., Kahneman, Knetsch, & Thaler, 1986; Van Dijk & Vermunt, 2000). Hence, they should not only give points to the other if they themselves are profiting from the asymmetry, but also take points from the other if the other is in the advantageous position. In other words, the fairness norm account predicts that the final distribution is always fair for individuals high in Honesty–Humility – no matter who was initially advantaged. In turn, the fairness norm account would imply an interaction between Honesty–Humility and the direction of asymmetry (advantage-self vs. advantage-other) in predicting the final share. Specifically, it would predict that individuals high in Honesty– Humility decrease their final share if they are initially advantaged but also increase their final share if they are initially disadvantaged. 1. Methods 1.1. Materials To measure Honesty–Humility, we used the German version (Moshagen, Hilbig, & Zettler, 2014) of the 60-item HEXACO Personality Inventory Revised (HEXACO-60; Ashton & Lee, 2009). For each of the six HEXACO dimensions, the inventory contains 10 items to be answered on a five-point Likert-type scale. Sample items for Honesty–Humility are ‘‘Having a lot of money is not especially important to me’’ and ‘‘I would never accept a bribe, even if it were very large’’. In our study, Cronbach’s a for Honesty–Humility was .79. As introduced above, individuals’ redistribution decision in a redistribution paradigm2 (that is, the final distribution of points in this paradigm) served as the main dependent variable – to differentiate between benevolence and internalized fairness norm. The 2 The paradigm is structurally similar to a variant of the dictator game in which dictators earn their endowment (Cherry, Frykblom, & Shogren, 2002). However, contrary to the standard dictator game, both individuals earn an endowment and, more importantly, there is asymmetry in the earnings (different payoff for the same effort or success).
93
B.E. Hilbig et al. / Personality and Individual Differences 80 (2015) 91–95
paradigm was designed to measure participants’ willingness to rectify an existing asymmetry in input–outcome-ratio, depending on the direction and the degree of the asymmetry. In the paradigm, participants were coupled with an unknown other.3 Both earned points by taking a quiz (15 multiple choice tasks with four response options each) on verbal and mathematical skills (similar to the 10Minute-Test; cf. Hilbig, Heydasch, & Zettler, 2014). As an exemplary task consider: ‘‘Which of the following fractions is smallest: 2/4, 2/5, 1/3, or 3/8?’’. For each correct answer in the quiz, participants (and the other) earned points. To ensure that solving the tasks was strenuous to some extent – thus eliciting a feeling of earning the points – we speeded the quiz with a limit of 5 min. Supporting the suitability of this time limit, participants in an independent lab-based prestudy (N = 63, 30 female, aged 18–27, M = 20.4, SD = 1.8 years) solved M = 8.8 (59%) tasks (SD = 2.8) correctly and only three participants (<5% of the sample) were able to solve 14 or all 15 tasks, despite substantial monetary incentives (1.00€ for each task solved, thus up to 15.00€ in 5 min). Also, in this pre-study, performance in the quiz was unrelated to the six HEXACO factors (as assessed via the German HEXACO-60), all |r| < .15. Following the quiz, participants learned how many tasks they had solved correctly and how many points this earned them. They also learned about the number of correct solutions given by the hypothetical other which was always the same as the participants’. As sketched above, this performance was then translated into hypothetical payoffs (points) asymmetrically. Specifically, participants received either more (advantage-self) or less (advantage-other) points than the other for the same number of correct responses. In addition, we varied the degree of asymmetry, such that an individual’s points were either doubled (moderate asymmetry) or quadrupled (severe asymmetry) as compared to the counterpart’s points. For illustration, Table 1 displays example payoffs (assuming 10 correct solutions) for the four experimental conditions to which participants were randomly assigned. Finally, participants were given the power to redistribute all the points between themselves and the other. That is, they could reduce the asymmetry by either giving points to the other or taking points from the other. As dependent measure, we used participants’ relative outcome, that is, the ratio of (i) a participant’s own points after the redistribution to (ii) the sum of all points earned by both individuals. The relative outcome hence mirrors a participant’s final share. For the sake of simplicity, we subtracted 0.5 from this proportion so that a value of 0 mirrors establishment of an equal split (i.e., no asymmetry left after redistribution). Any value larger than 0 indicates that participants gained beyond the equal split through their redistribution choice; any value smaller than 0 indicates the opposite (the other came off better than implied by the equal split). Note that, by implication, the same value of the relative outcome implies different redistribution choices across the two conditions. For example, a value of 0 indicates giving in the advantage-self condition, but taking in the advantage-other condition. 1.2. Procedure and participants The study was run online in close adherence to common standards for Internet-based experimenting (Reips, 2002). After providing informed consent and demographic information, participants completed the HEXACO-60. Next, they worked on the quiz in which they (and the other) earned the points. After the quiz, participants received information on their own and the other’s earnings and decided on the final distribution of points 3 Note that the paradigm was completely hypothetical in nature, meaning that participants were asked to imagine both the other person and the points earned and to be distributed. Thus the points did not convert to actual monetary stakes.
Table 1 Exemplary initial allocation of points for a participant with 10 correct solutions in the quiz, separated by experimental condition. Degree of asymmetry
Advantage-self
Advantage-other
Own points
Others’ points
Own points
Others’ points
Moderate Severe
40 80
20 20
20 20
40 80
Table 2 Means and standard deviations of the final proportion of participants’ points on all points (i.e., relative outcome) separated for the four experimental conditions. Degree of asymmetry
Moderate Severe
Advantage-self
Advantage-other
n
M
SD
n
36 36
0.11 0.12
0.12 0.15
36 38
M 0.03 0.06
SD 0.13 0.16
Note: a value of 0 indicates the equal split; values larger than 0 indicate that the final distribution was beneficial for the participant relative to the other (and vice versa for values below 0).
by giving, taking, or leaving the points untouched. As a manipulation check, we finally asked participants about the strenuousness of the quiz and further assessed the seriousness of participation. Participants were recruited from a community sample via mailing lists and social networks. Out of 185 participants who started the study, N = 146 (82 female, aged 18–62, M = 25.3, SD = 8.4) completed all tasks and were thus considered in the analyses.4 The majority (72.6%) of participants were students, 15.8% were employees. Participants were almost equally distributed across the experimental conditions, with n = 72 in the advantage-self (n = 36 moderate and n = 36 severe) and n = 74 in the advantage-other (n = 36 moderate and n = 38 severe) condition. 2. Results As intended by the chosen time limit – and supporting the ‘‘earning character’’ of the fairness paradigm – participants perceived the quiz to be moderately strenuous (M = 2.89, SD = 0.83, on a 5-point Likert-type scale). On average, they solved 8 out of 15 tasks (SD = 3.0) which is highly similar to the performance observed in the incentivized pilot study summarized above. In turn, participants and others earned a total of 67.7 points (SD = 28.5). In the redistribution of these points, 42% of participants took points from the other (on average 19.2 points, SD = 13.2), 31% gave points to the other (on average 14.8 points, SD = 8.9), and 27% left the distribution untouched. A majority of 51% of participants established an equal split, 32% advantaged themselves, and 17% advantaged the other. Correspondingly, after the redistribution, participants had 35.7 points on average (SD = 18.6) whereas the other had 31.9 points on average (SD = 17.9). For the relative outcome (i.e., the final proportion of participant’s points), Table 2 provides the means and standard deviations for all four experimental conditions. As can be seen (in a column-wise comparison), the relative outcome was larger when participants themselves profited from the asymmetry (M = 0.12, SD = 0.13) – as compared to the condition in which they were disadvantaged (M = 0.05, SD = 0.15). In particular, in the advantage-self condition, participants were less willing to shift 4 As indicated by a post hoc power analysis using G * Power (Faul, Erdfelder, Lang, & Buchner, 2007), this sample size yielded a high statistical power to detect small-tomedium sized (f2 = .10) main effects (in case of two predictors in total, 1 b = .99) as well as small-to-medium sized interaction effects (1 b = .93; significant increase in explained variance when adding an interaction term) in a multiple linear regression.
94
B.E. Hilbig et al. / Personality and Individual Differences 80 (2015) 91–95
Fig. 1. Relation between the relative outcome and Honesty–Humility (centered on the sample mean) separated for advantage-self (left) and advantage-other condition (right). The dotted line indicates the equal split distribution; the dashed line indicates the initial distribution prior to the redistribution choice. Confidence bands represent the 95% confidence intervals.
towards the equal split. This main effect was confirmed by a 2 (advantage: self vs. other) by 2 (degree of asymmetry: moderate vs. severe) ANOVA with F(1, 142) = 47.49, p < .001, gp2 = .25. In contrast, the severity of asymmetry (row-wise comparison in Table 2) did not influence participants’ relative outcome, F(1, 142) < 1, p = .658, gp2 < .01. That is, participants were as willing to establish the equal distribution if the asymmetry was moderate (M = 0.04, SD = 0.14) as compared to when the asymmetry was severe (M = 0.03, SD = 0.18). Finally, there was no interaction between the direction and the degree of asymmetry (F < 1). For this reason – and given that our main question refers to the effect of who is advantaged (i.e., self vs. other), not how severely someone is advantaged, on the redistribution decision – we discarded the severityfactor in the following analyses. To test the main hypotheses, we regressed the direction of asymmetry (dummy coded), Honesty–Humility, and their interaction on the relative outcome. First off, this analysis revealed a (positive) main effect of the direction of asymmetry, indicating a larger relative outcome in the advantage-self condition than in the advantage-other condition, b = .52, p < .001. Furthermore – in line with the benevolence account – we found a (negative) main effect of Honesty–Humility on the relative outcome, b = .24, p = .013. That is, high levels of Honesty–Humility were generally associated with smaller values in the relative outcome, irrespective of the direction of asymmetry. However, in contrast to the predictions of the fairness norm account, there was no interaction between Honesty–Humility and the direction of asymmetry, b = .06, p = .529.5 The results were comparable when controlling for the remaining HEXACO dimensions in a multiple regression analysis.6 As can be seen in Fig. 1, individuals low in Honesty–Humility left the asymmetry untouched whenever they benefited from it but strongly reduced the asymmetry (took points from the other) whenever it disadvantaged them. By contrast, their counterparts high in Honesty–Humility reduced the asymmetry especially whenever they originally profited from it (gave points to the other), but did not 5 Moreover, this interaction did not depend on the severity of asymmetry as indicated by a non-significant three-way-interaction between Honesty–Humility, direction, and severity of asymmetry, b = .02, p = .895. 6 The corresponding regression coefficients were b = .00 (Emotionality), b = .08 (Extraversion), b = .08 (Agreeableness), b = .11 (Conscientiousness), and b = .09 (Openness), all p > .19.
establish a full disadvantaged.
equal
split
whenever
they
were
initially
3. Discussion Fairness represents a vital aspect of interpersonal and societal functioning. Regarding individual differences in fairness and traits driving fair behavior, recent developments point to Honesty– Humility as a vital basic trait dimension (cf. Ashton et al., 2014). In line with its theoretical conceptualization, past research has consistently linked Honesty–Humility to fairness-related and pro-social behaviors. Notably – as indicated by our re-analysis of extant findings on economic behavior – Honesty–Humility is not only associated with pro-sociality in relative terms, but also with an inclination to establish fairness (i.e., equality) in absolute terms. However, different explanations may account for why these individuals show such a strong tendency to establish fairness. First, the HEXACO theory primarily suggests that benevolence drives these effects in that individuals high in Honesty–Humility avoid exploiting others. However, prior evidence cannot rule out an internalized (blind) fairness norm (i.e., a general preference for equality) in these individuals. To tease the two explanations apart, the present study assessed participants’ willingness to rectify an initial asymmetry in rewards, that is, unfair outcomes following a demanding task. Specifically, it was tested whether individuals differ in their willingness to redistribute outcomes conditional upon (a) whether they themselves or the other profited from the asymmetric distribution of rewards and (b) their individual level of Honesty–Humility. In line with the benevolence account, our results show that individuals high in Honesty–Humility reduce an asymmetry and establish an equal split if and only if they themselves are the ones profiting from the unfairness. If, in turn, it is to their disadvantage, they do not alter the asymmetric distribution as strongly and mostly refrain from taking points from the other (even though this would actually result in a more fair distribution). These results corroborate the non-exploitation aspect of Honesty–Humility as the motivating force behind fair behavior, but are incompatible with the alternative notion of a general (blind) fairness norm. Of course, we cannot claim that benevolence is the only explanation that might account for the observed pattern of results. However, among the two
B.E. Hilbig et al. / Personality and Individual Differences 80 (2015) 91–95
accounts that we derived from the extant literature, the data exclusively supports one (benevolence) and contradicts the other (blind fairness norm). In turn, our findings support the conceptualization of Honesty–Humility in the HEXACO model, demonstrating that individuals with high factor levels do not take advantage of their power even if they themselves are in a disadvantageous position. Extending prior research, our study is the first specifically designed to test why individuals with high levels in Honesty– Humility are inclined towards fair behavior. We hence provide a first toehold towards explaining the consistently observed link between Honesty–Humility and fairness-related behaviors. However, as a limitation, our findings are based on a hypothetical scenario without actual stakes, thus potentially limiting the conclusiveness of results due to socially desirable responding. Arguably, fairness is socially desirable (Hoffman, McCabe, Shachat, & Smith, 1994) as are high levels of Honesty–Humility (Hilbig, Moshagen, & Zettler, 2015). Nonetheless, as indicated by recent research, the relation between Honesty–Humility and non-exploitation seems to be equivalent in hypothetical scenarios vs. real decisions with actual stakes (Hilbig et al., 2015). In conclusion, the present study provides first evidence that the willingness to establish fairness shown by individuals high in Honesty–Humility can be traced back to benevolence and a reluctance to take advantage of others’ inferiority, but not to a general preference for equality in terms of adherence to a ‘‘blind’’ fairness norm. Given that Honesty–Humility explains a variety of socially relevant behaviors such as crime, delinquency, and cooperation (e.g., Hilbig, Glöckner, et al., 2014; Van Gelder & De Vries, 2014), it is important to understand these behavioral tendencies, shedding light on the ‘‘black box’’ between Honesty–Humility and pro-social behavior. Acknowedgements The work reported herein was supported by grants to the first author from the Baden-Württemberg Foundation (Germany) and the German Research Foundation (HI 1600/1-1). References Ackermann, K. A., Fleiß, J., & Murphy, R. O. (in press). Reciprocity as an individual difference. Journal of Conflict Resolution. http://dx.doi.org/10.1177/ 0022002714541854. Arneson, R. J. (1982). The principle of fairness and free-rider problems. Ethics, 92(4). Ashton, M. C., & Lee, K. (2007). Empirical, theoretical, and practical advantages of the HEXACO model of personality structure. Personality and Social Psychology Review, 11(2), 150–166. http://dx.doi.org/10.1177/1088868306294907. Ashton, M. C., & Lee, K. (2009). The HEXACO-60: A short measure of the major dimensions of personality. Journal of Personality Assessment, 91(4), 340–345. http://dx.doi.org/10.1080/00223890902935878. Ashton, M. C., Lee, K., & De Vries, R. E. (2014). The HEXACO Honesty–Humility, Agreeableness, and Emotionality factors: A review of research and theory. Personality and Social Psychology Review, 18(2), 139–152. http://dx.doi.org/ 10.1177/1088868314523838. Ben-Ner, A., Kong, F., & Putterman, L. (2004). Share and share alike? Gender-pairing, personality, and cognitive ability as determinants of giving. Journal of Economic Psychology, 25(5), 581–589. http://dx.doi.org/10.1016/S0167-4870(03)00065-5. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (rev. ed.). Hillsdale, NJ, England: Lawrence Erlbaum Associates Inc.. Cherry, T. L., Frykblom, P., & Shogren, J. F. (2002). Hardnose the dictator. American Economic Review, 92(4), 1218–1221. http://dx.doi.org/10.1257/ 00028280260344740. Dunlop, P. D., Morrison, D. L., Koenig, J., & Silcox, B. (2012). Comparing the Eysenck and HEXACO models of personality in the prediction of adult delinquency. European Journal of Personality, 26(3), 194–202. http://dx.doi.org/10.1002/ per.824. Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G⁄Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39(2), 175–191. http://dx.doi.org/10.3758/ BF03193146. Fehr, E., & Schmidt, K. M. (1999). A theory of fairness, competition, and cooperation. Quarterly Journal of Economics, 114(3), 817–868. http://dx.doi.org/10.1162/ 003355399556151.
95
Field, A. P. (2001). Meta-analysis of correlation coefficients: A Monte Carlo comparison of fixed- and random-effects methods. Psychological Methods, 6(2), 161–180. Hilbig, B. E., Glöckner, A., & Zettler, I. (2014). Personality and pro-social behavior: Linking basic traits and social value orientations. Journal of Personality and Social Psychology, 107(3), 529–539. http://dx.doi.org/10.1037/a0036074. Hilbig, B. E., Heydasch, T., & Zettler, I. (2014). To boast or not to boast: Testing the humility aspect of the Honesty–Humility factor. Personality and Individual Differences, 69, 12–16. Hilbig, B. E., Moshagen, M., & Zettler, I. (2015). Truth will out: Linking personality, morality, and honesty through indirect questioning. Social Psychological and Personality Science, 6, 140–147. http://dx.doi.org/10.1177/1948550614553640. Hilbig, B. E., Thielmann, I., Hepp, J., Klein, S., & Zettler, I. (2015). From personality to altruistic behavior (and back): Evidence from a double-blind dictator game. Journal of Research in Personality, 55, 46–50. http://dx.doi.org/10.1016/ j.jrp.2014.12.004. Hilbig, B. E., & Zettler, I. (2009). Pillars of cooperation: Honesty–Humility, social value orientations, and economic behavior. Journal of Research in Personality, 43(3), 516–519. http://dx.doi.org/10.1016/j.jrp.2009.01.003. Hilbig, B. E., Zettler, I., Leist, F., & Heydasch, T. (2013). It takes two: Honesty– Humility and Agreeableness differentially predict active versus reactive cooperation. Personality and Individual Differences, 54(5), 598–603. http:// dx.doi.org/10.1016/j.paid.2012.11.008. Hoffman, E., McCabe, K. A., Shachat, K., & Smith, V. (1994). Preferences, property rights, and anonymity in bargaining games. Games and Economic Behavior, 7(3), 346–380. Huseman, R. C., Hatfield, J. D., & Miles, E. W. (1985). Test for individual perceptions of job equity: Some preliminary findings. Perceptual and Motor Skills, 61(3), 1055–1064. http://dx.doi.org/10.2466/pms.1985.61.3f.1055 (Pt. 2). Huseman, R. C., Hatfield, J. D., & Miles, E. W. (1987). A new perspective on equity theory: The equity sensitivity construct. The Academy of Management Review, 12(2), 222–234. http://dx.doi.org/10.2307/258531. Kahneman, D., Knetsch, J. L., & Thaler, R. H. (1986). Fairness and the assumptions of economics. Journal of Business, 59(4), S285–S300. Kamas, L., & Preston, A. (2012). Distributive and reciprocal fairness: What can we learn from the heterogeneity of social preferences? Journal of Economic Psychology, 33(3), 538–553. http://dx.doi.org/10.1016/j.joep.2011.12.003. Kurzban, R., & Houser, D. (2001). Individual differences in cooperation in a circular public goods game. European Journal of Personality, 15(1), S37–S52. http:// dx.doi.org/10.1002/per.420 [Special issue]. Lee, K., & Ashton, M. C. (2004). Psychometric properties of the HEXACO personality inventory. Multivariate Behavioral Research, 39(2), 329–358. http://dx.doi.org/ 10.1207/s15327906mbr3902_8. Lee, K., Ashton, M. C., Morrison, D. L., Cordery, J., & Dunlop, P. D. (2008). Predicting integrity with the HEXACO personality model: Use of self- and observer reports. Journal of Occupational and Organizational Psychology, 81(1), 147–167. http:// dx.doi.org/10.1348/096317907X195175. Marcus, B., Lee, K., & Ashton, M. C. (2007). Personality dimensions explaining relationships between integrity tests and counterproductive behavior: Big five, or one in addition? Personnel Psychology, 60(1), 1–34. http://dx.doi.org/10.1111/ j.1744-6570.2007.00063.x. McCrae, R. R., & Costa, P. T. (1985). Updating Norman’s ‘‘adequacy taxonomy’’: Intelligence and personality dimensions in natural language and in questionnaires. Journal of Personality and Social Psychology, 49(3), 710–721. http://dx.doi.org/10.1037/0022-3514.49.3.710. Mellers, B. A., Haselhuhn, M. P., Tetlock, P. E., Silva, J. C., & Isen, A. M. (2010). Predicting behavior in economic games by looking through the eyes of the players. Journal of Experimental Psychology: General, 139(4), 743–755. http:// dx.doi.org/10.1037/a0020280. Moshagen, M., Hilbig, B. E., & Zettler, I. (2014). Faktorstruktur, psychometrische Eigenschaften und Messinvarianz der deutschen Version des 60-Item HEXACO Persönlichkeitsinventars. Diagnostica, 60(2), 86–97. Reips, U.-D. (2002). Standards for internet-based experimenting. Experimental Psychology, 49(4), 243–256. http://dx.doi.org/10.1026//1618-3169.49.4.243. Sullivan, B. A., Snyder, M., & Sullivan, J. L. (2008). The centrality of cooperation in the functioning of individuals and groups. In B. A. Sullivan, M. Snyder, & J. L. Sullivan (Eds.), Cooperation: The political psychology of effective human interaction (pp. 1–16). Malden: Blackwell Publishing. Thielmann, I., & Hilbig, B. E. (2014). Trust in me, trust in you: A social projection account of the link between personality, cooperativeness, and trustworthiness expectations. Journal of Research in Personality, 50(3), 61–65. http://dx.doi.org/ 10.1016/j.jrp.2014.03.006. Thielmann, I., Hilbig, B. E., & Niedtfeld, I. (2014). Willing to give but not to forgive: Borderline personality features and cooperative behavior. Journal of Personality Disorders, 28(6), 778–795. http://dx.doi.org/10.1521/pedi_2014_28_135. Van Dijk, E., & Vermunt, R. (2000). Strategy and fairness in social decision making: Sometimes it pays to be powerless. Journal of Experimental Social Psychology, 36(1), 1–25. http://dx.doi.org/10.1006/jesp.1999.1392. Van Gelder, J.-L., & De Vries, R. E. (2014). Rational misbehavior? Evaluating an integrated dual-process model of criminal decision making. Journal of Quantitative Criminology, 30(1), 1–27. http://dx.doi.org/10.1007/s10940-0129192-8. Zhao, K., & Smillie, L. D. (in press). The role of interpersonal traits in social decision making: Exploring sources of behavioral heterogeneity in economic games. Personality and Social Psychology Review. http://dx.doi.org/10.1177/ 1088868314553709.
Contents lists available at ScienceDirect
Personality and Individual Differences journal homepage: www.elsevier.com/locate/paid
From Honesty–Humility to fair behavior – Benevolence or a (blind) fairness norm? Benjamin E. Hilbig a,⇑, Isabel Thielmann a, Johanna Wührl b, Ingo Zettler c a
University of Koblenz-Landau, Germany University of Jena, Germany c University of Copenhagen, Denmark b
a r t i c l e
i n f o
Article history: Received 4 December 2014 Received in revised form 11 February 2015 Accepted 14 February 2015 Available online 7 March 2015 Keywords: Honesty–Humility Fairness Pro-social Benevolence HEXACO model Social norm
a b s t r a c t Recent research has identified trait Honesty–Humility as a vital determinant of a variety of cooperative, fair, or pro-social behaviors. However, it is less clear what drives the link between Honesty–Humility and the tendency to exhibit fair behavior which we first demonstrate in a meta-analysis of prior work using the dictator game. To close this gap, we test two potential explanations, namely benevolence versus an internalized (blind) fairness norm. In a redistribution paradigm, we implemented an unfair situation in which participants and their hypothetical counterparts received asymmetric rewards for the same performance in a quiz. The asymmetry was either to the participants’ advantage or disadvantage, manipulated experimentally. Finally, participants were empowered to redistribute all rewards at will. In line with the benevolence account, individuals high in Honesty–Humility established a fair distribution if they themselves were advantaged (thus making sure they do not exploit others) whereas they refrained from rectifying the asymmetry (by taking points from the other) if they were disadvantaged. Vice versa, the pattern cannot be accommodated by the idea of a (blind) fairness norm which would have implied that individuals high in Honesty–Humility establish an equal distribution irrespective of who is advantaged. Ó 2015 Elsevier Ltd. All rights reserved.
Cooperation represents a key aspect of social behavior, fostering benefits of groups rather than individual agents (cf. Sullivan, Snyder, & Sullivan, 2008). A basic ingredient of individuals’ inclination to cooperate is fairness (e.g., Arneson, 1982; Fehr & Schmidt, 1999), which – in distributive terms – denotes ‘‘the concern for how money or goods are distributed among individuals’’ (Kamas & Preston, 2012, p. 538). Understanding individual differences in fairness hence provides important insights into interpersonal as well as societal functioning. Correspondingly, many studies have investigated the relation between fairness and basic personality traits (e.g., Ben-Ner, Kong, & Putterman, 2004; Kurzban & Houser, 2001). In recent years, research relying on the HEXACO Model of Personality (comprising the traits Honesty–Humility, Emotionality, Extraversion, Agreeableness, Conscientiousness, and Openness to Experience; e.g., Ashton & Lee, 2007; Lee & Ashton, 2004) has provided a fruitful approach and a more fine-grained understanding of the traits influencing fairness. The HEXACO model has emerged from lexical studies across various languages, ⇑ Corresponding author at: Cognitive Psychology Lab, Department of Psychology, University of Koblenz-Landau, Fortstraße 7, 76829 Landau, Germany. Tel.: +49 (0)6341 280 34234. E-mail address: [email protected] (B.E. Hilbig). http://dx.doi.org/10.1016/j.paid.2015.02.017 0191-8869/Ó 2015 Elsevier Ltd. All rights reserved.
essentially representing a variant and extension of the classical Big Five framework (e.g., McCrae & Costa, 1985): Beyond factors that are equivalent across the two frameworks (Extraversion, Conscientiousness, and Openness), the HEXACO model encompasses variants of Emotionality (i.e., Neuroticism) and Agreeableness (see Ashton, Lee, & De Vries, 2014, for a recent review). In addition – and most prominently – the model proposes a sixth basic factor termed Honesty–Humility, specifically capturing individual differences in fairness. Essentially, Honesty–Humility represents ‘‘the tendency to be fair and genuine in dealing with others, in the sense of cooperating with others even when one might exploit them without suffering retaliation’’ (Ashton & Lee, 2007, p. 156). As such, high levels of Honesty–Humility stand for sincerity, fairness, greed-avoidance, and modesty whereas low levels imply dishonesty, unfairness, greed, and pretentiousness. In line with this theoretical conceptualization, Honesty–Humility relates negatively to crime, delinquency (e.g., Dunlop, Morrison, Koenig, & Silcox, 2012; Van Gelder & De Vries, 2014), and counterproductive work behavior (e.g., Marcus, Lee, & Ashton, 2007) but positively to integrity (Lee, Ashton, Morrison, Cordery, & Dunlop, 2008) and fairness-related behaviors such as pro-social distribution decisions (Ackermann, Fleiß, & Murphy, in press; Hilbig, Glöckner, & Zettler, 2014).
92
B.E. Hilbig et al. / Personality and Individual Differences 80 (2015) 91–95
In particular, several studies (Hilbig, Thielmann, Hepp, Klein, & Zettler, 2015; Hilbig & Zettler, 2009; Hilbig, Zettler, Leist, & Heydasch, 2013; Thielmann & Hilbig, 2014; Thielmann, Hilbig, & Niedtfeld, 2014) have linked Honesty–Humility to fair behavior in the dictator game (for a recent review of personality influences on various games, see Zhao & Smillie, in press). In this paradigm, an individual is asked to divide an endowment at will between herself and an unknown other. Across studies, individuals high in Honesty– Humility gave more, thus refraining from abusing their power to exploit the other. Summarizing these findings in a meta-analysis (N = 1231), the sample-size weighted average correlation (cf. Field, 2001) between Honesty–Humility and dictator game giving was r = .29 and thus a medium-sized effect (Cohen, 1988). Although these findings clearly demonstrate that individuals high in Honesty–Humility behave more pro-socially, the conclusions are merely relative. That is, such a positive association does not necessarily imply that individuals high in Honesty–Humility are also more inclined towards fair behavior in absolute terms – in the sense of sharing equally which most people (i.e., more than 80%) perceive to be the fair solution in the dictator game (Mellers, Haselhuhn, Tetlock, Silva, & Isen, 2010). To investigate this more directly, we re-analyzed the data referred to above, testing whether individuals high in Honesty–Humility are actually more likely to share equally in this game. Indeed, amongst individuals high in Honesty–Humility (i.e., those with values larger or equal to the median), 64.4% chose the equal split, whereas only 34.6% of their counterparts low in Honesty–Humility did so, OR = 3.42, 95% CI [2.71, 4.33], v2(1) = 109.48, p < .001.1 This finding clearly indicates that high levels of Honesty–Humility are not only related to more pro-sociality or non-exploitation in relative terms, but also to more fairness in absolute terms. However, although extant evidence linking Honesty–Humility to fair behavior is consistent, it remains unclear why individuals with high levels of Honesty–Humility are more inclined towards fair behavior. Specifically, the empirical picture allows for two alternative explanations. First, the HEXACO theory suggests that this pattern is essentially due to benevolence. That is, Honesty– Humility is conceptualized in terms of non-exploitation, implying that individuals high in Honesty–Humility should be motivated to avoid having an edge over others (i.e., receiving more for similar input compared to others), but not mind the opposite. In simple terms, they may share equally in the dictator game due to their benevolent nature – placing high importance on justice for others, without insisting on justice for themselves. Alternatively, extant evidence could also be accounted for by an internalized (‘‘blind’’) fairness norm (i.e., a general preference for equality) in individuals high in Honesty–Humility. In terms of equity sensitivity (Huseman, Hatfield, & Miles, 1985, 1987), individuals high in Honesty–Humility might generally prefer equivalent input–outcome-ratios both for others and themselves and thus insist that no-one is under- or over-rewarded. In simple terms, they may share equally in the dictator game because this is the social fairness norm (Mellers et al., 2010), thus placing high importance on justice for both others and themselves. The purpose of the present study was to disentangle these two explanations. The benevolence account implies a mechanism that may be summarized as ‘‘I need to ensure I do not get more than others; I should not take advantage of others’’ in individuals high in Honesty–Humility. The fairness norm account, by contrast, implies ‘‘I need to ensure we all get the same (for similar input)’’. So, to tease the two apart, we used a paradigm in which participants were asked to re-distribute points that they – and a 1 Regressing the dictator game allocation (equal split: yes vs. no) on the (continuous) Honesty–Humility scores in a logistic regression analysis yielded similar results, OR = 2.90, 95% CI [2.36, 3.55], p < .001.
hypothetical other – earned through exerting (cognitive) effort in a quiz. Establishing an unfair situation, there were asymmetric rewards in the form of different gains (points earned) for the same number of correct solutions. Thus, the same effort and success in the quiz came with different outcomes. This asymmetry was either in favor of the participant (advantage-self) or in favor of the hypothetical other (advantage-other). Finally, participants could redistribute all points at will between themselves and the other. That is, participants were empowered to give points to or take points from the other and thus to increase or reduce the asymmetry (or leave it unaltered). This redistribution paradigm allows for disentangling benevolence from an internalized fairness norm as the two make incompatible predictions: According to benevolence – and the HEXACO theory – individuals high in Honesty–Humility should use their redistribution-power to reduce asymmetry only if they themselves are in the advantageous position. That is, if participants get more points than the other for the same performance in the quiz, those high in Honesty–Humility should give points to the other to reduce this unfairness. However, if participants actually get fewer points in the quiz than the other, those high in Honesty–Humility should refrain from taking points from the other due to their unwillingness to abuse their power over the final distribution of points. Participants’ final share should therefore be lower for individuals with high levels of Honesty–Humility, irrespective of the direction of asymmetry. In statistical terms, the benevolence account would thus essentially imply a main effect of Honesty–Humility on the final share. By contrast, according to an internalized fairness norm, individuals high in Honesty–Humility should be motivated to establish fairness in general, that is, an equal split (e.g., Kahneman, Knetsch, & Thaler, 1986; Van Dijk & Vermunt, 2000). Hence, they should not only give points to the other if they themselves are profiting from the asymmetry, but also take points from the other if the other is in the advantageous position. In other words, the fairness norm account predicts that the final distribution is always fair for individuals high in Honesty–Humility – no matter who was initially advantaged. In turn, the fairness norm account would imply an interaction between Honesty–Humility and the direction of asymmetry (advantage-self vs. advantage-other) in predicting the final share. Specifically, it would predict that individuals high in Honesty– Humility decrease their final share if they are initially advantaged but also increase their final share if they are initially disadvantaged. 1. Methods 1.1. Materials To measure Honesty–Humility, we used the German version (Moshagen, Hilbig, & Zettler, 2014) of the 60-item HEXACO Personality Inventory Revised (HEXACO-60; Ashton & Lee, 2009). For each of the six HEXACO dimensions, the inventory contains 10 items to be answered on a five-point Likert-type scale. Sample items for Honesty–Humility are ‘‘Having a lot of money is not especially important to me’’ and ‘‘I would never accept a bribe, even if it were very large’’. In our study, Cronbach’s a for Honesty–Humility was .79. As introduced above, individuals’ redistribution decision in a redistribution paradigm2 (that is, the final distribution of points in this paradigm) served as the main dependent variable – to differentiate between benevolence and internalized fairness norm. The 2 The paradigm is structurally similar to a variant of the dictator game in which dictators earn their endowment (Cherry, Frykblom, & Shogren, 2002). However, contrary to the standard dictator game, both individuals earn an endowment and, more importantly, there is asymmetry in the earnings (different payoff for the same effort or success).
93
B.E. Hilbig et al. / Personality and Individual Differences 80 (2015) 91–95
paradigm was designed to measure participants’ willingness to rectify an existing asymmetry in input–outcome-ratio, depending on the direction and the degree of the asymmetry. In the paradigm, participants were coupled with an unknown other.3 Both earned points by taking a quiz (15 multiple choice tasks with four response options each) on verbal and mathematical skills (similar to the 10Minute-Test; cf. Hilbig, Heydasch, & Zettler, 2014). As an exemplary task consider: ‘‘Which of the following fractions is smallest: 2/4, 2/5, 1/3, or 3/8?’’. For each correct answer in the quiz, participants (and the other) earned points. To ensure that solving the tasks was strenuous to some extent – thus eliciting a feeling of earning the points – we speeded the quiz with a limit of 5 min. Supporting the suitability of this time limit, participants in an independent lab-based prestudy (N = 63, 30 female, aged 18–27, M = 20.4, SD = 1.8 years) solved M = 8.8 (59%) tasks (SD = 2.8) correctly and only three participants (<5% of the sample) were able to solve 14 or all 15 tasks, despite substantial monetary incentives (1.00€ for each task solved, thus up to 15.00€ in 5 min). Also, in this pre-study, performance in the quiz was unrelated to the six HEXACO factors (as assessed via the German HEXACO-60), all |r| < .15. Following the quiz, participants learned how many tasks they had solved correctly and how many points this earned them. They also learned about the number of correct solutions given by the hypothetical other which was always the same as the participants’. As sketched above, this performance was then translated into hypothetical payoffs (points) asymmetrically. Specifically, participants received either more (advantage-self) or less (advantage-other) points than the other for the same number of correct responses. In addition, we varied the degree of asymmetry, such that an individual’s points were either doubled (moderate asymmetry) or quadrupled (severe asymmetry) as compared to the counterpart’s points. For illustration, Table 1 displays example payoffs (assuming 10 correct solutions) for the four experimental conditions to which participants were randomly assigned. Finally, participants were given the power to redistribute all the points between themselves and the other. That is, they could reduce the asymmetry by either giving points to the other or taking points from the other. As dependent measure, we used participants’ relative outcome, that is, the ratio of (i) a participant’s own points after the redistribution to (ii) the sum of all points earned by both individuals. The relative outcome hence mirrors a participant’s final share. For the sake of simplicity, we subtracted 0.5 from this proportion so that a value of 0 mirrors establishment of an equal split (i.e., no asymmetry left after redistribution). Any value larger than 0 indicates that participants gained beyond the equal split through their redistribution choice; any value smaller than 0 indicates the opposite (the other came off better than implied by the equal split). Note that, by implication, the same value of the relative outcome implies different redistribution choices across the two conditions. For example, a value of 0 indicates giving in the advantage-self condition, but taking in the advantage-other condition. 1.2. Procedure and participants The study was run online in close adherence to common standards for Internet-based experimenting (Reips, 2002). After providing informed consent and demographic information, participants completed the HEXACO-60. Next, they worked on the quiz in which they (and the other) earned the points. After the quiz, participants received information on their own and the other’s earnings and decided on the final distribution of points 3 Note that the paradigm was completely hypothetical in nature, meaning that participants were asked to imagine both the other person and the points earned and to be distributed. Thus the points did not convert to actual monetary stakes.
Table 1 Exemplary initial allocation of points for a participant with 10 correct solutions in the quiz, separated by experimental condition. Degree of asymmetry
Advantage-self
Advantage-other
Own points
Others’ points
Own points
Others’ points
Moderate Severe
40 80
20 20
20 20
40 80
Table 2 Means and standard deviations of the final proportion of participants’ points on all points (i.e., relative outcome) separated for the four experimental conditions. Degree of asymmetry
Moderate Severe
Advantage-self
Advantage-other
n
M
SD
n
36 36
0.11 0.12
0.12 0.15
36 38
M 0.03 0.06
SD 0.13 0.16
Note: a value of 0 indicates the equal split; values larger than 0 indicate that the final distribution was beneficial for the participant relative to the other (and vice versa for values below 0).
by giving, taking, or leaving the points untouched. As a manipulation check, we finally asked participants about the strenuousness of the quiz and further assessed the seriousness of participation. Participants were recruited from a community sample via mailing lists and social networks. Out of 185 participants who started the study, N = 146 (82 female, aged 18–62, M = 25.3, SD = 8.4) completed all tasks and were thus considered in the analyses.4 The majority (72.6%) of participants were students, 15.8% were employees. Participants were almost equally distributed across the experimental conditions, with n = 72 in the advantage-self (n = 36 moderate and n = 36 severe) and n = 74 in the advantage-other (n = 36 moderate and n = 38 severe) condition. 2. Results As intended by the chosen time limit – and supporting the ‘‘earning character’’ of the fairness paradigm – participants perceived the quiz to be moderately strenuous (M = 2.89, SD = 0.83, on a 5-point Likert-type scale). On average, they solved 8 out of 15 tasks (SD = 3.0) which is highly similar to the performance observed in the incentivized pilot study summarized above. In turn, participants and others earned a total of 67.7 points (SD = 28.5). In the redistribution of these points, 42% of participants took points from the other (on average 19.2 points, SD = 13.2), 31% gave points to the other (on average 14.8 points, SD = 8.9), and 27% left the distribution untouched. A majority of 51% of participants established an equal split, 32% advantaged themselves, and 17% advantaged the other. Correspondingly, after the redistribution, participants had 35.7 points on average (SD = 18.6) whereas the other had 31.9 points on average (SD = 17.9). For the relative outcome (i.e., the final proportion of participant’s points), Table 2 provides the means and standard deviations for all four experimental conditions. As can be seen (in a column-wise comparison), the relative outcome was larger when participants themselves profited from the asymmetry (M = 0.12, SD = 0.13) – as compared to the condition in which they were disadvantaged (M = 0.05, SD = 0.15). In particular, in the advantage-self condition, participants were less willing to shift 4 As indicated by a post hoc power analysis using G * Power (Faul, Erdfelder, Lang, & Buchner, 2007), this sample size yielded a high statistical power to detect small-tomedium sized (f2 = .10) main effects (in case of two predictors in total, 1 b = .99) as well as small-to-medium sized interaction effects (1 b = .93; significant increase in explained variance when adding an interaction term) in a multiple linear regression.
94
B.E. Hilbig et al. / Personality and Individual Differences 80 (2015) 91–95
Fig. 1. Relation between the relative outcome and Honesty–Humility (centered on the sample mean) separated for advantage-self (left) and advantage-other condition (right). The dotted line indicates the equal split distribution; the dashed line indicates the initial distribution prior to the redistribution choice. Confidence bands represent the 95% confidence intervals.
towards the equal split. This main effect was confirmed by a 2 (advantage: self vs. other) by 2 (degree of asymmetry: moderate vs. severe) ANOVA with F(1, 142) = 47.49, p < .001, gp2 = .25. In contrast, the severity of asymmetry (row-wise comparison in Table 2) did not influence participants’ relative outcome, F(1, 142) < 1, p = .658, gp2 < .01. That is, participants were as willing to establish the equal distribution if the asymmetry was moderate (M = 0.04, SD = 0.14) as compared to when the asymmetry was severe (M = 0.03, SD = 0.18). Finally, there was no interaction between the direction and the degree of asymmetry (F < 1). For this reason – and given that our main question refers to the effect of who is advantaged (i.e., self vs. other), not how severely someone is advantaged, on the redistribution decision – we discarded the severityfactor in the following analyses. To test the main hypotheses, we regressed the direction of asymmetry (dummy coded), Honesty–Humility, and their interaction on the relative outcome. First off, this analysis revealed a (positive) main effect of the direction of asymmetry, indicating a larger relative outcome in the advantage-self condition than in the advantage-other condition, b = .52, p < .001. Furthermore – in line with the benevolence account – we found a (negative) main effect of Honesty–Humility on the relative outcome, b = .24, p = .013. That is, high levels of Honesty–Humility were generally associated with smaller values in the relative outcome, irrespective of the direction of asymmetry. However, in contrast to the predictions of the fairness norm account, there was no interaction between Honesty–Humility and the direction of asymmetry, b = .06, p = .529.5 The results were comparable when controlling for the remaining HEXACO dimensions in a multiple regression analysis.6 As can be seen in Fig. 1, individuals low in Honesty–Humility left the asymmetry untouched whenever they benefited from it but strongly reduced the asymmetry (took points from the other) whenever it disadvantaged them. By contrast, their counterparts high in Honesty–Humility reduced the asymmetry especially whenever they originally profited from it (gave points to the other), but did not 5 Moreover, this interaction did not depend on the severity of asymmetry as indicated by a non-significant three-way-interaction between Honesty–Humility, direction, and severity of asymmetry, b = .02, p = .895. 6 The corresponding regression coefficients were b = .00 (Emotionality), b = .08 (Extraversion), b = .08 (Agreeableness), b = .11 (Conscientiousness), and b = .09 (Openness), all p > .19.
establish a full disadvantaged.
equal
split
whenever
they
were
initially
3. Discussion Fairness represents a vital aspect of interpersonal and societal functioning. Regarding individual differences in fairness and traits driving fair behavior, recent developments point to Honesty– Humility as a vital basic trait dimension (cf. Ashton et al., 2014). In line with its theoretical conceptualization, past research has consistently linked Honesty–Humility to fairness-related and pro-social behaviors. Notably – as indicated by our re-analysis of extant findings on economic behavior – Honesty–Humility is not only associated with pro-sociality in relative terms, but also with an inclination to establish fairness (i.e., equality) in absolute terms. However, different explanations may account for why these individuals show such a strong tendency to establish fairness. First, the HEXACO theory primarily suggests that benevolence drives these effects in that individuals high in Honesty–Humility avoid exploiting others. However, prior evidence cannot rule out an internalized (blind) fairness norm (i.e., a general preference for equality) in these individuals. To tease the two explanations apart, the present study assessed participants’ willingness to rectify an initial asymmetry in rewards, that is, unfair outcomes following a demanding task. Specifically, it was tested whether individuals differ in their willingness to redistribute outcomes conditional upon (a) whether they themselves or the other profited from the asymmetric distribution of rewards and (b) their individual level of Honesty–Humility. In line with the benevolence account, our results show that individuals high in Honesty–Humility reduce an asymmetry and establish an equal split if and only if they themselves are the ones profiting from the unfairness. If, in turn, it is to their disadvantage, they do not alter the asymmetric distribution as strongly and mostly refrain from taking points from the other (even though this would actually result in a more fair distribution). These results corroborate the non-exploitation aspect of Honesty–Humility as the motivating force behind fair behavior, but are incompatible with the alternative notion of a general (blind) fairness norm. Of course, we cannot claim that benevolence is the only explanation that might account for the observed pattern of results. However, among the two
B.E. Hilbig et al. / Personality and Individual Differences 80 (2015) 91–95
accounts that we derived from the extant literature, the data exclusively supports one (benevolence) and contradicts the other (blind fairness norm). In turn, our findings support the conceptualization of Honesty–Humility in the HEXACO model, demonstrating that individuals with high factor levels do not take advantage of their power even if they themselves are in a disadvantageous position. Extending prior research, our study is the first specifically designed to test why individuals with high levels in Honesty– Humility are inclined towards fair behavior. We hence provide a first toehold towards explaining the consistently observed link between Honesty–Humility and fairness-related behaviors. However, as a limitation, our findings are based on a hypothetical scenario without actual stakes, thus potentially limiting the conclusiveness of results due to socially desirable responding. Arguably, fairness is socially desirable (Hoffman, McCabe, Shachat, & Smith, 1994) as are high levels of Honesty–Humility (Hilbig, Moshagen, & Zettler, 2015). Nonetheless, as indicated by recent research, the relation between Honesty–Humility and non-exploitation seems to be equivalent in hypothetical scenarios vs. real decisions with actual stakes (Hilbig et al., 2015). In conclusion, the present study provides first evidence that the willingness to establish fairness shown by individuals high in Honesty–Humility can be traced back to benevolence and a reluctance to take advantage of others’ inferiority, but not to a general preference for equality in terms of adherence to a ‘‘blind’’ fairness norm. Given that Honesty–Humility explains a variety of socially relevant behaviors such as crime, delinquency, and cooperation (e.g., Hilbig, Glöckner, et al., 2014; Van Gelder & De Vries, 2014), it is important to understand these behavioral tendencies, shedding light on the ‘‘black box’’ between Honesty–Humility and pro-social behavior. Acknowedgements The work reported herein was supported by grants to the first author from the Baden-Württemberg Foundation (Germany) and the German Research Foundation (HI 1600/1-1). References Ackermann, K. A., Fleiß, J., & Murphy, R. O. (in press). Reciprocity as an individual difference. Journal of Conflict Resolution. http://dx.doi.org/10.1177/ 0022002714541854. Arneson, R. J. (1982). The principle of fairness and free-rider problems. Ethics, 92(4). Ashton, M. C., & Lee, K. (2007). Empirical, theoretical, and practical advantages of the HEXACO model of personality structure. Personality and Social Psychology Review, 11(2), 150–166. http://dx.doi.org/10.1177/1088868306294907. Ashton, M. C., & Lee, K. (2009). The HEXACO-60: A short measure of the major dimensions of personality. Journal of Personality Assessment, 91(4), 340–345. http://dx.doi.org/10.1080/00223890902935878. Ashton, M. C., Lee, K., & De Vries, R. E. (2014). The HEXACO Honesty–Humility, Agreeableness, and Emotionality factors: A review of research and theory. Personality and Social Psychology Review, 18(2), 139–152. http://dx.doi.org/ 10.1177/1088868314523838. Ben-Ner, A., Kong, F., & Putterman, L. (2004). Share and share alike? Gender-pairing, personality, and cognitive ability as determinants of giving. Journal of Economic Psychology, 25(5), 581–589. http://dx.doi.org/10.1016/S0167-4870(03)00065-5. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (rev. ed.). Hillsdale, NJ, England: Lawrence Erlbaum Associates Inc.. Cherry, T. L., Frykblom, P., & Shogren, J. F. (2002). Hardnose the dictator. American Economic Review, 92(4), 1218–1221. http://dx.doi.org/10.1257/ 00028280260344740. Dunlop, P. D., Morrison, D. L., Koenig, J., & Silcox, B. (2012). Comparing the Eysenck and HEXACO models of personality in the prediction of adult delinquency. European Journal of Personality, 26(3), 194–202. http://dx.doi.org/10.1002/ per.824. Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G⁄Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39(2), 175–191. http://dx.doi.org/10.3758/ BF03193146. Fehr, E., & Schmidt, K. M. (1999). A theory of fairness, competition, and cooperation. Quarterly Journal of Economics, 114(3), 817–868. http://dx.doi.org/10.1162/ 003355399556151.
95
Field, A. P. (2001). Meta-analysis of correlation coefficients: A Monte Carlo comparison of fixed- and random-effects methods. Psychological Methods, 6(2), 161–180. Hilbig, B. E., Glöckner, A., & Zettler, I. (2014). Personality and pro-social behavior: Linking basic traits and social value orientations. Journal of Personality and Social Psychology, 107(3), 529–539. http://dx.doi.org/10.1037/a0036074. Hilbig, B. E., Heydasch, T., & Zettler, I. (2014). To boast or not to boast: Testing the humility aspect of the Honesty–Humility factor. Personality and Individual Differences, 69, 12–16. Hilbig, B. E., Moshagen, M., & Zettler, I. (2015). Truth will out: Linking personality, morality, and honesty through indirect questioning. Social Psychological and Personality Science, 6, 140–147. http://dx.doi.org/10.1177/1948550614553640. Hilbig, B. E., Thielmann, I., Hepp, J., Klein, S., & Zettler, I. (2015). From personality to altruistic behavior (and back): Evidence from a double-blind dictator game. Journal of Research in Personality, 55, 46–50. http://dx.doi.org/10.1016/ j.jrp.2014.12.004. Hilbig, B. E., & Zettler, I. (2009). Pillars of cooperation: Honesty–Humility, social value orientations, and economic behavior. Journal of Research in Personality, 43(3), 516–519. http://dx.doi.org/10.1016/j.jrp.2009.01.003. Hilbig, B. E., Zettler, I., Leist, F., & Heydasch, T. (2013). It takes two: Honesty– Humility and Agreeableness differentially predict active versus reactive cooperation. Personality and Individual Differences, 54(5), 598–603. http:// dx.doi.org/10.1016/j.paid.2012.11.008. Hoffman, E., McCabe, K. A., Shachat, K., & Smith, V. (1994). Preferences, property rights, and anonymity in bargaining games. Games and Economic Behavior, 7(3), 346–380. Huseman, R. C., Hatfield, J. D., & Miles, E. W. (1985). Test for individual perceptions of job equity: Some preliminary findings. Perceptual and Motor Skills, 61(3), 1055–1064. http://dx.doi.org/10.2466/pms.1985.61.3f.1055 (Pt. 2). Huseman, R. C., Hatfield, J. D., & Miles, E. W. (1987). A new perspective on equity theory: The equity sensitivity construct. The Academy of Management Review, 12(2), 222–234. http://dx.doi.org/10.2307/258531. Kahneman, D., Knetsch, J. L., & Thaler, R. H. (1986). Fairness and the assumptions of economics. Journal of Business, 59(4), S285–S300. Kamas, L., & Preston, A. (2012). Distributive and reciprocal fairness: What can we learn from the heterogeneity of social preferences? Journal of Economic Psychology, 33(3), 538–553. http://dx.doi.org/10.1016/j.joep.2011.12.003. Kurzban, R., & Houser, D. (2001). Individual differences in cooperation in a circular public goods game. European Journal of Personality, 15(1), S37–S52. http:// dx.doi.org/10.1002/per.420 [Special issue]. Lee, K., & Ashton, M. C. (2004). Psychometric properties of the HEXACO personality inventory. Multivariate Behavioral Research, 39(2), 329–358. http://dx.doi.org/ 10.1207/s15327906mbr3902_8. Lee, K., Ashton, M. C., Morrison, D. L., Cordery, J., & Dunlop, P. D. (2008). Predicting integrity with the HEXACO personality model: Use of self- and observer reports. Journal of Occupational and Organizational Psychology, 81(1), 147–167. http:// dx.doi.org/10.1348/096317907X195175. Marcus, B., Lee, K., & Ashton, M. C. (2007). Personality dimensions explaining relationships between integrity tests and counterproductive behavior: Big five, or one in addition? Personnel Psychology, 60(1), 1–34. http://dx.doi.org/10.1111/ j.1744-6570.2007.00063.x. McCrae, R. R., & Costa, P. T. (1985). Updating Norman’s ‘‘adequacy taxonomy’’: Intelligence and personality dimensions in natural language and in questionnaires. Journal of Personality and Social Psychology, 49(3), 710–721. http://dx.doi.org/10.1037/0022-3514.49.3.710. Mellers, B. A., Haselhuhn, M. P., Tetlock, P. E., Silva, J. C., & Isen, A. M. (2010). Predicting behavior in economic games by looking through the eyes of the players. Journal of Experimental Psychology: General, 139(4), 743–755. http:// dx.doi.org/10.1037/a0020280. Moshagen, M., Hilbig, B. E., & Zettler, I. (2014). Faktorstruktur, psychometrische Eigenschaften und Messinvarianz der deutschen Version des 60-Item HEXACO Persönlichkeitsinventars. Diagnostica, 60(2), 86–97. Reips, U.-D. (2002). Standards for internet-based experimenting. Experimental Psychology, 49(4), 243–256. http://dx.doi.org/10.1026//1618-3169.49.4.243. Sullivan, B. A., Snyder, M., & Sullivan, J. L. (2008). The centrality of cooperation in the functioning of individuals and groups. In B. A. Sullivan, M. Snyder, & J. L. Sullivan (Eds.), Cooperation: The political psychology of effective human interaction (pp. 1–16). Malden: Blackwell Publishing. Thielmann, I., & Hilbig, B. E. (2014). Trust in me, trust in you: A social projection account of the link between personality, cooperativeness, and trustworthiness expectations. Journal of Research in Personality, 50(3), 61–65. http://dx.doi.org/ 10.1016/j.jrp.2014.03.006. Thielmann, I., Hilbig, B. E., & Niedtfeld, I. (2014). Willing to give but not to forgive: Borderline personality features and cooperative behavior. Journal of Personality Disorders, 28(6), 778–795. http://dx.doi.org/10.1521/pedi_2014_28_135. Van Dijk, E., & Vermunt, R. (2000). Strategy and fairness in social decision making: Sometimes it pays to be powerless. Journal of Experimental Social Psychology, 36(1), 1–25. http://dx.doi.org/10.1006/jesp.1999.1392. Van Gelder, J.-L., & De Vries, R. E. (2014). Rational misbehavior? Evaluating an integrated dual-process model of criminal decision making. Journal of Quantitative Criminology, 30(1), 1–27. http://dx.doi.org/10.1007/s10940-0129192-8. Zhao, K., & Smillie, L. D. (in press). The role of interpersonal traits in social decision making: Exploring sources of behavioral heterogeneity in economic games. Personality and Social Psychology Review. http://dx.doi.org/10.1177/ 1088868314553709.